two and three dimensional display of myocardial ischemic “border zone” in dogs

Two and Three Dimensional Display of Myocardial lschemic

“Border Zone” in Dogs

ALDEN H. HARKEN, MD, FACC CLYDE H. BARLOW, PhD WESLEY R. HARDEN, III, MD BRITTON CHANCE, PhD

Philadelphia, Pennsylvania

From the Department of Cardiovascular Surgery and the Johnson Foundation, University of Penn- sylvania Medical School, Philadelphia, Pennsyl- vania. This study was supported in part by Grant 1 ROl HL-22315-01 from the National Institutes of Health, Bethesda, Maryland. Manuscript received May 9, 1978; revised manuscript received July 5, 1978, accepted July 5, 1978.

Address for reprints: Alden H. Harken, MD, Department of Cardiovascular Surgery, Hospital of the University of Pennsylvania, 3400 Spruce Street, Philadelphia, Pennsylvania 19104.

It is intuitively apparent that the ultimate fate of reversibly damaged, peri-ischemic “border zone” tissue should relate to individual patient survival. The purpose of this study was: (1) to describe a technique of assessing the adequacy of epicardial and myocardial oxygenation, and (2) to examine the extent and character of the peri-ischemic “border zone” in the dog after coronary arterial ligation. A diagonal branch of the left anterior descending coronary artery was ligated in four anesthetized open chest, neurohormonally intact dogs. The same diagonatcoronary artery was ligated in six isolated perfused canine heart preparations in which coronary partial pressures of oxygen and carbon dioxide, pH, blood flow and temperature were fixed. The ischemic zones were rapidly frozen and reduced nicotinamide adenine dinucleotide (NADH) fluorescence photographs were taken of the epicardium and at 0.5 mm depths into the myocardium. The distinction between perfused and ischemic myocardium is not apparent with the naked eye or natural light photography. Epicardial and myocardial oxidation-reduction status is well seen with NADH fluo- rophotography. In both the intact and perfused heart preparations the NADH-fluorescent (ischemic) border is jagged along all edges. Islands of perfused nonfluorescent tissue appear within the ischemic border. The transition between NADH-fluorescent ischemic cells and adjacent nonfluorescent tissue is less than 0.1 mm. The ischemic border is narrow. The distance between homogeneously NADH-fluorescent tissue and homogeneously nonfluorescent tissue (across the zone of island normoxia or microheterogeneity) may be as wide as 6 to 8 mm.

Abrupt coronary arterial occlusion in the experimental animal leads to regional hypoperfusion and local tissue hypoxia. Human coronary in- sufficiency is probably a less abrupt phenomenon and the degree of overlap between experimental and clinical myocardial ischemia is not clear. However, after 10 to 15 seconds of experimental myocardial hypoxia, TQ-ST segment changes occur1 and contractility decreases.2 The concurrent shift from aerobic and anaerobic metabolism permits analysis with direct tissue biopsy” or indirect venous efflux metabolite collec- tion.4J The spatial resolution of the biopsy technique has been limited by the size of the biopsy drill (usually 3 or 4 mm in diameter). Recently, the character and extent of the ischemic zone and the peri-ischemic border zone have been further examined in dogs.” The existence of a border zone of intermediate tissue injury has great clinical relevance and is currently being questioned.

A precise reflection of regional myocardial oxygen supply and demand ratio should permit better delineation of myocardial ischemia and justify further investigation into its pharmacologic and surgical modification. The purpose of this manuscript is (1) to describe a technique for assessing the adequacy of epicardial and myocardial oxygenation, and (2) to examine the extent and character of the peri-ischemic “border zone” in the dog after coronary arterial ligation.

954 December 1978 The American Journal of CARDIOLOGY Volume 42

ISCHEMIC BORDER ZONE-HARKEN ET AL.

