Preservation of cardiac myocytes subjected to different preconditionsA comparative morphometric study of beating fibrillating and cardioplegically arrested canine hearts.код для вставкиСкачать
THE ANATOMICAL RECORD 235425-435 (1993) Preservation of Cardiac Myocytes Subjected to Different Preconditions: A Comparative Morphometric Study of Beating, Fibrillating, and Cardioplegically Arrested Canine Hearts A. SCHMIEDL, Ph.A. SCHNABEL, J. RICHTER, G. MALL, AND H.J. BRETSCHNEIDER Department of Anatomy ( A S . , J.R.), Division of Electron Microscopy, Department of Physiology (H.J.B.), Division of Vegetative Physiology, and Pathophysiology, University of Gottingen, Gottingen, and Department of Pathology (Ph.A.S., G.M.), University of Heidelberg, Heidelberg, Germany ABSTRACT This study compares the ultrastructure of beating canine hearts with that of hearts subjected to different clinically common forms of cardiac arrest. The contraction state per test field was ascertained according to a specially developed classification. The volume density of myofibrils and the surface to volume ratio of mitochondria were used as parameters for cellular and mitochondrial swelling. Contraction bands were not found in any of the differently pretreated hearts. Following immersion fixation, contractions as well as over- and hypercontractions in beating, fibrillating, and St.Thomas-arrested hearts are significantly more pronounced than in HTK-arrested hearts. Cellular and mitochondrial volumes were similar in beating and fibrillating hearts. St. Thomas-perfusion significantly decreased cellular and mitochondrial volume compared to beating hearts, but these values were in the same range as in fibrillating hearts. Only HTKsolution actually led to a strong reduction of these compartments. Compared to immersion, perfusion fixation after coronary perfusion with cardioplegic solutions led to comparable cellular volumes, but significantly elevated the percentage of relaxed sarcomeres and significantly reduced mitochondrial swelling. The best structural preservation of myocytes was found after HTK-perfusion and perfusion fixation. Such ultrastructural quantitative and morphometrical parameters are powerful tools since results confirm that the degree of myocardial preservation depends on the method of cardiac arrest. This forms the basis for the choice of preconditions for subsequent ischemia. Furthermore, significant alterations of myocardial ultrastructure depend on a combination of the functional state of the heart, the method of cardioplegia, and the technique of fixation. 0 1993 Wiley-Liss, Inc. In open heart surgery, different methods of cardiac arrest are used. The degree of myocardial protection during the subsequent ischemia depends on the energetic and structural integrity of the heart (Paulussen et al., 1968; Bretschneider, 1980; Hearse et al., 1981; Hearse, 1988).Therefore, a knowledge of the structural preservation of myocytes after different forms of cardiac arrest without ischemic stress is useful. The better the structural and energetic preservation, the higher is the ischemic tolerance (Schaper et al., 1986; Gebhard et al., 1989) as well as the change of a successful outcome of an operation on an ischemic heart. Furthermore, an adequate pretreatment and a suitable fixation technique are necessary to obtain a well preserved myocardial ultrastructure that can serve as a control for structural alterations of myocytes caused by ischemic stress or various heart diseases. Diagnostic samples can only be taken as endomyocardial catheter biopsies from beating hearts. A comparison of biopsies 0 1993 WILEY-LISS, INC with cut samples obtained from arrested hearts is indeed plausible, because earlier investigations have shown that the contraction state of sarcomeres, the structural preservation of organelles, and the degree of cellular swelling in the working myocardium in biopsies do not significantly differ from that in cut samples obtained from hearts perfused for 60 minutes with HTK- or St. Thomas-solution and subsequent reperfusion (Schnabel et al., 1991). Myocardial samples in open heart surgery as well as biopsies from beating hearts are fixed by immersion (Schaper et al., 1980; Billingham, 1983). However, perfusion fixation which can only be performed in exper- Received August 19, 1991; accepted September 16, 1992. Address reprint requests to Dr. A. Schmiedl, Zentrum Anatomie, Abteilung Elektronenmikroskopie Kreuzbergring 36, D-3400 Gottingen, Germany. 