Contributions of electron microscopy to the study of myocardial ischaemia

Bull. Org. mond. Sant 1969, 41, 865-872 Bull. Wid Hlth Org. Contributions of Electron Microscopy to the Study of Myocardial Ischaemia H. ALEXANDER HEGGTVEIT, M.D.1 Electron-microscope studies of experimental models of myocardial ischaemia have provided basic information on the pathogenesis of hypoxic heart injury. Correlation of ultrastructural changes with biochemical data confirms the importance of catecholamine release and ionic shifts in the early evolution of ischaemic injury. An altered cellular metabolism induced by ischaemia causes rapid depletion ofglycogen and is followed quickly by alterations in the nucleus, the mitochondria and the sarcotubular system; the myofibril is the organelle most resistant to hypoxia. Postmortem autolysis mimics early ischaemic change very closely and it probably has an initial hypoxic basis. Significant hypoxic-autolytic changes may begin during the agonal state. The time elapsing and the techniques of tissue preservation are critical in determining the amount of artefact. At present it is unrealistic to expect to obtain acutely ischaemic human myocardium soon enough after death to be of value in the estimation of the degree or duration of ischaemia by electron-microscope techniques. Rapidly progressive autolytic changes preclude the meaningful morphological assessment of hypoxic change at the ultrastructural level. At magnifications provided by the light-micro- scope, heart muscle appears to react to many diverse types of injury in essentially the same way. In fact, it is often difficult to distinguish primary myocardial necroses with reactive inflammatory changes from primarily infectious or inflammatory processes with secondary degenerative changes in myocardial cells; this is particularly true in the later stages of cardiac lesions. Electron-miscroscope studies have shown that various metabolic and physicochemical distur- bances may induce quite dissimilar morphological alterations in the affected myocardial cells. These differences may be qualitative, quantitative or chro- nological and relate to patterns of change in individ- ual subcellular organelles. Thus, cellular necrosis, from an ultrastructural point of view, may evolve along a number of different pathways. Conversely, certain disparate pathogenic agents may damage the heart muscle in similar fashion, suggesting certain common pathways of cellular injury. Experimental models of human myocardial in- farction suitable for ultrastructural study may be created in laboratory animals by provoking altera- 1 Associate Professor of Pathology, Faculty of Medicine, University of Ottawa, Canada; and Pathologist, Ottawa General Hospital. tions in coronary or non-coronary (myocardial) parameters. Coronary ligature, reduced arterial oxygen saturation or the injection of cardiotoxic doses of catecholamines are examples ofsuch models. Each of these may be criticized as not being truly representative of ischaemic heart disease as it is seen in man. However, within limits, valid and useful information may be obtained from such experiments. One must also be aware of temporal, and possibly qualitative, differences in the reactivity of animal tissues to ischaemia. These are probably more apparent during the later development of lesions than in the early stages. In this evaluation of ischaemic and hypoxic changes in the myocardium, emphasis will be placed on manipulations of the coronary blood flow and arterial oxygen concentration. Particular conside- ration will be given to the changes occurring within the first 6 hours of ischaemia. Because of their intimate relationship to the pathogenesis of ischae- mic heart damage, attention will also be given to the effects of catecholamines and electrolyte depletion on the integrity of the myocardium. No attempt will be made here to assess other types of experimental cardiopathies (such as the electrolyte-steroid cardio- pathy) since their relevance to ischaemic heart 2441 865 H. A. HEGGTVEIT disease is