Methods

Myocardial ischemia in anesthetized open chest preparation: Four male mongrel dogs, 20 to 28 kg, were anesthetized with sodium pentobarbital, 25 mg/kg body weight. Systemic blood pressure was monitored with a femoral arterial cannula and continuously displayed on a multichannel recorder (Electronics for Medicine model CR 9). The animals were intubated and ventilated (12 ml/kg tidal volume) to maintain normal acid-base status. The heart was approached through a left thoracotomy. A diagonal branch of the left anterior descending coronary artery was ligated to produce a small zone of ischemia on the left ventricular free wall. The ischemic insult persisted for 5 minutes. Large Wollenberger tongs7 (8 by 8 cm) were cooled in liquid nitrogen and abruptly placed across the ischemic zone so that this area was in the center of the large cold metal plate. The frozen portion of the heart was rapidly amputated and the tongs plus attached left ventricle were plunged immediately into liquid nitrogen.

Myocardial ischemia in isolated perfused heart preparation: Six mongrel dogs, 19 to 22 kg, of both sexes were anesthetized with sodium pentobarbital, 25 mg/kg. The heart was approached through a right thoracotomy. The innominate artery and right atrium were cannulated for a period of blood and temperature equilibration on partial cardiopulmonary bypass. An isolated, perfused decompressed heart (Fig. 1) was prepared by ligating the superior and inferior venae cavae and cross-clamping the transverse aortic arch.s A thermistor probe was placed through a purse string suture in the right atrium for temperature measurement. A membrane-lung perfusion systems9 permitted fixed coronary blood flow, stable coronary arterial partial pressures of oxygen and carbon dioxide and pH, and constant blood temperature.

A left thoracotomy revealed the left ventricle. After a 30 minute period of stabilization, a diagonal branch of the left anterior descending coronary artery was ligated. The ischemic insult persisted for 5 minutes. Large Wollenberger tongs7 were placed across the left ventricle and the heart was rapidly frozen as previously described.

NADH fluorescence photography of frozen ischemic zones: The quick-frozen left ventricular epicardium and myocardium were then shaved or filed off in successive mea- sured layers. The shaving blade and file were precooled in liquid nitrogen.

Nicotinamide adenine dinucleotide (NADH) fluorescence photographs were taken of the epicardial and myocardial surfaces at each level. This photographic technique has previously been described in reports from this laboratory.re~ll NADH fluorescence photographs were taken with a Bronica S2A camera. Corning 9788 and Wratten 45 filters were placed over the camera lens to allow transmission of NADH fluorescence in the 430 to 510 nanometer (nm) region. NADH fluorescence excitation was provided by two 400 joule xenon flashtubes (EG and G FX-47C3) covered by Corning 5840 filters to provide 330 to 380 nm excitation.

Results

Anesthetized open chest animals: Mean systemic blood pressure before coronary arterial ligation was 115 f 10 torr (standard error of the mean) and 5 minutes after ligation was 105 f 9 torr. Heart rate was 120 f 8 before and 126 f 6 beats/min after ligation. Arterial oxygen tension (PO,) was 85 f 10 torr before and 84 f 11 after occlusion. Arterial pH was 7.48 f 0.02 with arterial PCO:! of 32 f 5 torr before and pH 7.47 f 0.015 with arterial PC02 of 30 f 6 torr after occlusion. Con-

ARTERIAL INFUSION

FIGURE 1. Isolated beating heart doing no external pressure or flow work. Arterial blood is infused by way of the innominate artery, courses retrograde down the ascending aorta and out the coronary arteries. Venous blood is collected from the left-sided Thebesian system and the coronary sinus and Thebesian systems on the right. Catheter de- compression of both ventricles prevents any external pressure or flow work. Fluorescence photographic instrumentation: A 350 psec flash from two 400 joule xenon flash tubes is directed at the left ventricle and filtered through 330 to 380 nm filters. Emitted fluorescence, designating reduced NADH, is filtered at 440 to 5 10 nm and a photograph taken. This technique permits delineation of the oxygen supply/demand ratio over the entire surface of the epicardium or frozen myocardial specimen.

ditions before and after arterial ligation were not dif- ferent.