426 A. SCHMIEDL ET AL. imental research is considered to be a n optimal procedure for tissue preservation if a collapse of capillaries is to be prevented by washing out the vascular bed free of blood with a prefixation solution (Forssmann e t al., 1977; Marino e t al., 1983; Schaper et al., 1985; Schnabe1 e t al., 1985, 1987). This study was designed: 1. To compare the preservation of the myocardial ultrastructure after application of different clinically proven cardioplegic solutions for cardiac arrest only. This will establish the method, in which the highest degree of structural preservation of myocytes after immersion and perfusion fixation is obtained. 2. To reconsider the myocardial ultrastructure of fibrillating and beating hearts in comparison with cardioplegically arrested hearts. 3. To find out whether the mitochondrial outer membrane area is influenced by the different contraction states of sarcomeres seen after different forms of cardiac arrest. MATERIALS AND METHODS The investigations were carried out on the left ventricular myocardium of canine hearts. Anesthesia and experimental procedure were standardized as described elsewhere (Preusse et al., 1981). composition and application of the cardioplegic solutions were performed corresponding to the original clinically accepted prescriptions (Hearse et al., 1981; Gebhard e t al., 1987).Samples from the subendocardial third of the free ventricular wall were taken using different procedures: 1. Needle biopsies from beating hearts (n = 7), obtained with TRU-CUT biopsy needles (Travenol, Deerfield, Illinois, USA), were fixed by immersion. 2. Samples measuring approximately 0.5 x 1 x 0.3 cm were cut with scissors and fixed by immersion: a) from fibrillating hearts (n = 6), within 1 minute after aortic cross clamping, immediately after the onset of fibrillation; b) from St. Thomas-arrested hearts (n = 9) after 4 minutes coronary perfusion with the St. Thomas-solution (Dr.F.Kohler, Alsbach, FRG); composition in mmol/l: 91.1 NaC1, 14.8 KC1, 15 MgCl,, 1.2 CaCl,, 25 NaHCO,, 1.2 procaine.HC1, 1.2 MgSO,, 1.2 KH,PO,; c) from HTK-arrested hearts (n = 9), after 11 minutes coronary perfusion with the HTK-solution (Dr.F.Kohler, Alsbach, FRG); composition in mmol/l: 15 NaCl, 9 KCl, 4 MgCl,, 180 histidine, 18 histidine.HC1, 30 mannitol, 2 tryptophane, 1 K-ketoglutarate. 3. Tissues obtained after 15 minutes perfusion fixation of a) St. Thomas-arrested hearts (n = 3) and b) HTK-arrested hearts (n = 9) were sectioned with a scalpel. In fixation solution each needle biopsy was cut with a razor blade into six to nine tissue blocks, each sample taken with scissors into 10-15 tissue blocks of about 1 mm3. Border zones possibly bearing bruise artifacts caused by the biopsy needle or scissors were rejected. Fixation always lasted for 2-3 hours a t room temperature with 1.5 M glutaraldehyde and 1.5 M para- formaldehyde in 0.1 M sodium cacodylate buffer (total osmolality: 700-800; effective osmolality, a measure for the osmotic effect of the fixative caused primarily by the buffer: 420 mOsm/kg H,O; Hayat, 1981). After fixation tissue blocks were rinsed in 0.1 M sodium cacodylate buffer with 4% sucrose. Postfixation for 2 hours with 1% OsO, in 0.1 M sodium cacodylate buffer containing 3.5%sucrose followed at room temperature. After a subsequent final rinsing in 0.1 M cacodylate buffer with 4% sucrose, dehydration in a graded series of ethanol, and treatment with propylene oxide, specimens were embedded in araldite. All steps from fixation to embedding were standardized, automated, and carried out as a continuous process (Histomat, bio-med, Theres, FRG). Five randomly selected blocks from each heart were cut according to the random sectioning system (Weibel, 1979) to obtain nonorientated semithin sections. Sections were stained with toluidine blue and examined by light microscopy to eliminate artifacts and areas within a distance of less than 500 pm from the endocardium. Ultrathin sections were stained with uranyl acetate and lead citrate. Qualitative and quantitative investigations were carried out with a n EM 10 (Zeiss, Oberkochen, FRG) equipped with a TV camera and monitor. For the morphometry, 50 test fields per ultrasection were obtained by systematic random sampling (Weibel, 1979). A 72 point lattice test system with test lines was projected to each field. The volume densities of mitochondria (VvMi), free sarcoplasm (Vvsp), myofibrils (VVMf),and nuclei (VVNc)as well as the surface to volume ratio of mitochondria (SvratioMi) were determined according to the point and intersection point counting method (Weibel, 1979; Mall et al., 1986; Schmiedl et al., 1990). All volume densities