Skip Navigation

Cardiovascular Research 1998 39(1):121-135; doi:10.1016/S0008-6363(98)00069-8
© 1998 by European Society of Cardiology
This Article
Right arrow Extract Freely available
Right arrow FREE Full Text (PDF) Freely available
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to My Personal Archive
Right arrow Download to citation manager
Right arrowRequest Permissions
Google Scholar
Right arrow Articles by Verdouw, P. D
Right arrow Articles by Duncker, D. J
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Verdouw, P. D
Right arrow Articles by Duncker, D. J
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us  
What's this?

Copyright © 1998, European Society of Cardiology

Animal models in the study of myocardial ischaemia and ischaemic syndromes

Pieter D Verdouw*, Mirella A van den Doel, Sandra de Zeeuw and Dirk J Duncker

Experimental Cardiology, Thoraxcenter, Cardiovascular Research Institute COEUR, Erasmus University Rotterdam, Rotterdam, The Netherlands

* Corresponding author. Experimental Cardiology, Thoraxcenter, Erasmus University Rotterdam, P.O. Box 1738, 3000 DR Rotterdam, The Netherlands. Tel.: +31-10-408-8029; Fax: +31-10-436-5607; E-mail: verdouw@tch.fgg.eur.nl

Received 29 October 1997; accepted 12 February 1998

KEYWORDS Myocardial ischaemia; Animal models; Ischaemic syndromes


   1 Introduction
Top
 1 Introduction
2 Myocardial ischaemia
 3 Ischaemic syndromes
 4 Future developments
 References
 
Although myocardial ischaemia has been one of the most extensively studied topics in cardiovascular research, its definition is still debated [1]. In the strictest sense ‘ischaemia’ derived from the Greek words, ischõ (to restrain) and haima (blood), means insufficient blood, and if we would adhere to this definition, all animal models which lack blood (e.g. isolated heart models) should by definition be termed ‘ischaemic'. Most investigators prefer to define ischaemia as an imbalance between the amount of oxygen and substrates supplied to the heart and the amount needed to perform normal function [1, 2]. The rationale behind this definition is that the myocardium strongly depends on oxygen to sustain adequate oxidative phosphorylation, the only metabolic process that is capable of providing sufficient high energy phosphates to maintain normal myocardial contraction. When oxygen supply to the heart becomes impaired there will be inadequate production of high energy phosphates with a resultant decline in myocardial contractility. When this condition develops, the myocardium starts to produce some, but not sufficient, high energy phosphates by anaerobic glycolysis, and lactate which is the end-product of this process starts to accumulate in the myocardium. Myocardial ischaemia is thus viewed as a condition in which an imbalance exists between oxygen-supply and oxygen-demand, leading to anaerobic metabolism and reduced contractile function. Central to the definition of ischaemia is that coronary flow is not only insufficient to allow adequate energy production, but that the impaired flow also results in impaired removal of metabolic waste products. In the large majority of cases in man myocardial ischaemia is confined to specific regions of the myocardium (regional ischaemia), because a stenosis in a coronary artery prevents adequate perfusion of the artery's distribution area (supply ischaemia). In its early stage of development, the stenosis can be sufficiently mild so that oxygen supply is adequate under resting conditions, but becomes flow-limiting during stress such as exercise. At the moment that ischaemia develops during exercise, coronary blood flow may be higher than under resting conditions, but the increase is not sufficient to meet the increased oxygen demand (demand ischaemia). The situations that the entire human heart becomes ischaemic (global ischaemia) are relatively rare, but may be seen during severe hypotension and during open-heart surgery when aortic-cross clamping is required.

1.1 Ischaemia versus hypoxia
In man at high altitude myocardial oxygen supply may become impaired, not because of a reduced coronary blood flow but because the oxygen content of the blood is decreased. Because coronary blood flow is not restricted, there is no accumulation of myocardial metabolic waste products. This condition of reduced oxygen delivery in the presence of unrestricted coronary inflow, has been termed hypoxia (or anoxia when no oxygen is present). Important in the interpretation of the data from ischaemia and hypoxia studies is that in the latter there is adequate removal of metabolic waste products.

1.2 Ischaemia versus infarction
Myocardial ischaemia does not only lead to changes in (global and regional) cardiovascular function and metabolism, but also to changes in homeostasis of electrolytes, neurohormones and ultrastructural features of the myocardium. These changes can be seen within the first few minutes of ischaemia and are reversible when perfusion is promptly restored. However, when ischaemia is maintained there is a gradual transition from reversible to irreversible injury as infarction develops. The distinction between ischaemia and infarction thus concerns the reversibility of the changes.

In this overview we will discuss some of the most frequently used animal models in the study of myocardial ischaemia (and hypoxia) and ischaemic syndromes, which are the consequence of prior ischaemia.


   2 Myocardial ischaemia
Top
 1 Introduction
 2 Myocardial ischaemia
3 Ischaemic syndromes
 4 Future developments
 References
 
2.1 Regional ischaemia
2.1.1 Experimental animal
In man, myocardial ischaemia is most commonly regional in nature and this condition has been widely studied in intact animals. In in vivo studies the dog has traditionally been the most frequently used animal species, but because of cost and social pressure against the use of dogs, the pig has become a more favoured animal only in recent years [3, 4]. This is surprising because Leonardo da Vinci already used pigs to demonstrate the movement of the heart during the cardiac cycle nearly five centuries ago [5]. However, the pig did not become a widely used animal in biomedical research because of problems with handling of the animal. For instance, Pavlov banned all pigs from the laboratory as he considered all pigs to be hysterical because of their shrieking [4]. In most studies young and thus growing domestic swine are used and the age-dependency of cardiovascular responses should therefore be kept in mind. The age of domestic pigs in acute studies usually ranges from 3 to 5 months, which is sufficiently old for maturation of central nervous system regulation of cardiovascular function [6, 7]. An interesting feature is the relation between heart and body weight. In man this ratio is about 5 g/kg, whereas it is twice as high in dogs. For pigs weighing between 25 and 30 kg the ratio is similar to that in man, but for animals exceeding 100 kg it is only half that value. The relative low heart weight in domestic pigs is probably the result of selective breeding during the last two centuries. Nowadays, one year old pigs reach body weights of 150 kg, whereas 200 years ago weights of only 40 kg were attained in twice that life span [8].

Smaller animals such as the rabbit and the rat have also been used to study regional myocardial ischaemia [9], but instrumentation of in situ myocardium for measurement of regional metabolism and function is limited due to the small size of the heart and coronary vessels. An additional problem is that instantaneous reperfusion may not always occur when an occlusion of a coronary artery is released and because quantitative assessment of phasic coronary blood flow is difficult, one must rely on visual inspection to verify reperfusion (disappearance of cyanosis). Nevertheless, the use of small animals has gained a large popularity in studies of ischaemic preconditioning [10, 11], studies employing techniques such as microdialysis [12]and studies of post-infarction remodelling [13, 14].

2.1.2 Collateral circulation
The early response to ischaemia depends to a large extent on the existence of a collateral circulation [15]. Especially the group of Schaper has shown that major differences exist in the extent of collateralization of the various species [16–18]. Thus, it is important to realize that dogs may have a well-developed coronary collateral circulation (which may vary considerably among individual animals), while collateral blood flow in pigs and baboons is extremely low (<5% of transmural blood flow) [16, 19–21]. From the other species used in the study of myocardial ischaemia, it has been well established that the collateral circulation of the rat is sparse and that the rabbit may show intraspecies differences [18, 22]. The guinea pig has such an extensive collateral network that normal perfusion is maintained after a coronary artery occlusion and infarction does not develop. Cats also have a well-developed collateral network, but this is not sufficient to prevent infarction [22, 23].