Isolated perfused heart preparation: In each perfused heart, coronary blood flow (pump flow) was regulated to produce a mean aortic (coronary ostial) pressure of 100 torr. This flow was absolutely fixed in each animal between 150 and 200 ml/min. Before and after coronary ligation, coronary blood flow did not change. There was no recognizable change in mean aortic pressure (100 f 2 torr) associated with coronary occlusion. Coronary arterial POP was 85 f 9 torr, PC02 was 38 f 2 torr and pH was 7.41 f 0.015. Coronary arterial oxygen and acid-base status did not change with ligation of the diagonal coronary artery. Right atria1 blood temperature remained 37 f 0.2” C during coronary occlusion.

NADH fluorescence photographs: After rapid freezing of each heart, NADH fluorescence photographs were taken of the epicardial ischemic zone and at se- quential levels down into the ischemic myocardium. Figure 2A is a natural white light photograph of an intramyocardial ischemic zone from a working heart. The epicardium has been shaved off to a depth of 1.5 mm. The distinction between perfused and ischemic myocardium is not well seen. Figure 2B is an NADH fluorescence photograph of the identical rapid-frozen intramyocardial ischemic zone. The course of the coronary artery, the level of the ligature and the hematoma are as noted. The distinction between the bright NADH fluorescence in the ischemic area and the nonfluorescent normoxic zone is evident.

December 1978 The American Journal of CARDIOLOGY Volume 42 955


NADH fluorescence photographs of a rapid-frozen isolated decompressed heart 5 minutes after ligation of a diagonal branch of the left anterior descending coronary artery are depicted in Figure 3. The outer 0.5 mm (Fig. 3A), 1.0 mm (Fig. 3B) and 1.5 mm (Fig. 3C) of epicardium has been shaved off. Fat and collagen fluoresce brightly. A peninsula of NADH fluorescence surrounds the ligated coronary artery distal to the occlusion. The NADH-fluorescent (“ischemic”) border is jagged along all edges. Within the ischemic area the NADH fluorescence is not homogeneous. Islands of nonfluorescent tissue appear within the ischemic border. An enlarged photograph (Fig. 3D) of the hetero- geneous fluorescence at the ischemic border reveals a central fluorescent core in many of the islands of nor- moxie tissue (also seen in Fig. 2B).

NADH fluorescence photographs of a rapid-frozen working heart 5 minutes after ligation of a diagonal branch of the left anterior descending coronary artery were examined. Again, the junction between NADH- fluorescent and nonfluorescent zones was jagged. Islands of nonfluorescent tissue within the ischemic border created a zone of microheterogeneity.

The junction at the jagged border of fluorescent tissue and the edge of each nonfluorescent (normoxic) island

A White light

is always abrupt. Gradations in intensity of NADH fluorescence are not seen. The distance between complete NADH fluorescence and complete nonfluo- rescence is in every instance less than the spatial resolution of this technique (less than 0.1 mm).

Discussion

Experimental myocardial ischemic tissue has been distinguished from adjacent normally perfused tissue with a variety of methods. ‘L-6 In each instance, the ca- pability of discerning regional heterogeneity is limited by the spatial resolution of each technique. The transition zone separating the core of an ischemic area from the surrounding normal tissue has traditionally been described as a critical “border zone” of reversibly damaged or jeopardized tissue.12 If this zone of reversibly damaged tissue is large, it is intuitively apparent, that its fate should ultimately relate to individual patient survival.lZi The converse may also be true.

During the past 2 decades, evidence has accumulated to suggest a narrow transition between normal and ischemic myocardium. Caulfield and Klionsky14 found a distance of 0.1 mm between pathologically normal and damaged myocardium. Fischl et al.‘” found blood flow in the central and lateral portions of an ischemic area

NADH Fluorescence

B I- Icrn -I

FIGURE 2. Rapid-frozen ischemic zone of canine left ventricle. A, white light photograph. The epicardium has been shaved off. The course of the coronary artery and level of the ligature are seen. It is difficult to appreciate the demarcation between normoxic and ischemic tissue. B, NADH fluorescence photograph. The ischemic zone is seen below the coronary ligature. The brightly fluorescent ischemic area is distinguished from the dark well oxygenated myocardium. The transition from NADH-fluorescent tissue to dark perfused tissue is abrupt (less than 0.1 mm). Islands of nonfluorescent tissue are seen within the zone of NADH ischemic fluorescence.



to be the same, and Linderls confirmed a very steep decrease in blood flow within a narrow surrounding border zone. Similarly, Hirzel et a1.e found homogeneous creatinine phosphokinase depletion throughout isch- emit myocardium with normal creatinine phosphokinase activity immediately adjacent to this area.