were related to the myocardial cells as reference unit. The VVMfis regarded, under the conditions investigated in this study (no myofibrillolysis), as a direct inverse parameter of cellular edema (Schnabel et al., 1990). The Sv ratioMi serves to estimate mitochondrial size and shape and thus mitochondrial swelling (Mall et al., 1986; Greve et al., 1988; Schmiedl et al., 1990). We choose to interpret the low values of VVMf and Sv ratioMias indicative of slight to moderate edema and the high values as a n absence of edema. The surface density (SvMi) is a relative parameter for the density of mitochondrial outer membranes (Greve et al., 1988). The quotient SVMi/VVMfis independent of the reference space and a measure for absolute changes of mitochondrial outer membranes (Schmiedl et al., 1990). In 30 test fields showing myocytes the contraction states of sarcomeres were determined according to the following classification, which has in part been described previously (Schnabel et al., 1990): 1. Relaxation: I-bands are always visible. 2. Contraction: I-bands have generally disappeared; Z-band spacing is slightly wider than the A-band visible under physiological conditions. 3. Relaxation/contraction (incomplete relaxation): Relaxed and contracted sarcomeres are found in one test field. The contraction states above are summarized as physiological forms. Those following are considered pathological forms of contraction: 427 PRESERVATION OF CARDIAC MYOCYTES 1. Overcontraction: There are submaximal contractions resulting in A-band compression and apparent thickening of Z-bands. 2. Hypercontraction: There are submaximal to maximal contractions, which lead to A-band compression and additionally to clumping of actin and myosin filaments. 3. Contraction band: Contraction bands are considered maximal shortening of sarcomeres which in addition to actidmyosin clumping lead to distortion of the contractile system. This formation is regarded as irreversible damage (Ganote, 1983). Sometimes rupturing of the cell, nuclear, and/or mitochondrial membranes also occurs. The definitions of the contraction state are in agreement with those of other authors (Vanderwee et al., 1981; Todd et al., 1985; Vander Heide et al., 1986). The different contraction states of the sarcomeres were evaluated in this study by means of electron microscopy, because an exact distinction, especially of the pathological forms of contraction is not possible using light microscopy (Todd et al., 1985; Ashraf and Rahamathulla, 1989). All results are given as mean values SEM, if not indicated otherwise. Significant differences were noted for P values of 0.05 or less using the Wilcoxon-MannWhitney-test for unpaired samples (U-test). * volume (S, ratioMi:7.9 -I- 0.1 pm2/pm3;P < 0.05).The cristae are densely packed and the matrix structure appears dark. St. Thomas-Arrested Hearts After Immersion Fixation (Fig. 2b, see also Figs. 4-6) Relaxation is found in nearly 2% of the test fields, overcontraction in 11.2 2 3.0%, and hypercontraction in 2.6 2 1.8%.Compared to HTK-arrested hearts after perfusion fixation (controls) the myocytes show a moderately enlarged cellular volume (VVMf:69.2 0.9 Vo1.-%,P < 0.05) and the mitochondria are moderately swollen (S, ratioMi:6.9 4 0.1 pm2/pm3;P < 0.05). Mitochondrial cristae are intact, frequent local clearing of matrix structure and occasional loss of matrix occur. * Fibrillating Hearts (Figs. 3a, 4-6) Relaxation of sarcomeres occurs in 5.7 * 4.0%of test fields, overcontraction in 16.7 4.9%,and hypercontraction in 7.3 2.4%.Cellular volume is moderately increased (VVMf:68.9 0.8 Vo1.-%;P < 0.05) and mitochondria are remarkably (Sv ratioMi:6.3 0.2 pm2/ pm3; P < 0.05) enlarged compared to HTK-arrested hearts after perfusion fixation. * * * Beating Hearts (Figs. 3b, 4-6) Relaxation of sarcomeres is found in less than 1%of test fields, overcontraction in 26.2 4.5%, and hypercontraction in 7.3 & 2.4% (Fig. 3b). Compared to the RESULTS controls, the volume of myocytes appears moderately HTK-Arrested Hearts After Perfusion Fixation increased (VVMf:66.6 & 0.9 Vo1.-%; P < 0.05; Fig. 5) (Fig. la, see also Figs. 4-6) and most of the mitochondria show remarkable swellThe sarcomeres are largely relaxed (76.8 7.2%); ing (Sv ratioMi: 6.0 0.1 pm2/pm3;P < 0.05, Fig. 6) over and hypercontractions are not found; and the high and intact cristae. Clearings and sometimes loss of mavalues of vvMG(76.4 0.4 vol.