less obvious and the mechanisms more complex. ULTRASTRUCTURAL ASPECTS OF EXPERIMENTAL "ANOXAEMIC " CARDIOPATHY Heart muscle has been studied by means of the electron microscope in the rat, rabbit, dog, cat and man under various conditions of anoxia, acute and chronic hypoxia, permanent and temporary ischae- mia, employing a variety of experimental techniques (Bryant, Thomas & O'Neal, 1958; Caulfield & Klion- sky, 1959; Albert et al., 1961; Bahr & Jennings, 1961; Hecht, Korb & David, 1961; Holscher, Just & Mer- ker, 1961; Miller et al., 1961; Burdette & Ashford, 1963; Hasper, 1964; Miller, Rasmussen & Klionsky, 1964; Ahn, 1965; Burdette & Ashford, 1965; Hausamen & Poche, 1965a; Herdson, Sommer & Jennings, 1965; Hort & DaCanalis, 1965; Jennings, Baum & Herdson, 1965; Sulkin & Sulkin, 1965; Weddell, Merker & Neubert, 1965; Grosgogeat et al., 1966; Lozada & Laguens, 1966; Buchner & Onishi, 1967; Korb & Totovic, 1967-1969; Jennings et al., 1969; Poche, 1969). As far as possible, the hypoxic and ischaemic changes will be described sequentially as they occur in the various subcellular organelles. Time intervals for the appearance of the different alterations vary somewhat from species to species and with the experimental method used. Glycogen Acute hypoxic injury of the myocardium, regard- less of the mode of induction, is accompanied by rapid disappearance of glycogen granules from the sarcoplasm beginning within 1-5 minutes. After 40 minutes of ischaemia, cellular depletion of glycogen is virtually complete (Jennings, Baum & Herdson, 1965). Some of the earlier studies on the ultrastruc- tural aspects of myocardial ischaemia made no reference to glycogenic alterations. Nuclei Nuclear changes are an early and prominent feature of acute hypoxia. While they have been reported as early as 5 minutes (Caulfield & Klionsky, 1959) and as late as 4 hours (Bryant, Thomas & O'Neal, 1958), widespread nuclear damage is usually well established after 15 minutes of permanent ischaemia (Jennings, Baum & Herdson, 1965; Miller et al, 1961; Grosgogeat et al., 1966). Early nuclear injury is evidenced by rarefaction of the nucleoplasm and coarse clumping and margination of the chro- matin inside the nuclear envelope. Actual rupture of the nuclear membrane is a late occurrence. Mitochondria Although some mild mitochondrial changes may begin after 10-20 minutes (Holscher, Just & Merker, 1961; Caulfield & Klionsky, 1959; Jennings, Baum & Herdson, 1965), swelling, vesiculation and decreas- ed matrix density with loss of intramitochondrial granules and distortion of cristae are usually not well developed until after 30 minutes. After 1 hour, further vacuolization, disintegration of cristae and increasing deposition of electron-dense material, presumably calcium, are apparent. Rupture of some mitochondrial membranes may take place after 1-2 hours (Grosgogeat et al., 1966) but many remain intact for much longer (Jennings, Baum & Herdson, 1965). Sarcotubular system Swelling, widening or vacuolization of the com- ponents of the sarcoplasmic reticulum follows abruptly upon the mitochondrial changes and may be seen after 30 minutes. This is accompanied by intracellular oedema manifested by separation of myofibrils and enlarged, clear subsarcolemmal and perinuclear spaces devoid of glycogen particles. Myofibrils The contractile elemcnts are the organelles most resistant to anoxic injury. Diminished myocardial contractility may be reflected by early elongation or relaxation of sarcomeres but the myofibrils usually maintain their structural integrity until the later stages of hypoxic myocardial damage. Some authors have described transient contraction bands (Caulfield & Klionsky, 1959; Grosgogeat at al., 1966) and minor Z-band changes (Bahr & Jennings, 1961; Holscher, Just & Merker, 1961), while others have reported focal fraying of myofilaments in the early stages of hypoxia (Buchner & Onishi, 1967; Hecht, Korb & David, 1961). The myofibrils, however, generally remain intact during the first 6 hours of ischaemia (Grosgogeat et al., 