In animals that do not have an extensive native collateral circulation, coronary collaterals can be induced by a variety of methods. These include permanent complete coronary artery occlusion, brief (2 min) repeated complete coronary artery occlusions at regular intervals for a number of weeks [24–26], progressive coronary artery occlusion by means of an ameroid constrictor [16, 27], partial coronary artery occlusion without or in combination with exercise [28–31](exercise alone does not promote collateral growth [29, 30], or embolization of the microvasculature with non-radioactive microspheres with a diameter of 25 µm [32]. In young growing pigs a mild fixed coronary stenosis becomes flow limiting as the animals grow older and collaterals develop [28]. Bloor et al. [33]have shown that in pigs collateral blood flow can eventually account for 80% of the needed perfusion at baseline and 50% during exercise.

It is quite obvious that when a coronary artery is completely occluded in a species with a sizeable collateral network, determination of residual coronary perfusion is mandatory to appreciate the severity of ischaemia. Labelled (radioactive [34, 35]or coloured [36, 37]) microspheres are the first choice for measuring collateral blood flow, as this technique permits the determination of both total collateral flow and its transmural distribution. The latter is important because from earlier studies we know that, in particular, the subendocardial perfusion is a major determinant of regional function [38]. For species with a negligible collateral circulation regional perfusion data during ischaemia are less mandatory. For a description of other methods to determine collateral blood flow and growth the reader is referred to ref. [18].

Although pigs and baboons differ from dogs in their collateral blood flow, they both represent distinct groups of patients with coronary artery disease as a large number of these patients may have few collaterals during the early phase of their disease. On the other hand, in patients with a longstanding history of coronary artery lesions, collaterals may develop and dogs may be a more appropriate model for their condition.

2.1.3 Experimental conditions
Because the pathophysiological responses to ischaemia and also the response of pharmacological agents may be influenced by the presence of anaesthetic agents [39–42]and acute surgical trauma [43], chronically instrumented awake animal models are the first choice to mimic myocardial ischaemia in man. In addition to the possibility of studying the animals without interference of anaesthesia and acute surgical trauma, the awake animal model has the major advantage that it can be used in studies requiring long-term interventions or follow-up and during physiological stress as produced by exercise. Furthermore, if only used in short-term studies, a number of different protocols can be performed in the same animal, provided that sufficient time is (can be) taken to prevent interference from previous interventions. The elaborate surgery under sterile conditions, the high cost of the implanted transducers and probes (which have a limited life span due to chronic implantation), the cost of housing, the initial therapy to minimize the unavoidable discomfort related to the recovery from surgery and some of the interventions have limited the widespread use of awake animals.

In the majority of studies, ischaemia is therefore produced in anaesthetized animals, either open- or closed-chest. The open-chest preparation has the advantage that regional function and metabolism can be studied in detail. In open-chest animals severe ischaemia is almost invariably produced by complete occlusion of a coronary artery by inflation of a balloon placed around the vessel. If one wants to avoid thoracotomy, intracoronary devices have to be used to occlude a vessel. The closed-chest model has the advantage that tissue trauma is minimized, but the approach requires fluoroscopy for proper positioning of the obstruction. The most frequently employed method is inflation of a balloon placed at the tip of a catheter [44].

In order to produce mild or moderate ischaemia, the flow in the coronary artery can be reduced to a fixed percentage of resting blood flow. In open-chest animals, this can be achieved by a hydraulic occluder positioned around the coronary artery [45], tightening a J-shaped screw clamp [46]or other mechanical devices [47, 48]. These methods have the disadvantage that control of blood flow is difficult and requires continuous measurement of blood flow with flow probes or the use of microspheres at intervals. In order to circumvent large variations in flow (e.g. by changes in perfusion pressure or movement of the occlusion devices), more advanced methods may be used which control perfusion pressure [49]or perfuse the coronary vascular bed by an extracorporeal system using the animal's own blood [50, 51]. Flow reductions can also be achieved by intracoronary placement of a hollow plug attached to a catheter [52, 53]. The lumen of a hole, drilled in the length of the plug, provides the limited perfusion when the catheter is so far advanced into the coronary artery that the plug occludes the vessel. An advantage of this method is that it can also be used in closed chest animals and that the lumen of the catheter allows post-stenotic pressure measurements. A disadvantage is the risk that the lumen of the plug may become obstructed due to formation of blood clots, unless precautions are taken. Furthermore, the site of the obstruction in the artery cannot be determined in advance, which may lead to considerable variations in the size of the ischaemic areas between individual animals.

A disadvantage of all above described methods is that they produce concentric stenoses, while in man the majority (~70%) of stenoses are eccentric. At the site of obstruction, vessels with a fixed concentric constriction do not respond to vasodilators, while vessels with eccentric stenoses are capable to increase their narrowed lumen, thereby enlarging coronary blood flow. There have been a few attempts to employ eccentric stenoses by partial inflation of an intraluminal balloon [54]. The drawback of this approach is that it is very difficult to keep the balloon (eccentric stenosis) in place. Movement of the balloon inside the artery will cause variation in the geometry of the residual lumen and thereby in the severity of the obstruction.

The left anterior descending coronary artery is most frequently used to produce regional myocardial ischaemia, as it permits selective sampling of the regional coronary vein for metabolic studies. Nevertheless, left circumflex coronary artery occlusions have also been used [55, 56]. It is well known that in man and pigs proximal occlusion of the left anterior descending coronary artery leads to a greater impairment of global left ventricular function than occlusion of the left circumflex coronary artery, which is believed to be caused by the greater area at risk. Hoit and Lew [57]have, however, shown that for the same areas at risk the compensatory increase in function in the adjacent non-ischaemic area is different for the left anterior descending coronary artery than for the left circumflex coronary artery. This illustrates that not only the size of the area at risk, but also its location within the left ventricle determines the global functional consequences of ischaemia.

Blood samples collected from the coronary sinus are quite often used for metabolic studies. These samples are not only contaminated with blood from adjacent non-ischaemic myocardium, but in pigs there is also contamination with blood from the left azygous vein, which drains into the coronary sinus. As a result of this, oxygen saturation in the coronary sinus of pigs may be around 65 to 70%, while in the local coronary vein it is usually around 20% or even less [58]. However, the use of coronary venous samples for studying metabolism is decreasing, because of recent developments in microdialysis, NMR spectroscopy and positron emission tomography [59–61].

2.2 In vivo models of coronary arterial thrombosis
In patients, myocardial ischaemia develops quite often after formation of a thrombus at the site of a stenosis. Models mimicking this process have been developed in the past to study the process of thrombus formation, its prevention by pharmacological agents or the effectiveness of thrombolytic agents. These models are not very useful for the study of well-defined ischaemia and the evaluation of cardioprotective therapies, because of variation in time to occlusive thrombus formation and uncontrollable duration and severity of ischaemia. These models are therefore not dealt with in this overview; for an extensive review of these models the reader is referred to [62].