-%),and of the svratioMi trix structure are visible. Compared to contracted or (8.4 0.1 pm /pm3) indicate unswollen myocytes and relaxed areas (compare Figs. 3a, la), the free sarcomitochondria. The mitochondria are somewhat elon- plasm is reduced in over-and hypercontracted areas gated, the cristae are very densely packed, and the (Fig. 3b), and the mitochondrial damage is much more dark matrix generally does not show clearings. Be- pronounced-extreme swelling, striking loss of matrix cause of the excellent structural integrity of the myo- structure, distinct fragmentation of cristae, and somecytes and the well preserved ultrastructure this group times even cristolysis are apparent. is regarded as control. The other groups investigated The SVMishows significant differences among all exhibit more or less pronounced alterations of the con- groups of this study. The quotient of SVMito VVMf traction state as well as of cellular and mitochondrial shows no statistical differences between the groups volume when compared to these controls. (Table 1). This means that the differences in size of mitochondria are caused by alterations of their volume St. Thomas-ArrestedHearts After Perfusion Fixation without significant alterations of their outer mem(Fig. lb, see also Figs. 4-6) brane area (comp. Schmiedl et al., 1990). 21.1 8.5%of the test fields exhibit relaxation and DISCUSSION only 0.3%overcontraction. Hypercontractions are not seen. The volume density of myocytes (VvMc 73. 9 At the moment of sampling, the hearts of the groups 1.7 Vo1.-%) does not show significant differences com- investigated are not subjected to any ischemic stress. pared to the HTK-arrested hearts after perfusion fixa- Because of this, the overall adenosine triphosphate (ATP) and phosphocreatine tissue concentrations in tion (controls). Mitochondria1volume (Svratio,,: 7.7 0.6 pm2/pm3)is significantly reduced compared to the samples taken from the hearts show no critical deficit controls. The mitochondrial matrix shows only spo- and may be sufficient for the reaction to contractile stimuli (Billingham, 1983) as well as for structural radic clearings. preservation metabolism. Thus, the partly significant HTK-Arrested Hearts After Immersion Fixation alterations of the myocardial ultrastructure may not be (Fig. 2a, see also Figs. 4-6) the result of energy deficits. The sarcomeres are relaxed in 30.7 * 7.9%and overBeating-Fibrillating Hearts contracted in less than 0.6%of the test fields. HyperThe extracellular electrolyte concentrations of contractions are not seen. In general, compared to the controls the myocytes are not too swollen (VVMf:76.3 k C a + + ,N a + , and K + in beating or fibrillating hearts 0.6 Vo1.-%), but the mitochondria show an increased correspond to physiological levels. The extremely low * * * * * * * * 428 A. SCHMIEDL ET AL. Fig. 1, a: Myocardial ultrastructure after coronary perfusion with HTK-solution and subsequent perfusion fixation serving as control in this study. x 15,000. b: Myocardial ultrastructure after coronary perfusion with St. Thomas-solution and subsequent perfusion fixation. x 15,000. degree of relaxation and the remarkably higher percentage of pathological contractions may be regarded as overreactions due to the presence of physiological calcium levels in these groups. Beating and fibrillating hearts show comparable morphometric values and comparable physiological forms of contraction (compare Figs. 4-6; U-test). The fact that the greater portion of test fields show hypercontractions in beating hearts may be partly related to the kind of samples taken in this group. The small size of these biopsies leads to a n insufficient suspension of the contractile apparatus. In combination with the mechanically in- duced depolarization caused by cutting the samples (Adomian et al., 1978; Thomson and Torp, 1979; Billingham, 1983), local overreactions to contractile stimuli such a s the first contact to fixation solution (Schnabe1 et al., 1990, 1991) may occur. Neither in beating hearts nor in arrested hearts do irreversible forms of contractions (Baroldi et al., 1977; Ganote, 1983) occur. Such forms may be caused by one or more combined detrimental factors. These factors may be alterations caused by underlying cardiac disease (Armiger and Smeeton, 1986), hypothermia (Karch and Billingham, 19861, reoxygenation after anoxic or ischemic periods 429 PRESERVATION OF CARDIAC MYOCYTES Fig. 2. a: Myocardial ultrastructure after coronary perfusion with HTK-solution and subsequent immersion fixation. x 15,000. Sarcomeres are relaxed. b: Myocardial ultrastructure after coronary perfusion with St. Thomas-solution and subsequent immersion fixation; x 15,000. (Ganote and Humphry 1985; Ashraf and Rahamathulla, 