1966). Intercalated discs Dehiscence of the intercalated discs is not usually present in the early hours of ischaemia according to most reports but is found as a late or agonal change. Buchner & Onishi (1967), however, believe that early oedematous separation or widening of the intercalated discs (demonstrated in rats after 10-30 minutes of hypoxia) is responsible for heart dilata- 866 ELECTRON MICROSCOPY IN THE STUDY OF MYOCARDIAL ISCHAEMIA tion and failure in the course of acute hypoxia. (Conversely, the narrowing of the discs after the hypoxic condition has been abolished causes the size of the heart to diminish. Poche (1969) also consi- ders this to be an important mechanism in cardiac dilatation. Sarcolemma Rupture of the plasma membrane with loss of organelles into the interstitial space is a late occur- rence in ischaemic heart injury, although separation with focal ruptures of the outer layer has been seen after 4 hours (Grosgogeat et al., 1966). Other structures Small lipid droplets arranged in rows between the mitochondria appear in increased numbers within the first hour of ischaemia (Korb & Totovi6, 1967) but may be seen as soon as 10 minutes after anoxic cardiac arrest (Holscher, Just & Merker, 1961). No meaningful observations on changes in lysosomes, ribosomes or Golgi apparatus have been made in the various studies of myocardial hypoxia. In the extracellular compartment, the only notable early change is variable oedematous swelling of the capil- lary endothelial cells. Discussion Acute hypoxic changes appear to be completely reversible up to 30 minutes with irreversible damage a constant occurrence after 60-90 minutes of oxygen deprivation (Jennings et al., 1965; Wedell et al., 1965). In their recent report of coronary ligation and posterior papillary muscle infarction in dogs, Jennings, Sommers & Herdson, (1969) showed that the ischaemic cells were reversibly injured for periods of 15-18 minutes. A few isolated cells or groups of cells became irreversibly injured (necrotic) after 20 minutes of ischaemia. About 50% of the cells in the posterior papillary muscle were dead after 40 min- utes, and after 60 minutes of ischaemia most of the cells were dead. Moderate hypothermia protects against the development or limits the severity of hypoxic changes (Wedell, Merker & Neubert, 1965; Holscher, Just & Merker 1961; Burdette & Ashford, 1963; Albert et al., 1961), although severe cold itself may damage the myocardium (Sarajas, 1964; Heggtveit, unpublished data). Studies of temporary coronary ligation in the rat, with restoration of blood flow, revealed a more rapid development and progression of the lesions than after permanent vascular occlusion (Herdson, Som- mers & Jennings, 1965; Hort & DaCanalis, 1965). Herdson, Sommers & Jennings, (1965) suggested that tissue-electrolyte disturbances develop more rapidly with restored flow. It is of interest that exten- sive mitochondrial calcification is usually not seen after permanent coronary ligation but is readily apparent after temporary occlusion. Repeated bouts of hypoxia in mice induced par- ticularly severe changes with the end-stage represen- ted by an almost complete loss of myocardial cell differentiation (Hasper, 1964); this again points to the possible participation of altered tissue-electrolyte distribution. Rats maintained under conditions of chronic hypoxia showed no depletion of myocardial glycogen and no significant nuclear changes, but intense mitochondrial swelling, intracellular oedema and myofibrillar alterations were apparent (Sulkin & Sulkin, 1965). These changes were completely reversible with slow reacclimatization to normal oxygen tension. Older animals were found to be much more susceptible to such changes (Sulkin & Sulkin, 1967). Chronic or repeated bouts of hypoxia are associated with increased formation and accu- mulation of lipofuscin pigment which is possibly an end-product of lysosomal activity. Anoxic cardiac arrest (H6lscher, Just & Merker, 1961; Burdette & Ashford, 1963; Miller et al., 1961; Miller, Rasmussen & Klionsky, 