2.3 Global ischaemia (in vitro models)
The consequences of acute myocardial ischaemia can often be studied more easily under well-controlled conditions in isolated than in in situ heart models. Langendorff has to be credited for being the first to devise a method to investigate the mechanical activity of an isolated heart [63]. The principle has been described extensively [64]and is based on forcing blood or an oxygenated fluid into the coronary arteries through a cannula implanted in the aorta. Initially, Langendorff used the model to study contractile function, but already before the turn of the century he reported on the relationship between phasic coronary blood flow and the rhythmic contraction of the heart [65]. Already very early, it was noted that coronary flow was considerably higher in hearts perfused with saline-like substances, than in hearts perfused with blood at the identical perfusion pressure; the reason being the higher viscosity of blood and consequently the higher coronary vascular resistance. The original Langendorff preparation underwent many modifications, which permitted studies under well-controlled conditions such as constant perfusion, perfusion at constant perfusion pressure or perfusion via a donor animal [66, 67].

A critical point in setting up the model remains the time interval between removing the heart from the animal and fixation in the perfusion system, as ATP and creatine phosphate start to decline within seconds after the heart is removed from the animal. After coronary perfusion with oxygenated blood has been re-established, a stabilisation period of at least 15 to 30 minutes must be allowed before baseline values can be taken. The stability of the preparation is usually assessed from heart rate, left ventricular pressure and coronary blood flow (weaning of the reactive hyperaemia after the period of ischaemia). Since in vivo experiments have revealed that brief periods of ischaemia may already lead to ischaemic syndromes such as stunning and preconditioning and as these phenomena are also studied in isolated heart models, one must be aware that the period between removing the heart from the animal and the re-establishment of coronary perfusion and also the duration of the stabilisation period could influence the outcome of such studies.

In comparison with the in vivo models, the perfusion pressures of the isolated hearts are lower. In in vivo experiments diastolic pressures are usually in the range of 70 to 85 mmHg. When saline-like solutions are used, such high pressures cannot be used in isolated hearts because tissue oedema will develop due to the lower coronary vascular resistance (leading to higher intracapillary filtration pressures), in conjunction with the low colloid osmotic pressure of these solutions. The use of whole blood could circumvent these problems initially, but will lead to complications caused by foaming, clotting, haemolysis and thereby considerably shortens the period during which the preparation remains stable. Most investigators have therefore added substances to their perfusion medium, which increase the osmotic pressure or increase tissue pressure, by immersing the heart in an organ bath filled with perfusion medium. Interstitial oedema contributes to nonuniform perfusion when perfusion is restored after ischaemia, which is a limitation when recovery of post-ischaemic function or ischaemic preconditioning is studied. Microvascular compression does not only result from interstitial oedema but may also result from the development of myocardial contracture [68, 69]. The ensuing vascular deformation can partially be prevented by placing a fluid-filled balloon inside the left ventricular cavity [70].

The isolated heart model is well suited to study the effects of acute myocardial ischaemia or hypoxia [71]. Different degrees of ischaemia can be produced by completely stopping or partially restricting perfusion, while different degrees of hypoxia can be produced by removing oxygen completely or partially from the perfusion medium. An advantage of the isolated heart model is that one can study in a very simple manner whether the consequences of ischaemia are due to either lack of oxygen supply or accumulation of metabolites resulting from anaerobic metabolism [72]. Depending on the hypothesis that is investigated it may either be an advantage or a disadvantage that the influence of the haemodynamic and neurohumoral factors, which are present in the intact animal, are eliminated. However, caution must be exerted when interpreting the effects of pharmacological interventions on acute ischaemia. In such studies quite often concentrations of pharmacological agents are used that are not tolerated by the intact animal. For instance, it has been concluded from isolated heart studies that dihydropyridine derivatives protect the myocardium by a negative inotropic action [73], while in vivo experiments clearly demonstrate that the doses used in the isolated heart studies cannot be used in the intact animal because of severe hypotension secondary to their vasodilator actions [74].

The Langendorff preparation does not permit the left ventricle to eject the perfusate (cardiac output) and is therefore a non-working model. Some 30 years ago Neely et al. [75]developed a work performing isolated heart model, which was capable of ejecting perfusate. In order to do so, the perfusate is now supplied by a cannula inserted into the left atrium and left ventricular outflow can be monitored, while left atrial pressure or aortic pressure can be controlled. The advantage of the working heart model is that it allows the construction of cardiac function curves using Windkessel like models under a wide variety of conditions such as during the post-ischaemic recovery period.

Although functional studies can be performed in isolated heart models, they are particularly useful for the study of metabolic pathways because the content of the substrates in the perfusion solution (usually Krebs–Henseleit bicarbonate buffer) can be altered, while the metabolic products can be determined in the effluent. Finally, at the end of the experiment myocardial tissue is easily accessible for biochemical and histological analysis. Rat and rabbit hearts are most frequently used in isolated heart studies, but the use of hearts from larger animals such as pigs is not uncommon [76].

In addition to using the whole heart, some models have employed only the intraventricular septum [77]. For a review of this and other simple models the reader is referred to ref. [2].

Finally, although isolated heart studies can provide useful information about the events that occurs during ischaemia and early reperfusion, it should never be forgotten that extrapolation of the results to what is actually happening in regional ischaemia in man should occur with extreme care. For instance, the perfusion medium of the isolated hearts lacks blood components such as leukocytes, which have been implicated in the development of myocardial injury associated with ischaemia and reperfusion. Furthermore, with global ischaemia one eliminates the interaction between the myocytes of different regions of the heart, which may have an impact on the ultimate function of regionally ischaemic myocardium [78].

2.4 Isolated cardiac myocytes
During the last 25 years isolated myocytes have emerged as a new experimental model and their applications are still evolving. Among them are studies under anoxic or ischaemic conditions, but routinely isolated myocytes are most frequently used in electrophysiological and biochemical experiments. Because the cells are free from surrounding tissue, all changes can be attributed to processes occurring within the cell. Cardiac myocytes are available as either cultured embryonic or adult cells or freshly isolated adults cells. Details of the different types and isolation procedures have been described extensively (see [79]). An important determinant of the quality of the preparation is the oxygen consumption, which must match that of the intact quiescent heart [80, 81]. Contraction of the myocytes can be assessed by optical (e.g. laser diffraction, photodiode array edge detectors, and direct visual imaging) or mechanical detectors [82–84]. Critical for all studies is the viability and stability of the preparation. It is therefore mandatory to determine that the percentages of viable cells at the beginning and at the end of the study do not differ when interventions are performed that do not produce irreversible injury. Membrane damage can be evaluated by release of cytoplasmic enzymes such as lactate dehydrogenase and creatine phosphokinase [85]. Another reason for concern is that functional changes may occur during the isolation procedures. For instance, Linden et al. [86]have shown that exposing myocytes to catecholamines leads to down-regulation of the number of receptors within minutes.


   3 Ischaemic syndromes
Top
 1 Introduction
 2 Myocardial ischaemia
 3 Ischaemic syndromes
4 Future developments
 References
 
3.1 Stunning
In 1975 Heyndrickx et al. [87]reported that recovery of regional contractile function after a 10-minute coronary artery occlusion in conscious dogs was not immediate but took hours to days. In contrast, the electrocardiogram normalized almost instantaneously. This delayed recovery of post-ischaemic function, which initially received little attention and was even considered to be an experimental artefact [88], was later termed stunning [89]and the search for its mechanism has greatly contributed to our current knowledge of events occurring during ischaemia and reperfusion. By definition, function of stunned myocardium ultimately recovers without intervention. Stunning is therefore only clinically relevant for those patients in which global left ventricular function is already compromised through other cardiovascular conditions such as previous infarction.

Because myocardial stunning is the consequence of a period of reversible ischaemia, it follows that every model that is used for the study of brief ischaemia can serve as a model for stunning by merely showing that contractile function will recover when the period of reperfusion is extended sufficiently.