1989), global ischemia combined with partial aerobiosis of marginal layers, catecholamine release, and calcium overload (Schnabel et al., 1990). Cardioplegic-ArrestedHearts The degree of pathological contractions as well as swelling of myocytes and mitochondria after St. Thomas-cardioplegia are comparable with that of fibrillating hearts and significantly more pronounced than after HTK-cardioplegia (P < 0.05; Fig. 4). This may be due to the Na+ and Ca+ composition of the St. Thomas-solution which almost corresponds to that of the extracellular fluid, whereas in the HTK-solution these electrolyte concentrations are reduced to intracellular levels (Gebhard et al., 1989). Coronary perfusion with St. Thomas-solution a t 4°C or HTK-solution a t 8°C does not lead to an overall temperature in the subendocardial third of the left ventricle of substantially less than 10°C (Gebhard et al., 19871, and does not bring about cooling contractions seen in isolated myocytes of rabbits by an abrupt lowering of superfu+ 430 A. SCHMIEDL ET AL. Fig. 3. a: Myocardial ultrastructure of fibrillating hearts fixed by immersion. x 15,000. b: Myocardial ultrastructure of beating hearts fixed by immersion showing overcontractions (above) and hypercontractions (below). x 15,000. sate temperature to values of less than 5°C (Hryshko et al., 1989). experimental heart surgery as a resuscitation solution (Marino et al., 1983; Schaper et al., 1985; Schnabel et al., 1987), and the Forssmann-solution (Forssmann et Prefixation Solutions al., 1977) and cardioplegic-solutions (Schaper et al., In experimental and clinical research various prefix- 1986; Schnabel et al., 1985, 1987; Schmiedl et al., ation solutions are used for cardiac arrest and follow- 19901, which do not lead to specific interactions with ing perfusion fixation. Well known are the modified the fixative (Paulussen et al., 1968;Hearse et al., 1981; Tyrode-solutionserving without addition of procaine in Schaper et al., 1986; Schnabel et al., 1990). Neither = 431 PRESERVATION O F CARDIAC MYOCYTES = nypercontracticn V v Nuclel overcontraction a V v Lipid droplets Contraction 0 V" 0 Relaxation / Contraction V, VOLUME DENSITIES ( V v ) CONTRACTION STATE [%I 100- 3: 0 Vv Sarcoplasm Mitochondria Myofibrils 1007 80- 60- 80 - 60 - - 40 - 40 20 - 20- O d ... ..... B U T . FIERILL ARREST. ARREST. (n.71 L ( n - 6 ) ln-9) In-91 J IMMERSION FIXATION - THOUAS nnt ARREST. A I I C S T . ( n i l 1 In=)) PERFUSION FIXATION Fig. 4. Contraction states after different pretreatments. 0- THOMAS HTK FIBRILL ARREST ARREST BEAT (n=7) (n=6)(n=9) (n=9) \ / IMMERSION FIXATION U - THOMAS HTK ARREST ARREST (n=3) ( n = 9 ) PERFUSION FIXATION Fig. 5. Volume densities after different pretreatments. *Significant differences between the control group (HTK-arrested hearts and perfusion fixation) and the other groups investigated. precipitation of elements nor specific damage to the structure occurs. Immersion and Perfusion Fixation After Different Preperfusions Perfusion with Tyrode-solution and a subsequent cardiac arrest with KC1 lead to significant differences in the volume densities in the right ventricle of the cat after immersion and perfusion fixation (Marino e t al., 1983). After Tyrode-perfusion of canine hearts and immersion fixation more or less pronounced spots of poorly preserved ultrastructure occur. Collapsed capillaries with swollen endothelia, neighboring myocytes with over- to hypercontracted sarcomeres, swollen mitochondria with loss of matrix granules, clearings, and loss of matrix space are seen (Richter et al., 1984; Schnabel et al., 1984; Fig. 7a,b). Langendorff-perfusion of rat hearts with Tyrode-solution containing procaine and subsequent immersion or perfusion fixation lead to similar ultrastructural inhomogeneities in myocytes (Fig. 7c,d). These spots are the result of a n insufficient equilibration process combined with a lack of oxygen in these areas (Schnabel et al., 1984,1985).Because of the observed spots a comparison with other methods may be restricted. Cardiac arrest after coronary perfusion with the not clinically applied Forssmann-solution used for screening studies (composition: 154 mmol/l NaC1, 25 g/1 polyvinylpyrrolidone [PVP], 0.25 g/l heparin, 5 g/1 procaine.HC1; Forssmann et al., 1977) and immersion fixation show a n ultrastructure which is comparable with HTK-arrested hearts and immersion fixation (Figs. 2a, 8a) with one exception: between well preserved myocytes, myocytes with swollen mitochondria with clearings and loss of matrix occur (Fig. 8c). A subsequent perfusion fixation leads to