1964; Ahn, 1965) or exposure to hypobaric pressure (Hausamen & Poche, 1965a) results in rapid myocardial hypoxic changes similar to those produced by coronary ligation (Bryant, Thomas & O'Neal, 1958; Caulfield & Klionsky, 1959; Hecht, Korb & David, 1961; Jennings, Baum & Herdson, 1965; Grosgogeat et al., 1966; Korb & Totovic, 1967). Potassium arrest of the heart, on the other hand, caused no obvious glycogenic, mitochondrial or nuclear changes up to 1 hour (H6lscher, Just & Merker, 1961; Miller et al., 1961). This may have been due to a local fixative effect of potassium, the abrupt cessation of contrac- tion or a possible protective effect of the potassium ion. ULTRASTRUCTURE OF ACUTE HYPOXIA IN THE HUMAN HEART For obvious reasons there have been few ultra- structural studies of acute hypoxic changes in the human myocardium. Burdette & Ashford (1965) observed early hypoxic changes in atrial appendages cross-clamped for 15-30 minutes prior to surgical excision. Mitochondrial abnormalities, nuclear chromatin clumping and intracellular oedema were 867 H. A. HEGGTVEIT the principal features. Biopsy specimens obtained during cardiac surgery have revealed that mito- chondrial swelling can occur after only a few minutes (Meessen, 1968). Ifthe metabolism returns to normal, the changes are quickly reversed. ULTRASTRUCTURAL ASPECTS OF CHRONIC ISCHAEMIA IN THE HUMAN HEART The chronically ischaemic human heart is readily accessible to morphological studies in the living state. Pieces of cardiac, muscle are easily obtained in the course of myocardial revascularization proce- dures or during surgical excision of ventricular aneurysms or akinetic segments of the ventricular wall. Although ultrastructural studies of the myo- cardium in cases undergoing open-heart surgery have been reported (Bjork & Hultquist, 1967; Mees- sen, 1968), there are apparently no published reports dealing specifically with such ischaemic hearts. It is our experience that the chronically ischaemic, but non-infarcted, heart is structurally abnormal. The alterations include mitochondrial swelling, vacuoli- zation and loss of matrix density with occasional giant forms. The nucleoplasm may be less dense with some evidence of early clumping and margina- tion. There is no significant glycogen depletion and lipofuscin pigment is frequently seen. The changes are similar to the effects of chronic hypoxia in the rat heart (Sulkin & Sulkin, 1965). If hypertrophy is also present, alterations in the size, number and spatial arrangement of myofibrils and quantitative changes in the mitochondrial population may be observed. Unfortunately, these abnormalities are variable and of a marginal nature, so that they may fall into the realm of technical variation and subjec- tive interpretation. It is perhaps fair to state, how- ever, that there are fine structural alterations in the chronically ischaemic myocardium reflecting a metabolic abnormality which might " condition ", " sensitize " or otherwise influence the reactivity of such hearts to the effects of superimposed acute ischaemic episodes. ULTRASTRUCTURAL ASPECTS OF EXPERIMENTAL MAGNESIUM DEPLETION Ultrastructural alterations in the heart muscle of magnesium-deficient rats may be detected after 5 days of dietary magnesium depletion, preceding overt histological evidence of necrosis by the same length of time (Heggtveit, Herman & Mishra, 1964; Heggt- veit, 1965). The earliest lesions are characterized by the swelling and vacuolization of mitochondria with distortion and condensation of cristae. Concomitant deposition of electron-dense material occurs on and between the cristae of many mitochondria. Calcium salts having a microcrystalline structure similar to hydroxyapatite accumulate to fill entire mitochondria, which mat together to form mineralized masses occupying the greater portion of individual myo- cardial cells. Magnesium-deficient mitochondrial changes are followed later by fragmentation and loss of myofibrils together with the formation of intra- cellular spaces containing clusters of translocated