3.1.1 In vivo models
Studies in chronically instrumented awake animals (dogs, pigs or baboons) are most suitable for the study of stunning, as the time course of recovery of the post-ischaemic function can be followed for periods up to weeks or months in the absence of complicating factors such as anaesthesia, artificial ventilation and variations in temperature (see [90]). Because the degree of stunning is determined in part by the ischaemic burden i.e. duration and severity of the flow (O2) deficit [91, 92], determination of residual perfusion during ischaemia is mandatory in animal species with an extensive coronary collateral circulation such as the dog. Without collateral flow data, the effect of an intervention on myocardial stunning cannot be properly interpreted. An additional advantage of the conscious animal model is that it can serve as its own control, provided that sufficient long periods are allowed between two successive stunning protocols. However, before pharmacological interventions are evaluated in such models, it is mandatory to establish that the successive occlusions produce the same degree of stunning. With respect to this, one has to keep in mind that Shen and Vatner [93]recently reported significant species differences for the same experimental protocols. Thus, while in pigs the degree of stunning was the same after two consecutive occlusions separated by 24 hours, it was less after the second occlusion in dogs, an observation which might be related to recruitment of collateral flow in the latter species. Because of cost and complexity of the awake animal models, most investigators use (open or closed chest) anaesthetized animals, thereby accepting that full recovery of function cannot be achieved in the limited time span that is available. An important difference between the awake and anaesthetized models involves the degree of stunning and the time course of recovery [42, 91]. Thus, when the same experimental protocols are used, post-ischaemic dysfunction is more severe in pentobarbital anaesthetized dogs than in awake dogs [42], to which the less significant production of oxygen derived free radicals in awake animals may contribute. An explanation might be that pentobarbital anaesthesia, in part via altering haemodynamic conditions, results in higher oxygen demand at the onset of ischaemia and thus increases the total ischaemic burden [42]. The effect of anaesthesia on myocardial oxygen demand and its importance for the degree of stunning is supported by canine studies in which halothane [94]and isoflurane [95]anaesthesia, were compared to fentanyl [94]and morphine/{alpha}-chloralose/urethane anaesthesia [95]. Volatile anaesthesia was found to result in better recovery of contractile function following a 15-minute coronary artery occlusion. Another variable that influences functional recovery of stunned myocardium is temperature [42]. Consequently, a rigorous control of body temperature is mandatory when studying myocardial stunning. Open-chest animal preparations are more susceptible to temperature variations than closed chest animals.

Most investigators have used a single coronary artery occlusion, which produces stunning, provided that its duration is sufficiently long [96], but multiple occlusions have also been used. Of interest is that when a large number of occlusions is used there is a progressive loss in the relation between the degree of stunning and residual perfusion during ischaemia [97], a relation which is prominently present when a single 15-minute coronary occlusion is used [91]. Furthermore, partial coronary artery occlusions of sufficient duration [98, 99], or in the presence of exercise-induced ischaemia [100]also produce stunning. Finally, rapid ventricular pacing for 24 to 48 hours leads to post-pacing dysfunction which has all the characteristics of global myocardial stunning [101, 102], although pacing did not lead to demonstrable ischaemia. Global stunning can also be produced by cardioplegia and sequential regional ischaemia. To this end, Heyndrickx et al. [103]applied one hour flow reductions of the left circumflex coronary artery and the left anterior descending coronary artery at 30-minute intervals.

Recovery of post-ischaemic contractile function is sometimes also examined in models which employ periods of ischaemia that are sufficiently long to produce a mixture of both reversible and irreversible myocardial injury [104–106]. In these studies regional post-ischaemic function will not return to pre-ischaemia values and functional data obtained during the first hours of reperfusion period do not reflect recovery of stunned myocardium and are therefore difficult to interpret.

3.1.2 Isolated perfused hearts
Even more so than the in situ anaesthetized animal model, the isolated perfused heart model is limited by the relatively short period in which recovery of post-ischaemic function can be studied. Because full recovery of post-ischaemic function is rarely observed in isolated heart studies, one must include other pertinent features of stunning such as absence of irreversible injury, response to inotropic stimulation and the rebound in creatine phosphate to supranormal levels [107, 108]to assure that one is dealing with models of pure stunning. With respect to this, it should be noted that durations of ischaemia that do not result in myocardial necrosis under in vivo conditions, may already produce irreversible damage in isolated perfused hearts. For instance, Borgers et al. [109]observed that even a 15-minute period of global ischaemia in isolated rabbit hearts irreversibly damaged a significant number of myocytes.

In isolated hearts, stunning has been produced by single or repeated brief periods of total global ischaemia, by transient occlusion of a coronary artery producing regional ischaemia [110], and by pacing-induced tachycardia in the presence of a lowered coronary perfusion pressure thereby producing global demand ischaemia [111]. In addition to hearts of rats, also hearts of rabbits [109, 112], ferrets [113]and pigs [76]have been used in isolated heart models to study stunning.

3.1.3 Limitations of functional parameters
Understanding the limitations of the parameters that are used to describe contractile function is essential for proper assessment of the degree of stunning. Because in vitro studies usually employ models of global myocardial ischaemia, global cardiac function parameters such as left ventricular developed pressure (isovolumically beating hearts) and cardiac output (working hearts) are used to assess the degree of stunning. In in vivo experiments regional stunning is most commonly assessed by measurement of local contractile function, such as regional segment shortening or wall thickening. These parameters depend on pre- and afterload, which may be different from baseline during the post-ischaemic period. In addition, the afterload-dependency is more pronounced in stunned compared to normal myocardium [114], so that when the effect of an intervention on stunning is investigated, functional measurements should be made under identical or at least very similar loading conditions. One approach to correct for decreased afterload in in vivo preparations is by partially clamping the aorta or by inflation of a balloon positioned in the aorta. Alternatively, changes in left ventricular pressure can be included in the assessment of contractile performance by using parameters derived from the left ventricular pressure-volume (time varying elastance concept) or left ventricular pressure-segment length (or wall thickness) relations such as external work and end systolic elastance [114–116].

3.2 Ischaemic preconditioning
In 1985 Murry et al. [117]described that in normal dog myocardium the development of irreversible damage was delayed when a sustained coronary artery occlusion was preceded by brief periods of reversible myocardial ischaemia, a phenomenon which was termed ischaemic preconditioning. To date ischaemic preconditioning is the most potent endogenous mechanism by which infarct size can be limited. Since the original description by Murry et al. [117], numerous studies have been undertaken to elucidate the mechanism of ischaemic preconditioning and to investigate whether the protection by ischaemic preconditioning also applies to other endpoints than infarct size (e.g. arrhythmias and recovery of function following reversible ischaemia) and to demonstrate its existence and relevance in the clinical setting [118, 119]. When infarct is taken as endpoint, the clinical importance of ischaemic preconditioning has yet to be convincingly demonstrated. The lack of the definite proof for the occurrence of ischaemic preconditioning in man arises from the fact that in patients the accurate assessment of determinants of infarct size such as the area at risk, the residual (collateral) blood flow, the duration and completeness of the infarct-producing coronary artery occlusion, and also the characteristics of the preceding ischaemia that serves as preconditioning-stimulus, are extremely difficult [120]. Furthermore, patients have usually pre-existing laesions and a long history of angina pectoris before infarction develops, which gives rise to further caution in extrapolating data obtained in normal animal hearts. Two important but unanswered questions regarding the clinical implication of ischaemic preconditioning therefore are: ‘does recurrent angina lead to a preconditioned state?’ and ‘can the as yet incompletely known mechanism of preconditioning be mimicked pharmacologically?' Cohen et al. [121]have shown that awake rabbits become tolerant to multiple episodes of ischaemic preconditioning. With respect to this it is important to investigate if tolerance develops when pharmacological agents, that mimic ischaemic preconditioning, are administered chronically.