Characteristics of the ischemic border zone: The results of the present study delineate the ischemic border fluorophotographically. NADH was used as a natural intracell marker that, when fluorescent, indi- cated insufficient tissue oxygen supply. Incident light is filtered to the wavelength for NADH excitation, and

FIGURE 3. NADH fluorescence photographs of a rapid-frozen isolated decompressed heart 5 minutes after ligation of a diagonal branch of the left anterior descending coronary artery. A, the outer 0.5 mm of epicardium has been shaved off. Fat and vascular collagen fluoresce brightly. Zones of bare perfused ventricular myocardium are nonfluorescent. Small areas of NADH ischemic fluorescence are seen distal to the ligature. 6, 1.0 mm of epicardium shaved off. The peninsula of NADH fluorescence distal to the coronary ligature is larger with increasing distance into the myocardium. The transition between perfused and ischemic tissue is abrupt and the borderline is jagged. Islands of normoxic tissue are seen within the ischemic zone. C, 1.5 mm of epicardium shaved off. The NADH fluorescent ischemic zone continues to increase in size with distance into the myocardium. The abrupt, jagged ischemic border is again evident as are the normoxic islands within the ischemic zone. D, same specimen as in C, magnified X2. The abrupt transition from NADH ischemic fluorescence to nonfluorescent normoxic tissue is confirmed.


ISCHEMIC BORDER ZONE-HARKEN ET AL

FIGURE 4. Same specimen as in Figure 3 with 8 mm of epicardium and myocardium shaved off. The NADH-fluorescent ischemic zone persists and is well seen. Within the adjacent zone of perfused tissue is a diffuse area of NADH fluorescence. This is interpreted as freezing artifact. During the several seconds required to freeze this tissue, oxygen demand exceeds supply and NADH becomes reduced. A smaller zone of freezing artifact is evident at 1.5 mm in Figure 3C.

fluorescence between 430 and 510 nm represents accu- mulation of NADH from ischemic tissue.l7,ls The transition between NADH-fluorescent and nonfluorescent zones is always abrupt (less than 0.1 mm). The ischemic margin is not smooth but jagged. Within the border of NADH-fluorescent tissue there are islands of nonfluorescent cells measuring 0.5 mm in diameter. These islands have a central fluorescent core. Figure 2B represents an abrupt transition between perfused (NADH-nonfluorescent) and ischemic (NADH-fluorescent) tissue. Within the ischemic zone, we interpret this figure to indicate coronary collateral vessels, each surrounded by an island of normoxic tissues that have been cut in cross section. The core of these islands fluoresce because of the presence of vascular collagen. The cells just outside these islands are as fluorescent, and therefore as ischemic and as jeopardized, as the NADH-fluorescent tissue just within the other outer ischemic margin. Viable islands at the ischemic border have been demonstrated histologically by Jenningsi and confirmed with an autoradiographic technique by Cohn et a1.20 That group21 recently developed three dimensional reconstructions of serial sections through the ischemic border. They found no true viable islands but, instead, interdigitating peninsulas of normoxic tissue probably surrounding coronary collateral vessels in the border zone. Evidence from our study conforms exactly with these observations.

Transition from normally perfused and hypoxic tissue at the border zone: Sugano et a1.22 titrated NADH with oxygen and found a rapid reduction of NAD over a small oxygen gradient at very low oxygen concentrations. As oxygen concentration decreases from low7 to 10-s molar, NAD converts from 80 percent oxidized to 70 percent reduced. Fluorophotographically, tissues either do or do not appear to fluoresce. There is very little grey zone because the oxygen tension gradient is so steep at the ischemic border. In this study, intra- arterial oxygen tension was approximately 85 torr. NAD becomes reduced and fluoresces when the ambient oxygen tension falls below lop7 molar (0.07 torr).23 The distance between a perfused arteriole and NADH-fluorescent tissue is less than 0.1 mm. Thus, the oxygen gradient is at least (probably greater than) 0.1 torr/ micron.