mitochoiidria, lipid droplets, residual glycogen particles and dilated components of the sarcoplasmic reticulum. Nuclear chromatin clumping occurs only in the late stages of frank necrosis after severe sarcoplasmic damage. Magnesium deficiency is associated with a distur- bance of calcium metabolism and a competitive or antagonistic interrelationship between calcium and magnesium ions has been suggested (Kruse, Schmidt & McCollum, 1934; Alcock & Maclntyre, 1960). Uncoupling of oxidative phosphorylation is known to occur in the magnesium-deficient state (Balt- scheffsky, 1957; Vitale et al., 1957) and it is postula- ted that disruption of magnesium-dependent, intra- mitochondrial enzyme systems involved in energy production play a major role in the subsequent development of the lesions (Heggtveit, Herman & Mishra, 1964; Heggtveit, 1965). The lesions of mag- nesium deficiency may be enhanced by cold stress (Heggtveit, Herman & Mishra, 1964) or by thyro- xine administration (Susin & Herdson, 1967). It has been recognized for some time that magnesium salts (e.g., magnesium chloride) exert a protective or modifying influence on the development of certain types of cardiac necroses (Bajusz, 1962, 1965). The recent work of Lehr and his associates (Lehr, Kru- kowski & Col6n, 1966; Lehr, 1969) indicates that intracellular depletion of magnesium may be a common biochemical denominator in the early pathogenesis of cardiac necroses induced by various methods. The subsequent evolution of the lesions may relate, in large part, to this initial ionic disequi- librium. Hochrein et al. (1967) have also shown the rapid loss of intracellular magnesium in acute hypoxic states. ULTRASTRUCTURAL ASPECTS OF EXPERIMENTAL POTASSIUM DEPLETION It has been recognized for some time that potas- sium plays a role in experimental cardiac necrosis 868 ELECTRON MICROSCOPY IN THE STUDY OF MYOCARDIAL ISCHAEMIA (Prioreschi, 1967). Electron-microscope observations on the potassium-depleted rat heart have been reported by several investigators (Poche, 1958, 1969; Molnar, Larson & Spargo, 1962; Maurat et al., 1965) and indicate that an important factor is the alteration of membrane permeability induced by electrolyte imbalance. Varied contractility of adja- cent myocytes and imbibition of fluid by the myocar- dial cells are prominent features of potassium- deficiency lesions. Although extensive myofibrillar loss takes place in both magnesium and potassium depletion, the change is earlier and more evident in the latter condition. The mitochondria, on the other hand, appear to be reasonably well preserved in the potassium-deficient lesion, or only slightly altered. The predominance of sarcoplasmic abnormalities with little or no nuclear loss has been interpreted as a degenerative phenomenon (cytoplasmic dedifferen- tiation) rather than a focal necrosis (Molnar, Larson & Spargo, 1962). Maurat et al. (1965) regard the fine structural changes of potassium depletion as non-specific and suggest that potassium deficiency may be a common underlying factor in various types -of myocardial injury. However, it would seem that the subcellular changes relating to potassium loss occur somewhat later in the evolution of myocardial necrosis (Lehr, Krukowski & Colon, 1966; Lehr, 1969). Potassium efflux correlates better with myofibrillar damage and cellular dedifferentiation than with the early changes in the mechanism of energy release. Potassium salts (e.g., potassium chloride) have also been shown to have a protective effect on the development of various cardiac necroses (Bajusz, 1962; Prioreschi, 1967). ULTRASTRUCTURAL ASPECTS OF EXPERIMENTAL CATECHOLAMINE CARDIOPATHY In their study of isoproterenol-induced myocar- dial necrosis, Ferrans et al. (1964) found thickening and increased density of the Z-bands ½/2-hour after administration of the drug, but- obvious myofibrillar disintegration and the formation of lipid droplets were later developments. Mitochondrial swelling, vesiculation and cristolysis, as well as dilatation of the sarcoplasmic reticulum, were