a homogeneous preservation of the ultrastructure, which resembles that after HTK-perfusion and perfusion fixation (Figs. l a , 8b). The PVP should minimize volume changes of cells and of the interstitial spaces (Forssmann et al., 1977). After preperfusion with %.Thomas-solution and perfusion fixation, the myocytes are somewhat less swollen compared to immersion fixation, and no inhomogeneities of myocardial ultrastructure are seen (Figs. lb , 5). Preperfusion with HTK-solution and immersion fixation show a homogeneously well preserved myocardial ultrastructure which is comparable to that following perfusion fixation (Fig. 2a). The volume of mitochondria in all arrested hearts fixed by perfusion is significantly reduced (Fig. 6). The significant increase in relaxation after perfusion fixation (Fig. 4) can be explained mechanically by the effect of perfusion pressure. The perfusion pressure primarily stretches 432 A. SCHMIEDL ET AL. SURFACE TO VOLUME RATIO MITOCHONDRIA [v 2 / vm3] 9- Z2 SEM 8- 7- * 6- z 5- 4- 3- TABLE 1. Mitochondrial surface density per myocardial cells (Sv or. er volume density of myofibrils (SvMI/Vv flyor d&erent functional states anzpretreatments' Functional state and pretreatment Fibrillating hearts Immersion fixation St. Thomas-arrested hearts Immersion fixation St. Thomas-arrested hearts Perfusion fixation HTK-arrested hearts Immersion fixation HTK-arrested hearts Perfusion fixation SyMl (pm /pm3) 1,39' * 0.05 1,46' * 0.11 1,48 * 0,11 1.51 * 0.03 1.52 0.03 S"Mpv (I*m /,my f 2,02 * 0.12 2,12 f 0.18 1,96 * 0.16 2.04 * 0.13 1.99 f 0.14 'Values are x ? SD. 'Significant differences between the control group (HTK-arrested hearts and perfusion fixation) and the other groups investigated. fluid from sarcoplasm to mitochondria, which are remarkably more swollen than in physiologically contracted areas or by a pressing out of fluid into the interstitial space. In pathologically contracted regions the pronounced mitochondrial swelling is usually associated with further alterations of the mitochondrial deterioration (Sink e t al., 1979). The differences in mitochondrial shape between pathologically contracted and relaxed areas are obvious (Figs. 1-3) and lead to the question of whether mitochondrial swelling is due to a n increase in volume alone or in volume and surthe capillaries according to the "garden hose phenom- face. Linear regression analysis of the Sv ratioMi and enon" and secondarily stretches the sarcomeres of the the mean mitochondrial volume (VVMiNVMf)was permyocytes (Arnold et al., 1968).The degree of relaxation formed with logarithmically transformed values: In Sv depends also, however, on the perfusion pressure cho- ratioMi = In k - x . In VvMJVvMf For the values of the groups investigated there is a n inverse relationship (r sen. Thus, the functional state of the heart, the form of = -0.89, P < 0.02). Exponent x yields information as cardiac arrest, as well as the fixation technique may to whether swelling is caused by a change in volume or influence the preservation of myocardial ultrastruc- surface or both (Schmiedl et al., 1990). The empirically ture. The improvement of the preservation of the myo- determined exponent x (which varies between 1 = no cardial ultrastructure after perfusion fixation depends changes in the surface and 113 = dominant changes in further on the solution used for preperfusion. The de- the surface) is equal to 0.7 compared to a value for x of gree of improvement of the myocardial ultrastructural 0.9 obtained during ischemia; cf. Schmiedl et al. (1990). protection after perfusion fixation compared to immer- This lower value indicates that the increase in mitosion fixation depends on the capacity of the preperfu- chondrial volume is indeed caused by rounding off andl sion solution to guarantee a sufficient wash out from or smoothing of mitochondrial surface but also by a the vascular bed of all blood, combined with a dilata- seemingly slight change or possible overestimation of tion of capillaries. For the evaluation of pathologically mitochondrial surface. Mitochondrial shape may also altered myocardial ultrastructure HTK preperfused depend on the state of contraction, especially pathologhearts may serve as a n unaltered control. ical contraction. However, neither these differences nor swelling influence mitochondrial outer membrane Contraction State and Morphometry area markedly (Table 1). Physiological alterations of the contraction state do ACKNOWLEDGMENTS not lead to significant alterations of volume shifts The authors are indebted to Prof. Dr. M.M. Gebhard (Figs. 