The transition between NADH-fluorescent and nonfluorescent zones is always abrupt. In dogs, however, the distance between homogeneously nonfluorescent and homogeneously fluorescent areas (no interspersed perfused islands) may be 6 to 8 mm. A similar exami- nation of regional myocardial oxygen supply was carried out in the perfused rat and rabbit heart.24 The border zone was examined fluorophotographically with com- puterized light-guide scanning and with microanalytical biochemical techniques. The transition from 80 percent oxidized NAD to 80 percent reduced NADH occurred within a distance of 150 p. This was confirmed by biopsy microanalysis of NADH concentrations.24

No difference in the pattern of the ischemic border zone (epicardial or myocardial) was observed in specimens from the isolated perfused hearts and in hearts in intact open chest animals. The area of the ischemic zone clearly expanded down into the myocardium toward the endocardium in both preparations.

Problem of freezing artifact: We were concerned about the problem of freezing artifact. Intramyocardial tissue that was not frozen sufficiently rapidly might become reduced during the freezing process, producing spuriously large zones of NADH-fluorescent tissue. This problem has been evaluated in quick-frozen liver2”,2s and brain.2s,27 The investigators found stable oxidation-reduction potentials within tissue up to 1.0 mm below the surface of liquid nitrogen-fixed specimens. In our study, the freezing artifact is visible (Fig. 4). This specimen was taken 8 mm below the epicardial surface. Adjacent to the NADH-fluorescent ischemic zone is a blush of fluorescence due to slow freezing. This blush is also visible in the sections of the same heart taken at 1.5 mm (Fig. 3, C and D). However, the pattern of the ischemic zone and border remains clear. We believe, therefore, that freezing artifacts do not contribute to the NADH-fluorescent pattern of sections taken within 1 mm of the epicardial surface.

Implications: Tissue metabolic microheterogeneity is normal in many organs. The traditional approach to the delineation of ischemic and peri-ischemic zones has been to average the biochemical,4,s histochemicals and electrophysiologicl characteristics of a myocardial



sample volume. Until recently, the spatial resolution of these techniques was limited by the sample size (usually 3 to 4 mm). Fluorophotographic methods permit high resolution two dimensional evaluation of epicardial and myocardial oxidation-reduction states. It appears that the intramyocardial ischemic border in the dog may be defined in two ways. First, the distance between NADH-fluorescent ischemic cells and adjacent non-

1.

2.

3.

4.

5.

6.

7.

8.

9.

10.

11.

12.

13.

14.

15.

fluorescent cells is less than 0.1 mm. Second, the distance between homogeneously NADH-fluorescent tissue and homogeneously nonfluorescent tissue (across the zone of island normoxia) may be as wide as 6 to 8 mm.

Acknowledgment

We are grateful to David Reale for technical assistance and to Nancy Wells for secretarial assistance.