established at 2 hours. No great loss of glycogen granules was noted with isoproterenol, in contrast to epinephrine, and no mention was made of nuclear changes. The lesions were attributed to an exaggerated demand for oxygen imposed by a stimulatory effect on the myocardial metabolism. Similar, but less severe, alterations were described by Korb (1965) and by Hausamen & Poche (1956b). In addition, these authors found depletion of sarco- plasmic glycogen. The myocardial damage was considered to result from a brief oxygen deficiency and from an electrolyte disturbance (Korb, 1965), or from coronary insufficiency due to a hypotensive effect (Hausamen & Poche, 1965b). Sobel et al. (1966) have demonstrated an uncoupling effect of catecho- lamines on oxidative phosphorylation in rat heart mitochondria. Lehr and his associates (Lehr, Krukowski & Col6n, 1966; Lehr, 1969) have correlated tissue- electrolyte changes with fine structural alterations seen after treatment with various adrenergic amines. They found early shifts in ionic distribution accom- panying, and even preceding, mitochondrial damage which was evident at 2 hours. Rapid loss of myo- cardial magnesium was associated with the accumu- tion of calcium. A rise in sodium and drop in inorganic phosphate occurred within 24 hours whereas tissue-potassium levels remained initially remarkably constant. Ultrastructural studies sup- port the view that catecholamine release and elec- trolyte loss contribute to the pathogenesis of hypoxic and ischaemic myocardial damage and help to explain the sequence of changes observed. DELINEATION OF ISCHAEMIC CHANGES AT THE ULTRASTRUCTURAL LEVEL A problem made evident by electron:-microscopy is the apparent lack of delineation of an ischaemic area in the earliest stages of injury. Many histoche- mical techniques used for the demonstration of ischaemia are based on the delimitation of such zones. In theory, there should be a defined territory of damage from the earliest moments after the acute episode. In practice, however, many cells may survive for a time in the central parts of ischaemic zones (Herdson, Sommers & fennings, 1965; Jen- nings et al., 1969). Ultrastructural studies of diverse forms of myocardial injury have repeatedly demon- strated morphologically intact cells adjacent to irreversibly damaged or overtly necrotic cells (Heg- gtveit, Herman & Mishra, 1964; Grosgogeat et al., 1966; Caulfield & Klionsky, 1959; Molnar, Larson & Spargo, 1962; Ferrans et al., 1964). Not all cells of a target tissue are equally and simultaneously susceptible to injury. Within individual damaged cells, not all organelles are uniformly involved, e.g., damaged mitochondria mingle with normal mito- 869 H. A. HEGGTVEIT chondria, some myofibrils remain intact while others are fragmented, and zones of segmental damage may be unenvenly distributed in the sarcoplasm. In the earliest stages of injury it is therefore not possible to map out on a fine structural basis a defined area of infarction. (It should be mentioned here that electron-microscopy has distinct limitations in the sampling or screening of large areas of tissue.) From a quantitative point of view, the degree of organelle involvement and cellular injury generally increases with the passage of time and the persistence of ischaemia, so that a defined territory of damage eventually becomes manifest. POSTMORTEM AUTOLYSIS AND IATROGENIC DAMAGE Any consideration of experimental myocardial ischaemia, particularly when ultrastructural data are extrapolated to man, must take into account the factor of postmortem autolysis. Autolytic changes mimic early ischaemic changes quite closely and include loss of glycogen, alongation of sarcomeres, clumping of nuclear chromatin and the appearance of increased numbers of lipid droplets in the sarco- plasm (Bryant, Thomas & O'Neal, 1958; Hibbs & Black, 1963). These changes develop less rapidly and show more widespread uniformity of organelle in- volvement than in ischaemia (Bryant, Thomas & O'Neal, 1958). Such autolytic alterations, while