4-6). The reduced free sarcoplasm within pathological contractions could be caused by a shift of fluid and Prof. Dr. C.J. Preusse for providing the organ mafrom one area to another within one cell or by a shift of terial. We are grateful for valuable discussions held Fig. 6. Surface to volume ratio after different pretreatments. *Significant differences between the control group (HTK-arrested hearts and perfusion fixation) and the other groups investigated. PRESERVATION O F CARDIAC MYOCYTES Fig. 7. Myocardial ultrastructure of dog hearts after Tyrode-perfusion and immersion fixation (a, b) and of rat hearts and perfusion fixation (c, d). a: Significant ultrastructural damage is not seen. x 7,500. b Spot with hypercontracted myocytes. The collapsed capil- 433 lary shows a swollen endothelium. x 7,500. c: Sarcomeres are relaxed, mitochondria show a n intact ultrastructure. x 7,500. d: Spot with swollen mitochondria, showing local loss of matrix structure. x 10,000. 434 A. SCHMIEDL ET AL. Fig. 8. Myocardial ultrastructure after coronary perfusion with Forssmann-solution and immersion fixation (a, c) or perfusion fixation (b). a: Myocytes with contracted sarcomeres and unswollen mi- tochondria. x 7,500. b Myocytes with relaxed sarcomeres and unswollen mitochondria. x 7,500. c: Spot with swollen mitochondria showing loss of matrix structure. x 7,500. PRESERVATION O F CARDIAC MYOCYTES with Dr. M. Fleckenstein. We wish to thank S. Ebel, A. Gerken, E. Ehbrecht, H. Rode, and M. Scheumann for their skillful technical assistance in the electron microscopic work, E. Neumeyer for graphical assistance, H. Altmann for assistance in preparing the manuscript, and C. Maelicke, BSc, for correcting our cumbersome English. This work was supported by the DFG, SFB 330-Organprotektion. LITERATURE CITED Adomian, G.E., M.M. Laks, and M.E. Billingham 1978 The incidence and significance of contraction bands in endomyocardial biopsies from normal human hearts. Am. Heart J., 95:348-351. Armiger, L.C., and W.M.I. Smeeton 1986 Contraction-band necrosis: Patterns of distribution in the myocardium and their diagnostic usefulness in sudden cardiac death. Pathology, 18:289-295. Arnold, G., F. Kosche, E. Miessner, A. Neitzert, and W. Lochner 1968 The importance of the perfusion pressure in the coronary arteries for the contractility and the oxygen consumption of the heart. Pfliigers Arch., 299:339-356. Ashraf, M., and P.M. Rahamathulla 1989 Beneficial effects of low doses of ethanol on reoxygenation injury following anoxia in rat hearts. Basic Res. Cardiol., 84:378-387. Baroldi, G., M.D. Silver, W. Cixfeld, and D.C. McGregor 1977 Irreversible myocardial damage resembling catecholamine necrosis secondary to acute coronary occlusion in dogs: Its prevention by propanolol. J . Mol. Cell. Cardiol., 9:687-691. Billingham, M.E. 1983 The role of endomyocardial biopsy in the diagnosis and treatment of heart disease. In: Cardiovascular Pathology, Vol. 2. M.D. Silver, ed. Churchill Livingstone, New York, pp. 1205-1224. Bretschneider H.J. 1980 Myocardial protection. Thorac. Cardiovasc. Surg., 28:295-302. Forssmann, W.G., S. Ito, E. Weihe, A. Aoki, M. Dym, and D.W. Fawcett 1977 An improved perfusion fixation method for the testis. Anat. Rec., 188:307-314. Ganote, C.E. 1983 Contraction band necrosis and irreversible myocardial injury. J. Mol. Cell. Cardiol., 15:67-73. Ganote, C.E., and St.M. Humphrey 1985 Effects of anoxic or oxygenated reperfusion in globally ischemic isovolumic perfused rat hearts. Am. J . Pathol., 120:129-145. Gebhard, M.M., H.J. Bretschneider, E. Gersing, and Ph.A. Schnabel 1987 Bretschneider's histidine-buffered cardioplegic solution: Concept, application and efficiency. In: Myocardial Protection in Cardiac Surgery. A.J. Roberts, ed. Marcel Dekker, New York and Basel, pp. 95-119. Gebhard, M.M., H.J. Bretschneider, and Ph.A. Schnabel 1989 Cardioplegia: Principles and problems. In: Physiology and Pathophysiology of the heart, ed. 2. N. Sperelakis, ed. Kluwer, Dordrecht, pp. 655-669. Greve, G., S. Rotevatn, K. Berstad, H. Jodalen, K. Grong, and L. Stangeland 1988 Effects of verapamil and timolol on cellular morphometric changes in cat hearts with regional ischemia. Virchows Arch. [A], 412:291-299. Hayat, M.A. 1981 Fixation for Electron Microscopy. Academic Press, New York. Hearse, D.J. 1988 The protection of the ischemic myocardium: Surgical success v clinical failure? Progr. Cardiovasc. Dis., 30t381402. Hearse, D.J., M.V. Braimbridge, and P. Jynge 1981 Protection of the Ischemic Myocardium. Raven Press, New York. Hrvshko. L.V.. V. Stiffel. and D.M. Bers 1989 Rauid cooline contrac" tures as a n index of sarcoplasmatic reticulum calcium zontent in rabbit ventricular myoytes. Am. J . Physiol. (Heart Circ. Physiol.) 