The choice of the experimental animal does not appear to matter because positive results have been obtained in each species investigated in which infarct size was chosen as endpoint. These species now include dogs [117], pigs [12], sheep [123], rabbits [11], rats [10], marmots [124]and ferrets [125]. There is some evidence that the mechanism underlying ischaemic preconditioning may not be the same in all species [126]. For instance, the role of activation of protein kinase C is still a very much debated issue [127, 128]. Similar controversies exist about the activation of ATP sensitive K+ channels and adenosine. However, such data should not play a decisive role in the choice of the experimental species as long as it has not been established whether such a mechanism could be operative in man.

At variance with infarct size limitation, studies using other endpoints such as ventricular arrhythmias or attenuation of stunning have yielded less uniform results. Protection by ischaemic preconditioning against arrhythmias has conclusively been reported in studies using rats, but its protective effect in large animals is controversial [129, 130]. It also appears that in the same species, the stimuli required to precondition the heart against necrosis and arrhythmias may be different [131]. Presently, it is unclear if these different results are due to differences in species or to differences in experimental protocols. The age of the animal might need attention as Tani et al. [132]reported that in middle aged rats (50 weeks) the beneficial effects of ischaemic preconditioning against reperfusion ventricular fibrillation and the rate of sarcoplasmic reticular Ca2+ uptake were reversed compared to young rats (12 weeks). Although it has been well established that senescent animals (and also patients) are more vulnerable to the consequences of ischaemia than young animals, it should be taken into account that the observations by Tani et al. [132]were made in isolated hearts, while using an endpoint of ischaemic preconditioning that is still controversial. Attenuation of stunning by ischaemic preconditioning is also still controversial. It now appears that classical preconditioning does not protect against stunning, but that the second window of protection, which occurs 24 to 72 hours after the preconditioning stimulus has been applied, offers protection in large animals [133].

Principally, all models used in the study of myocardial ischaemia are suited to study ischaemic preconditioning. In order to mimic the clinical situation as closely as possible awake animals are preferred, but producing myocardial infarction in animals without anaesthesia meets almost insurmountable resistance from many ethical committees on animal experimentation. As a matter of fact there are only very few studies in awake animals in which ischaemic preconditioning with infarct size as endpoint has been investigated [121, 133].

In in vivo studies, a wide variety of anaesthesia regimens have been used, quite often depending on the species. Because the precise mechanism of ischaemic preconditioning is unknown, one should carefully describe the agents used to anaesthetize animals as anaesthetic agents may affect the magnitude of protection. This is supported by studies of Haessler et al. [134]who showed that using identical experimental protocols, protection in pentobarbital anaesthetized rabbits was significantly greater than in isoflurane- or ketamine/xylazine-anaesthetized rabbits, which was not caused by differences in area at risk, temperature, heart rate or arterial blood pressure. Infarct sizes of the control groups were reported to be similar, but there was a clear tendency for the isoflurane-treated animals to have smaller infarcts, an observation, which has later been firmly established by Cope et al. [135]. A single 5-minute coronary artery occlusion, which was an effective preconditioning stimulus in pentobarbital anaesthetized dogs without premedication of the opioid analgesic butorphanol [136, 137], was not sufficient to protect the myocardium after premedication of dogs with butorphanol [138], the threshold for preconditioning was increased as two episodes of 5-minute occlusions were required to provide an effective preconditioning stimulus.

Ytrehus et al. [139]have reported that infarct size development and the protection by ischaemic preconditioning are very similar for in situ and isolated Krebs–Henseleit bicarbonate buffer-perfused rabbit heart. Downey and Yellon [140]have pointed out that this observation strongly supports the use of the isolated heart models, because in these models the pharmacology of preconditioning can be better studied as administration of drugs can be precisely controlled and drugs that are not tolerated when administered systemically can be administered directly to the heart. However, of the rat heart, which is also used in in situ experiments and isolated heart preparations, it is unknown whether the degree of protection is similar in these two models [141]. Isolated cardiomyocytes [142, 143]also eliminate the interaction with non-cardiac tissue. The major attraction of this model is that it is very inexpensive and therefore permits studies using agents and techniques, which cannot be afforded otherwise.

However, the lack of innervation and extracardiac factors in isolated hearts and myocytes may be a shortcoming as evidence is now accumulating that brief supply or demand ischaemia in other organs (kidney, small intestine and skeletal muscle) is capable of delaying myocardial infarct development produced by a coronary artery occlusion [144, 145]. Therefore, it appears that to fully appreciate the processes by which the myocardium can be protected, the intact animal is the most appropriate model.

3.2.1 Experimental design
The experimental protocol of an ischaemic preconditioning study consists of five distinct phases (i) stabilization period after surgery, (ii) the preconditioning stimulus, (iii) a period of intermittent reperfusion, (iv) a prolonged period of test ischaemia during which infarction develops and (v) a period of reperfusion at the end of which the effect of the preconditioning stimulus on infarct size is assessed. Each of these phases require some comment;

(i) Sandhu et al. [146]observed in anaesthetized rabbits that a single 5-minute coronary artery occlusion limited infarct size development during a subsequent 30-minute occlusion when the preconditioning stimulus was given shortly (5 minutes) after the surgical procedures were completed, but was ineffective when the stabilization period was extended to 30 minutes. The reason for this observation is unknown but the authors speculated that the stress associated with surgery may have resulted in systemic release of catecholamines, a possible mechanism involved in ischaemic preconditioning [147]. This observation may be important for the design of experimental protocols, in which different groups of animals are studied (for instance comparison of the protective effect of a single versus multiple brief occlusions). Some investigators avoid differences in the duration of the experimental protocol by adjusting the duration of the stabilization period. But they should thus be aware that this could blur their data.

(ii) In the vast majority of studies, the preconditioning stimulus consists of one or multiple brief periods of ischaemia produced by abrupt occlusion and reperfusion of a coronary artery [132]. The preconditioning stimulus should produce no or only negligible necrosis. Depending on the severity of the flow reduction, partial coronary artery occlusions, alone [148]or in the presence of atrial pacing in order to produce demand ischaemia [149], can be used to precondition the myocardium.

(iii) Intermittent reperfusion between the preconditioning stimulus and the period of test ischaemia is required when complete occlusions are used to precondition the myocardium, but may not be necessary when a partial occlusion is used to either precondition the myocardium [150, 151]or during the test period of ischaemia [152]. The duration of the intermittent reperfusion period is an important determinant for the outcome of preconditioning studies as, depending on the species, protection is lost after a few hours (classic preconditioning or first window of protection). However, in some species protection reappears after 24–48 hours (second window of protection), illustrating that ischaemic preconditioning is, at least in some species, a biphasic phenomenon [153, 154].