well-established at 1 hour, do not progress substan- tially between 1 and 10 hours (Hibbs & Black, 1963). Although these changes still permit worthwhile ultrastructural observations to be made in postmor- tem cases of human cardiomyopathy (e.g., Hibbs et al., 1965), it is our experience that they prevent meaningful morphological assessmrent of carly ischaemic injury at the electron-microscopic level. It is unrealistic to expect to obtain acutely ischae- mic human myocardium soon enough after death to be of value in the estimation of the duration of ischaemia by electron-microscope techniques. From this point of view, deaths due to acute myocardial ischaemia would seem to fall into two categories: first, persons who die unattended or outside hospital and in whom the postmortem interval is too great; second, cases in which the acute attack receives medical attention, possibly from a coronary-care unit close at hand. This latter group now receives the benefit of vigorous cardiac resuscitative proce- dures which, in themselves, may be responsible for significant morphological alterations in the myocar- dium (Wolfe, Dudley & Wallace, 1968). Such damage will obviously obscure the interpretation of any underlying ischaemic changes. ACKNOWLEDGEMENTS This paper was prepared while the author was serving as a consultant to the World Health Organization on cardiovascular diseases. He wishes to thank Dr Z. Fejfar, Dr P. Y. Hatt and Dr D. Spiro for their advice and comments on the manuscript. The author's investigations have been supported in part by a grant from the Ontario Heart Foundation. RESUME APPORTS DE LA MICROSCOPIE ELECTRONIQUE A L'ETUDE DE L'ISCHtMIE DU MYOCARDE L'etude au microscope electronique de mod&les expe- rimentaux d'ischemie du myocarde a foumi nombre de renseignements fondamentaux sur la pathogenie des lesions cardiaques dues a l'hypoxie. La correlation entre les modifications ultrastructurales et les donnees bio- chimiques confirme l'importance du role joue par la liberation des catecholamines et les deplacements d'ions (perte de magnesium, de potassium et de phosphate; afflux de calcium, de sodium et d'eau) au debut de l'evolution des lesions ischemiques. Une alteration du metabolisme cellulaire, provoquee par l'ischemie, entraine une deperdition rapide de glycogene bientot suivie d'alterations du noyau, des mitochondries et du systeme sarcotubulaire. La myofibrille est l'organelle la plus resistante a l'hypoxie. La gravite des modifications dues A l'hypoxie et la rapidite avec laquelle elles se developpent dependent de la maniere dont elles ont ete provoquees et varient suivant l'espece animale utilisee. 870 ELECTRON MICROSCOPY IN THE STUDY OF MYOCARDIAL ISCHAEMIA 871 L'autolyse post mortem imite tres fidelement les pre- mieres alterations causees par l'ischemie et il est, en fait, probable qu'elle est initialement d'origine hypoxique. Des changements importants provoques par l'hypoxie et l'autolyse peuvent commencer a se manifester pendant l'agonie. Le temps ecoule avant l'examen et les methodes de conservation des tissus revetent une importance capitale en ce qui conceme la frequence des artefacts. A l'heure actuelle, il serait chimerique d'esp6rer obtenir un myocarde humain en etat d'ischemie aigue dans un delai suffisamment court apres la .mort pour qu'il puisse servir a evaluer par microscopie electronique le degre ou la duree de l'ischemie. Les alterations dues a l'autolyse progressent a une vitesse qui interdit toute estimation morphologique valable des changements survenus au niveau ultrastructural du fait de l'hypoxie. Peut-etre parviendra-t-on A apporter aux techniques de fixation et d'inclusion des perfectionnements qui per- mettront de mieux analyser le materiel humain. A l'heure actuelle, il pourrait etre possible d'etablir un lien entre les aspects des bandes du sarcomere et des parametres mecaniques tels que la pression de remplissage du cceur humain post mortem. De plus, il ne manque pas d'autres etudes A faire sur l'ischemie experimentale du myocarde, en associant notamment