257(26):H1369-H1377. Karch, St.B., and M.E. Billingham 1986 Myocardial contraction bands revisited. Hum. Pathol. 17:9-13. Mall, G., T. Mattfeld, H.-J. Mobius, and R. Leonard 1986 Stereologic study on the rat heart in chronic alimentary thiamine defi- 435 ciency-absence of myocardial changes despite starvation. J. Mol. Cell. Cardiol., 18:635-643. Marino, T.A., S.R. Houser, F.G. Martin, and A.R. Freeman 1983 An ultrastructural morphometric study of the papillary muscle of the right ventricle of the cat. Cell Tissue Res., 230.543-552. Paulussen F., G. Hiibner, D. Grebe, and H.J. Bretschneider 1968 Die Feinstruktur des Herzmuskels wahrend einer Ischamie mit Senkung des Energiebedarfs durch spezielle Kardioplegie. Klin. Wochenschr., 46:165-171. Preusse, C.J., M.M. Gebhard, and H.J. Bretschneider 1981 Myocardial equilibration processes and myocardial energy turnover during initiation of artificial cardiac arrest with cardioplegic solution-reasons for a sufficiently long cardioplegic perfusion. Thorac. Cardiovasc. Surg., 29:71-76.. Richter, J., Ph.A. Schnabel, M.M. Gebhard, T. Pomykaj, and H.J. Bretschneider 1984 Effects of perfusion with different solutions prior to fixation on myocardial fine structure. Acta Anat., 1201 1-2:59. Schaper, J., E. Meister, and G. Stammler 1985 Ultrastructural morphometric analysis of myocardium from dogs, rats, hamsters, mice and from human hearts. Circ. Res., 56:377-391. Schaper, J., H.H. Scheld, U. Schmidt, and F. Hehrlein 1986 Ultrastructural study comparing the efficacy of five different methods of intraoperative protection in the human heart. J . Thorac. Cardiovasc. Surg., 92:47-55. Schaper, J., F. Schwarz, H. Kittstein, E. Kreisel, B. Winkler, and F.W. Hehrlein 1980 Ultrastructural evaluation of the effects of global ischemia and reperfusion on human myocardium. Thorac. Cardiovasc. Surg., 28:337-342. Schmiedl, A,, Ph.A. Schnabel, G. Mall, M.M. Gebhard, D.H. Hunneman, J . Richter, and H.J. Bretschneider 1990 The surface to volume ratio of mitochondria, a suitable parameter for evaluating mitochondrial swelling. Correlations during the course of myocardial global ischemia. Virchows Arch. [A], 416:305-315. Schnabel, Ph.A., F. Bach, U. Bartels, B. Ramsauer, J. Richter, A. Schmiedl, M.M. Gebhard, and H.J. Bretschneider 1991 Limitations of a n evaluation of subendocardial structural protection using needle-biopsies. Z. Herz Thorax Gefapchir, 5.53-67. Schnabel, Ph.A., M.M. Gebhard, T. Pomykaj, C.J. Preusse, and H.J. Bretschneider 1984 Evaluation of left ventricular myocardial biopsies under different experimental conditions. Acta Anat., 1201 1 -2:66. Schnabel, Ph.A., J . Richter, H.M. Piper, T. Pomykay, R. Spahr, B. Spieckermann, P.G. Spieckermann, and H.J. Bretschneider 1985 Perfusion fixation of rat hearts. Verh. Anat. Ges., 79:315-318. Schnabel, Ph.A., A. Schmiedl, M.M. Gebhard, Th. Pomykaj, J . Richter, and H.J. Bretschneider 1987 Optimierung der Perfusionsfixierung des Herzens durch Kardioplegie. Verh. Anat. Ges., 81: 141-142. Schnabel, Ph.A., A. Schmiedl, B. Ramsauer, U. Bartels, M.M. Gebhard, J . Richter, and H.J. Bretschneider 1990 Occurrence and prevention of contraction bands in Purkinje fibres, transitional cells and working myocardium during global ischemia. Virchows Arch. LA], 417:463-471. Sink J.D., G.L. Pellom, W.D. Currie, W.R. Chitwood, R.C. Hill, and A.S. Wechsler 1979 Protection of mitochondrial function during ischemia by potassium cardioplegia: Correlations with ischemic contracture. Circulation, 60 (Suppl. 1):1159-1163. Todd G.L., G. Baroldi, G.M. Pieper, F.C. Clayton, and R.S. Eliot 1985 Experimental catecholamine-induced myocardial necrosis. I Morphology, quantification and regional distribution of acute contraction band lesions. J . Mol. Cell. Cardiol., 17:317-338. Thomson, H.K., and A. Torp 1979 Value and limitations of human myocardial biopsy. J Mol. Cell. Cardiol., 11:467-475. Vander Heide, R.S., J.P. Angelo, R.A. Altschuld, and C.E. Ganote 1986 Energy dependence of contraction band formation in perfused hearts in isolated adult myocytes. Am. J . Pathol., 125t5568. Vanderwee, M.A., St.M. Humphrey, J.B. Gavin, and L.C. Armiger 1981 Changes in the contractile state, fine structure and metabolism of cardiac muscle cells during the development of rigor mortis. Virchows Arch. [Bl, 35~159-167. Weibel, E.R. 1979 Stereological Methods and Practical Methods for Biological Morphometry, Vol. 1. Academic Press, New York.