(iv) As far as the duration of the infarct producing sustained coronary artery occlusion is concerned, one should take into account that infarct size development is different in the various species [155]. Ischaemic preconditioning has been shown in animals with [117]and without [122]a collateral circulation, implying that the extent of the collateral circulation is not crucial for the occurrence of ischaemic preconditioning. However, collateral blood flow is a major determinant of infarct size and in order to obtain a proper assessment of infarct size collateral blood flow should always be determined in species with a collateral circulation. Body temperature is another important determinant of infarct size development with the magnitude of the effect depending on the duration of the coronary artery occlusion [156–158]. The need for rigorous control of temperature also follows from the observation that protection by ischaemic preconditioning is extended to longer occlusion durations when the experiments are performed at lower body temperature [159].

Abrupt and complete reperfusion after the sustained occlusion has been a feature of all preconditioning studies except in the study by Kapadia et al. [160]. The latter studied ischaemic preconditioning in closed chest pigs by subjecting them to instrumentation with a percutaneous transluminal angioplasty catheter on which an artificial stenosis (82% diameter reduction and 30% flow reduction) was mounted just proximal to the balloon. Because the stenosis was left in place during the entire experimental protocol, reperfusion was incomplete. Although the stenosis was not severe enough to precondition the myocardium by itself, the presence of the stenosis did not prevent preconditioning by multiple balloon inflations. The presence of a stenosis during reperfusion has yielded conflicting reports on infarct size as no changes [161, 162]as well as increases [163, 164]have been reported. In trying to explain these differences, one must keep in mind that the presence of a stenosis during reperfusion could lead to both persistent endocardial ischaemia as well as a limitation of the reperfusion-induced damage by preventing an excessive burst of oxygen-derived free radicals. Nevertheless, a nice feature of the model by Kapadia et al. [160]is that it resembles a clinical condition (patients with a coronary artery stenosis which becomes totally obstructed by thrombus and is treated with thrombolysis) more closely than any of the abrupt total occlusion – total reperfusion models.

(v) The vast majority of studies use the ratio of infarcted area (IA) and the area at risk (AR) to assess infarct size limitation by ischaemic preconditioning. In the further analysis it is then assumed that this ratio is independent of the area at risk i.e. the relation between AR and IA is proportional. Several groups of investigators have however shown that this is true for rats [165], but not for pigs [151]and rabbits [165]i.e. the linear regression line relating infarcted area and area at risk has a zero-intercept on the AR-axis. This implies that the mathematical description of this relationship is IA=a*AR + b or IA/AR=a + b/AR. Especially for small AR, IA/AR depends on AR. Thus unless AR exceeds 20% of the left ventricular mass, IA/AR should be used with care, in particular when studies are compared in which the areas at risk differ substantially.

3.3 Hibernation
While myocardial stunning and ischaemic preconditioning were originally laboratory observations, the concept of myocardial hibernation originates from clinical observations that in some patients with coronary artery disease and chronic left ventricular dysfunction, wall function improved after coronary bypass surgery [166]. The decreased wall motion before surgery was therefore viewed as a chronic adaptive response to a reduced coronary blood flow, with the reduction in contractility being reversible upon reinstatement of normal myocardial perfusion. The clinical importance of hibernation is that viability of myocardium is preserved due to restoration of aerobic metabolism, in the presence of a chronic blood flow reduction [167]. In an earlier issue of this journal, Hearse [1]made therefore a distinction between physiological and biochemical ischaemia. In the latter condition the myocardium is metabolically distressed and cell death will evolve when this condition is maintained for a prolonged period of time. Hibernating myocardium can therefore be biochemical ischaemic for only brief periods, while its function is depressed chronically because of subnormal flow i.e. the myocardium is chronically physiologically ischaemic [167]. The most difficult modelling problem in hibernation is therefore how to achieve a state of reduced flow (function) which does not cause infarction.

Hibernation should, by definition, be studied in chronic models, which implies the use of awake animals in which blood flow in a coronary artery is chronically and sufficiently reduced such that the reduced coronary blood flow leads to a permanent regional dysfunction (which can be assessed continuously by sonomicrometry) but not to cell death. Regional metabolism should also be monitored continuously in the distribution area of the obstructed coronary artery and preferentially also in the normal myocardium to determine the nature of the metabolic adaptation. Essential for these models is that the function of the hibernating myocardium must recover after blood flow is normalized. Finally, at the end of the study the post-hibernating myocardium should be examined for the absence of cell death. None of the current models have met all these requirements, although in several studies chronically instrumented animals (pigs and dogs) have been employed in which regional dysfunction was produced for several days to months by partial narrowing of a coronary artery [52, 98, 168–171]. A further limitation of the existing models is that concentric stenoses have been applied, thereby excluding a potential role of the local intact endothelium as would have been possible if eccentric stenoses were used. Because of the lack of coronary collaterals the use of the pigs appears to be advantageous, when the flow restriction in the native coronary artery can be controlled. However, Millard et al. [28]have shown that coronary collaterals also develop in pigs with a fixed stenosis.

Based on the results of some studies in man [172, 173]and in animals [174], some investigators have hypothesized that chronic hibernation is nothing else than chronic stunning resulting from repetitive stress-induced ischaemia, rather than a chronic downregulation of function secondary to a decreased coronary blood flow. If true, flow might be normal or almost normal under resting conditions and even be increased during exercise-induced ischaemia, while function should return to normal after a sufficiently long stress-free period. To test this hypothesis continuous assessment of regional myocardial function in patients with hibernating myocardium is required. If confirmed, brief repetitive coronary artery occlusions in the presence and possibly even in the absence of a mild stenosis, could produce hibernation.

In several studies, hibernation has been produced by partially restricting coronary artery flow during periods ranging from only a few hours up to a few days [175–179]. In these ‘short-term hibernation models’ the essential characteristics of chronic hibernation cannot be studied. The usefulness of these short-term models is therefore limited, but they may provide some insight into the triggers leading to hibernation. However, for ‘each short-term hibernation model' it should be shown that chronic hibernation develops if the model is used in chronic studies. Otherwise the putative triggers identified in the ‘short-term hibernation models' cannot be related to the development of hibernation. Isolated heart models using sequences of no-flow and low-flow ischaemia have also been employed to study metabolic adaptation as a possible trigger for hibernation [180]. This model is not only limited by the short time in which it can be studied, but has also the disadvantage that it produces global ischaemia, and lacks the interaction with the other organs (neural reflexes) which may be vital in the development of hibernation. For these reasons isolated heart models and other less well advanced models such as isolated cardiomyocytes appear not to be very useful in the study of hibernation at the present time.


   4 Future developments
Top
 1 Introduction
 2 Myocardial ischaemia
 3 Ischaemic syndromes
 4 Future developments
References
 
Molecular biologists have provided physiologists in recent years with new animal models of cardiovascular disease by altering the mammalian genome through introduction or modification of genes. The mouse has become the animal of choice because their genome is well characterized, their cost is relatively low and a large series of animals can be obtained in a short period of breeding [181]. Assessment of cardiac function in these transgenic mice (e.g. gene-targeted knock-out) is essential for understanding the functional consequences of the manipulation of the genome. Because in the past measurements of cardiovascular function in mice have been rare, the possibilities to determine cardiac function in this species are just developing [182]. In recent years, however, we have witnessed the development of miniaturized Langendorff preparations of isolated (work performing) mouse hearts for measurement of left ventricular contractility and relaxation [183, 184], in situ measurements of aortic and left ventricular pressure [185, 186], the use of microsphere and dilution techniques for the determination of blood flow and volumes [187], the application of echocardiography for assessment of left ventricular dimensions, wall thickness, mass, circumferential fibre shortening and left ventricular wall stress-shortening relationships [188–190], and the application of Doppler echocardiography to obtain tracings from the left ventricular outflow tract for estimating peak blood velocity [189]. Desai et al. [191], employing techniques, previously used in large animals, have described the feasibility of chronic monitoring haemodynamic and metabolic parameters in conscious mice at rest and during exercise.