la microscopie electronique a d'autres techniques comme, par exemple, la cytochimie et l'autoradiographie. REFERENCES Ahn, K. (1965) Far East J. Anesth., 5, 74 Albert, H. M., Kattine, A., Glass, B. A. & Spurlock, B. 0. (1961) Circulation, 24, 875 Alcock, N. & Maclntyre, I. (1960) Biochem J., 76, 19P Bahr, G. F. & Jennings, R. B. (1961) Lab. Invest., 10, 548 Bajusz, E. (1962) Z. Ernahrungsw., 2, 229 Bajusz, E. (1965) The terminal electrolyte-shift mechanism in heart muscle; its significance in the pathogenesis and prevention of necrotizing cardiomyopathies. In: Bajusz, E., ed., Electrolytes and cardiovascular diseases, Basel and New York, Karger, vol. 1, p. 274 Baltscheffsky, H. (1957) Biochem. Biophys. Acta (Amst.), 25, 382 Bjork, V. 0. & Hultquist, G. (1967) Scand. J. thorac. cardiovasc. Surg., 1, 27 Bryant, R. E., Thomas, W. A. & O'Neal, R. M. (1958) Circulat. Res., 6, 699 Buchner, F. & Onishi, S. (1967) Beitr. path. Anat., 135, 153 Burdette, W. J. & Ashford, T. P. (1963) Ann. Surg., 158, 513 Burdette, W. J. & Ashford, T. P. (1965) J. thorac. cardio- vasc. Surg., 50, 210 Caulfied, J. & Klionsky, B. (1959) Amer. J. Path., 35, 489 Ferrans, V. J., Hibbs, R. G., Black, W. C. & Weilbaecher, D. G. (1964) Amer. Heart J., 68, 71 Grosgogeat, Y., Scebat, L., Renais, J. & Lenegre, J. (1966) Arch. Mal. Ceur, 59, 203 Hasper, B. (1964) Beitr. path. Anat., 130, 321 Hausamen, T.-U. & Poche, R. (1965a) Virchows Arch. path. Anat., 339, 212 Hausamen, T.-U. & Poche, R. (1965b) Virchows Arch. path. Anat., 339, 225 Hecht, A., Korb, G. & David, H. (1961) Virchows Arch. path. Anat., 334, 267 Heggtveit, H. A., Herman, L. & Mishra, R. K. (1964) Amer. J. Path., 45, 757 Heggtveit, H. A. (1965) The cardiomyopathy of magne- sium-deficiency. In: Bajusz, E., ed., Electrolytes and cardiovascular diseases, Basel and New York, Karger, vol. 1, p. 204 Herdson, P. B., Sommers, H. M. & Jennings, R. B. (1965) Amer. J. Path., 46, 367 Hibbs, R. G. & Black, W. C. (1963) Anat. Rec., 147, 261 Hibbs, R. G., Ferrans, V. J., Black, W. C., Weilbaecher, D. G., Walsh, J. J. & Burch, G. E. (1965) Amer. Heart J., 69, 766 Hochrein, H., Kuschke, H. J., Zaqqa, Q. & Fahl, E. (1967) Klin. Wschr., 45, 1093 Holscher, B., Just, 0. H. & Merker, H. J. (1961) Surgery, 49, 492 Hort, W. & DaCanalis, S. (1965) Virchows Arch. path. Anat., 339, 61 Jennings, R. B., Baum, J. H. & Herdson, P. B. (1965) Arch. Path., 79, 135 Jennings, R. B., Sommers, H. M., Herdson, P. B. & Kattenbach, J. P. (1969) Ann. N.Y. Acad. Sci., 156, 61 Korb, G. (1965) Virchows Arch. path. Anat., 339, 136 Korb, G. & Totovic, V. (1967) Virchows Arch. path. Anat., 342, 85 Korb, G. & Totovic, V. (1969) Ann. N.Y. Acad. Sci., 156, 48 Kruse, H. D., Schmidt, M. M. & McCollum, E. V. (1934) J. biol. Chem., 106, 553 Lehr, D., Krukowski, M. & Col6n, R. (1966) J. Amer. med. Ass., 197, 105 Lehr, D. (1969) Ann. N.Y. Acad. Sci., 156, 344 Lozada, B. B. & Laguens, R. P. (1966) Cardiologia, suppl. 1, vol. 49, p. 33. Maurat, J. P., Mercier, J. N., Ledoux, C. & Hatt, P. Y. (1965) Arch. Mal. Cawur, 58, 1004 Meesen, H. (1968) Amer. J. Cardiol., 22, 319 Miller, D. R., Rasmussen, P., Klionsky, B., Cossman, F. P. & Allbritten, F. F., Jr (1961) Ann. Surg., 154, 751 872 H. A. HEGGTVEIT Miller, R. D., Rasmussen, P. & Klionsky, B. (1964) Ann. Surg., 159, 208 Molnar, Z., Larson, K, & Spargo, B. (1962) Arch. Path., 74, 339 Poche, R. (1958) Virchows Arch. path. Anat., 331, 165 Poche, R. (1969) Ann. N.Y. Acad. Sci., 156, 34 Prioreschi, P. (1967) Canad. med. Ass. J., 96, 1221 Sarajas, H. S. S. (1964) Amer. J. Cardiol., 13, 355 Sobel, B., Jequier, E., Sjoerdsma, A. & Lovenberg, W. (1966) Circulat. Res., 19, 1050 Sulkin, N. M. & Sulkin, D. F. (1965) Lab. Invest., 14, 1523 Sulkin, N. M. & Sulkin, D. F. (1967) J. Geront., 22, 485 Susin, M. & Herdson, P. B. (1967) Arch. Path., 83, 86 Vitale, J. J., Hegsted, D. M., Nakamura, M. & Connors, P. S. (1957) J. biol. Chem., 228, 573 Wedell, J., Merker, H.-J. & Neubert, D. (1965) Virchows Arch. path. Anat., 338, 355 Wolfe, W. G., Dudley, A. W., Jr & Wallace, A. G. (1968) Arch. Surg., 96, 123