References

Holland RP, Brooks H: TQ-ST segment mapping: critical review and analysis of current concepts. Am J Cardiol 40:110-129, 1977 Mills SA, Jobsis FF, Seaber AV: A fluorometric study of oxidative metabolism in the in vivo canine heart during acute ischemia and hypoxia. Ann Surg 186:193-200, 1977 Cox JL, McLaughlin VW, Flowers NC, Horan LG: The ischemic zone surrounding acute myocardial infarction. Its morphology as detected by dehydrogenase staining. Am Heart J 76:650-659, 1968 Opie LH, Owen P: Effect of glucose-insulin-potassium infusions on arteriovenous differences of glucose and of free fatty acids and on tissue metabolic changes in dogs with developing myocardial infarction. Am J Cardiol 38:3 1 O-32 1, 1976 Opie LH, Bruyneel K, Owen P: Beneficial effects of glucose, potassium and insulin on tissue metabolic changes within first hour of myocardial infarction in the baboon. Circulation 52:49-57, 1975 Hirzel HO, Sonnenblick EH, Kirk ES: Absence of a lateral borderzone of intermediate creatinine phosphokinase depletion surrounding a central infarct 24 hours after acute coronary occlusion in the dog. Circ Res 41:673-683, 1977 Wollenberger A, Ristau 0, Schoffa G: Eine einfache Technik der extrem schnellen Abkuhlung grosserer Gewebestucke. Pflugers Arch 270:399-412, 1960 Harken AH, Woods M: Coronary arterial pH as a determinant of myocardial oxygen consumption. Surgery 81:696-700, 1977 Harken AH: Hydrogen ion concentration and oxygen uptake in an isolated canine hindlimb. J Appl Physiol 40:1-5, 1976 Barlow CH, Chance B: lschemic areas in perfused rat hearts: measurement by NADH fluorescence photography. Science 193:909-910, 1976 Barlow CH, Harken AH, Chance B: Evaluation of cardiac ischemia by NADH fluorescence photography. Ann Surg 186:737-740, 1977 Sobel BE, Shell WE: Jeopardized, blighted and necrotic myocardium. Circulation 47:215-216, 1973 Braunwald E, Maroko PR: The reduction of infarct size-an idea whose time (for testing) has come. Circulation 47:206-209, 1974 Caulfield J, Klionsky B: Myocardial ischemia and early infarction: electron microscopic study. Am J Pathol 35:489-523, 1959 Fischl SJ, Sonnenblick EH, Kirk ES: Absence of a border ischemic zone following acute coronary occlusion (abstr). Circulation 50:

16.

17.

18.

19.

20.

21

22.

23.

24.

25.

26.

27.

Suppl lll:lll-172, 1974 Linder E: Measurements of normal and collateral coronary blood flow by close-arterial and intramyocardial injection of krypto@ and xenon.133 Acta Physiol Stand 272:5-12, 1966 Chance B, Williams GR: Respiratory enzymes in oxidative phos- phorylation. Ill. The steady state. J Biol Chem 217:409-427, 1955 Page E, Polimeni PI, Zak R: Myofibrillar mass in rat and rabbit muscle: correlation with sterological measurements in normal and hypertrophied heart. Circ Res 30:430-439, 1972 Jennings RB: Early phase of myocardial ischemic injury and infarction. Am J Cardiol 24:753-765, 1969 Cohn PF, Kirk ES, Downey JM, Sonnenblick EH, Gorlin R: Auto- radiographic evaluation of myocardial collateral circulation in the canine heart. Cardiovasc Res 7:181-185, 1973 Factor SM. Sonnenbltck EH. Kirk ES: Histoloaical borderzone of acute myocardial infarction: islands or peninsulas? Circulation 56:Suppl lll:lll-71A, 1977 Sugano S, Oshino N, Chance B: Mitochondrial functions under hypoxic conditions: the steady states of cytochrome c reduction and of energy metabolism. Biochem Biophys Acta 347:340-358, 1974 Chance B, Oshino N, Sugano T, Mayevsky A: Basic principles of tissue oxygen determination from mitochondrial signals. Adv Exp Med Biol37A:277-292, 1973 Chance B, Barlow CH, Ha&grove J, Nakase Y, Quistorff B, Matschinsky F, Mayevsky A: Microhetercgeneities of redox states of perfused and intact organs. In, O’Hara Biochemical Research Symposium on the Role of Compartmentation in Metabolic Regu- lation: Microenvironmental Aspects, Dallas, Texas, January 16 and 17, 1978 (Srere PA, Estabrook RW. ed). New York, Academic Press, in press Faupel RP, Seitz HJ, Tarnowski W, Theimann V, Weiss CH: The problem of tissue sampling from experimental animals with respect to freezing techniques, anoxia, stress and narcosis: a new method for sampling rat liver tissue and the physiological values of gly- colytic intermediates and related compounds. Arch Biochem Biophys 148:509-522, 1972 Guistorif B: A mechanical device for the rapid removal and freezing of liver or brain tissue from unanesthetized and nonparalyzed rats. Anal Biochem 68: 102-l 18, 1975 Veech RL, Harris RL, Veloso D, Veech EH: Freeze-blowing: a new technique for the study of brain in vivo. J Neurochem 20:183-188, 1973


two and three dimensional display of myocardial ischemic “border zone” in dogs

Documents