The purpose of the aforementioned studies was to improve the ability to examine the phenotypic changes after gene targeted manipulations rather than the study of myocardial ischaemia and ischaemic syndromes. These transgenic models will, however, become useful for the study of processes involved in myocardial ischaemia and ischaemic syndromes, once we have learned to apply our knowledge and techniques obtained in large animal studies.

Time for primary review 29 days.


   Acknowledgements
 
This study has been supported by grants N.H.S. 95.103 and D96.024 from the Netherlands Heart Foundation. Dr Duncker is supported by an ‘Academie Onderzoeker’ fellowship from the Royal Netherlands Academy of Arts and Sciences.


   References
Top
 1 Introduction
 2 Myocardial ischaemia
 3 Ischaemic syndromes
 4 Future developments
 References
 

  1. Hearse D.J. Myocardial ischaemia: can we agree on a definition for the 21st century? Cardiovasc Res (1994) 28:1737–1744.[Free Full Text]
  2. Reimer KA, Jennings RB. Myocardial ischemia, hypoxia and infarction. In: Fozzard HA, Haber E, Jennings RB, Katz AM, Morgan HE, editors. The Heart and Cardiovascular System. New York: Raven Press, 1992:1875–1973.
  3. Bustad LK, McClennan RO. Swine in biomedical research. Pacific Northwest Laboratory, Richland, WA, 1966.
  4. Tumbleson ME. Swine in biomedical research. New York and London: Plenum Press, 1986.
  5. Da Vinci L. Instituto Graphico de Agostini. New York: Reynal, 1956:383.
  6. Buckley N.M, Gootman P.M, Yellin E.L, Brazeau P. Age-related cardiovascular effects of catecholamines in anesthetized piglets. Circ Res (1979) 45:282–292.[Abstract/Free Full Text]
  7. Gootman P.M, Goodman N, Buckley B.J. Maturation of central autonomic control of circulation. Fed Proc (1983) 42:1648–1655.[ISI][Medline]
  8. Hanson L.E. Fifty years progress in swine nutrition. J Animal Sci (1958) 17:1029–1057.[Abstract/Free Full Text]
  9. Selye H, Bajusz E, Grasso S, Mendel P. Simple techniques for the surgical occlusion of coronary vessels in the rats. Angiology (1960) 11:398–407.[Free Full Text]
  10. Liu Y, Downey J.M. Ischemic preconditioning protects against infarction in rat heart. Am J Physiol (1992) 263:H1107–H1112.[ISI][Medline]
  11. Cohen M.V, Liu G.S, Downey J.M. Preconditioning causes improved wall motion as well as smaller infarcts after transient coronary occlusion in rabbits. Circulation (1991) 84:341–349.[Abstract/Free Full Text]
  12. Obata T, Hosokawa H, Yamanaka Y. In vivo monitoring of norepinephrine and OH generation on myocardial ischemic injury by dialysis technique. Am J Physiol (1994) 266:H903–H908.[ISI][Medline]
  13. Pfeffer J.M, Pfeffer M.A, Fletcher P.J, Braunwald E. Progressive ventricular remodeling in rat with myocardial infarction. Am J Physiol (1991) 260:H1406–H1414.[ISI][Medline]
  14. Pfeffer J.M, Pfeffer M.A. Angiotensin converting enzyme inhibition and ventricular remodeling in heart failure. Am J Med (1988) 84:37–44.[ISI][Medline]
  15. Harken A.H, Simson M.B, Hasilgrove J, Wetstein L, Harden W.R 3rd, Barlow C.H. Early ischemia after complete coronary ligation in the rabbit, dog, pig, and monkey. Am J Physiol (1981) 241:H202–H210.[ISI][Medline]
  16. Schaper W. The collateral circulation of the heart. In: Black DAK, editor, Clinical Studies. Amsterdam: North Holland Publishing Company, 1971.
  17. Schaper W, Jageneau A, Xhonneux R. Development of collateral circulation in the pig and the dog heart. Cardiologia (1967) 51:321–335.[Medline]
  18. Schaper W, Bernotat-Danielowski S, Nienaber C, Schaper J. Collateral Circulation. In: Fozzard HA, Haber E, Jennings, RB, Katz AM, editors. The Heart and Cardiovascular System. New York: Raven Press, 1992:1427–1464.
  19. White F.C, Bloor C.M. Coronary collateral circulation in the pig: correlation of collateral flow with coronary bed size. Basic Res Cardiol (1981) 76:189–196.[CrossRef][ISI][Medline]
  20. Shen Y.T, Fallon J.T, Iwase M, Vatner S.F. Innate protection of baboon myocardium: effects of coronary artery occlusions and reperfusion. Am J Physiol (1996) 270:H1812–H1818.[Medline]
  21. Crozatier B, Ross J Jr., Franklin D, Bloor C.M, White F.C, Tomoike H, McKown D.P. Myocardial infarction in the baboon: regional function and the collateral circulation. Am J Physiol (1978) 235:H413–H421.[ISI][Medline]
  22. Winkler B, Sass S, Binz K, Schaper W. Myocardial blood flow and myocardial infarction in rats, guinea, pigs and rabbits (Abstract). J Mol Cell Cardiol (1984) 16:48.
  23. Schaper W. Experimental infarcts and the microcirculation. In: Hearse DJ, Yellon DM, editors. Therapeutic Approaches to Myocardial Infarct Size Limitation. New York: Raven Press, 1984:79–90.
  24. Franklin D, McKown D, McKown M, Hartley J, Caldwell M. Development and regression of coronary collaterals induced by repeated, reversible ischemia in dogs (Abstract). Fed Proc (1981) 40:339.
  25. Yamamoto H, Tomoike H, Shimokawa H, Nabeyama S, Nakamura M. Development of collateral function with repetitive coronary occlusion in a canine model reduces myocardial reactive hyperemia in the absence of significant coronary stenosis. Circ Res (1984) 55:623–632.[Abstract/Free Full Text]
  26. Kersten J.R, McGough M.F, Pagel P.S, Tessmer J.P, Warltier D.C. Temporal depence of coronary collateral development. Cardiovasc Res (1997) 34:306–312.[Abstract/Free Full Text]
  27. Litvak J, Siderides L.E, Vineberg A.M. The experimental production of coronary artery insufficiency and occlusion. Am Heart J (1957) 53:505–518.[CrossRef][ISI][Medline]
  28. Millard R.W. Induction of functional coronary collaterals in the swine heart. Basic Res Cardiol (1981) 76:468–473.[CrossRef][ISI][Medline]
  29. Scheel K.W, Ingram L.A, Wilson J.L. Effects of exercise on the coronary and collateral vasculature of beagles with and without coronary occlusion. Circ Res (1981) 48:523–530.[Abstract/Free Full Text]
  30. Sanders M, White F.C, Peterson T.M, Bloor C.M. Effects of endurance exercise on coronary collateral blood flow in miniature swine. Am J Physiol (1978) 234:H614–H619.[ISI][Medline]
  31. Cohen M.V, Yipintsoi T, Scheuer J. Coronary collateral stimulation by exercise in dogs with stenotic coronary arteries. J Appl Physiol (1982) 52:664–671.[Abstract/Free Full Text]
  32. Chilian W.M, Mass H.J, Williams S.E, Layne S.M, Smith E.S, Scheel K.W. Microvascular occlusions promote coronary collateral growth. Am J Physiol (1990) 258:1103–1111.
  33. Bloor CM, White FC, Lammers RJ. Cardiac ischemia and coronary blood flow in swine. In: Stanton HC, Mersmann HJ, editors. Swine in cardiovascular research. Boca Raton, Florida: C.R.C. Press Inc, 1986:87–104.
  34. Domenech R.J, Hoffman J.I, Noble M.I, Saunders K.B, Henson J.R, Subijanto S. Total and regional coronary blood flow measured by radioactive microspheres in conscious and anesthetized dogs. Circ Res (1969) 25:581–596.[Abstract/Free Full Text]
  35. Heyman M.A, Payne B.D, Hoffman J.I, Rudolph A.M. Blood flow measurements with radionuclide-labeled particles. Prog Cardiovasc Dis (1977) 20:55–79.[CrossRef][ISI][Medline]
  36. Kowallik P, Schulz R, Guth B.D, et al. Measurement of regional myocardial blood flow with multiple colored microspheres. Circulation (1991) 83:974–982.[Abstract/Free Full Text]
  37. Hale S.L, Alker K.J, Kloner R.A. Evaluation of nonradioactive, colored microspheres for measurements of regional myocardial blood flow in dogs. Circulation (1988) 78:428–434.[Abstract/Free Full Text]
  38. Gallagher K.P, Matsuzaki M, Koziol J.A, Kemper W.S, Ross J Jr. Regional myocardial perfusion and wall thickening during ischemia in conscious dogs. Am J Physiol (1984) 247:H727–H738.[ISI][Medline]
  39. Manders W.T, Vatner S.F. Effects of sodium pentobarbital anesthesia on left ventricular function and distribution of cardiac output in dogs, with particular reference to the mechanism for tachycardia. Circ Res (1976) 39:512–517.[Abstract/Free Full Text]
  40. Zimpfer M, Manders W.T, Barger A.C, Vatner S.F. Pentobarbital alters compensatory neural and humoral mechanisms in response to hemorrhage. Am J Physiol (1982) 243:H713–H721.[ISI][Medline]
  41. Merin R.G, Verdouw P.D, De Jong J.W. Myocardial functional and metabolic responses to ischemia in swine during halothane and fentanyl anesthesia. Anesthesiology (1982) 56:84–92.[ISI][Medline]
  42. Triana J.F, Li X.Y, Jamaluddin U, Thornby J.I, Bolli R. Post-ischemic myocardial ‘stunning’. Identification of major differences between the open-chest and the conscious dog and evaluation of the oxygen radical hypothesis in the conscious dog. Circ Res (1991) 69:731–747.[Abstract/Free Full Text]
  43. Ihara T, Shannon R.P, Komamura K, et al. Effects of anaesthesia and recent surgery on diastolic function. Cardiovasc Res (1994) 28:325–336.[Abstract/Free Full Text]
  44. Corday E, Lang T.W, Meerbaum S, et al. Closed chest model of intracoronary occlusion for study of regional cardiac function. Am J Cardiol (1974) 33:49–59.[CrossRef][ISI][Medline]
  45. Debley V.G. Miniature hydraulic occluder for zero blood flow determinations. J Appl Physiol (1971) 31:138–139.[Free Full Text]
  46. De Jong J.W, Verdouw P.D, Remme W.J. Myocardial nucleoside and carbohydrate metabolism and hemodynamics during partial occlusion and reperfusion of pig coronary artery. J Mol Cell Cardiol (1977) 9:297–312.[CrossRef][ISI][Medline]
  47. Elzinga W.E, Skinner D.B. Hemodynamic characteristics of critical stenosis in canine coronary arteries. J Thorac Cardiovasc Surg (1975) 69:217–222.[Abstract]
  48. Hosko M.J, Gross G.J, Warltier D.C. Technique for precise, graded arterial stenosis and occlusion. Bas Res Cardiol (1977) 72:651–659.[CrossRef][ISI][Medline]
  49. Van der Meer J.J, Reneman R.S. An improved technique to induce a standardized functional stenosis of a coronary artery. Eur Surg Res (1972) 4:407–418.[ISI]
  50. Liedtke A.J, DeMaison L, Eggleston A.M, Cohen L.M, Nellis S.H. Changes in substrate metabolism and effects of excess fatty acids in reperfused myocardium. Circ Res (1988) 62:535–542.[Abstract/Free Full Text]
  51. Schulz R, Rose J, Martin C, Brodde O.E, Heusch G. Development of short-term myocardial hibernation. Its limitation by the severity of ischemia and inotropic stimulation. Circulation (1993) 88:684–695.[Abstract/Free Full Text]
  52. Fedele F.A, Gewirtz H, Capone R.J, Sharaf B, Most A.S. Metabolic response to prolonged reduction of myocardial blood flow distal to a severe coronary artery stenosis. Circulation (1988) 78:729–735.[Abstract/Free Full Text]
  53. Haase J, Di Mario C, Slager C.J, et al. In-vivo validation of on-line and off-line geometric coronary measurements using insertion of stenosis phantoms in porcine coronary arteries. Cathet Cardiovasc Diagn (1992) 27:16–27.[ISI][Medline]
  54. Ichikawa Y, Yokoyama M, Akita H, Fukuzaki H. Constriction of a large coronary artery contributes to serotonin-induced myocardial ischemia in the dog with pliable coronary stenosis. J Am Coll Cardiol (1989) 14:449–459.[Abstract]
  55. Theroux P, Ross J Jr., Franklin D, Kemper W.S, Sasayama S. Coronary arterial reperfusion. III. Early and late effects on regional myocardial function and dimensions in conscious dogs. Am J Cardiol (1976) 38:599–606.[CrossRef][ISI][Medline]
  56. Theroux P, Ross J Jr., Franklin D, Kemper W.S, Sasayama S. Regional myocardial function in the conscious dog during acute coronary occlusion and responses to morphine, propranolol, nitroglycerin, and lidocaine. Circulation (1976) 53:302–314.[Abstract/Free Full Text]
  57. Hoit B.D, Lew W.Y. Functional consequences of acute anterior vs. posterior wall ischemia in canine left ventricles. Am J Physiol (1988) 254:H1065–H1073.[ISI][Medline]
  58. Sassen L.M.A, Duncker D.J.G.M, Gho B.C.G, Diekmann H.W, Verdouw P.D. Haemodynamic profile of the potassium channel activator EMD 52692 in anaesthetized pigs. Br J Pharmacol (1990) 101:605–614.[ISI][Medline]
  59. Headrick J.P, Emerson C.S, Berr S.S, Berne R.M, Matherne G.P. Interstitial adenosine and cellular metabolism during beta-adrenergic stimulation of the in situ rabbit heart. Cardiovasc Res (1996) 31:699–710.[Abstract/Free Full Text]
  60. Massie B.M, Schwartz G.G, Garcia J, Wisneski J.A, Weiner M.W, Owens T. Myocardial metabolism during increased work states in the porcine left ventricle in vivo. Circ Res (1994) 74:64–73.[Abstract/Free Full Text]
  61. McFalls E.O, Baldwin D, Palmer B, Marx D, Jaimes D, Ward H.B. Regional glucose uptake within hypoperfused swine myocardium as measured by positron emission tomography. Am J Physiol (1997) 272:H343–H349.[ISI][Medline]
  62. Bush L.R, Shebuski R.J. In vivo models of arterial thrombosis and thrombolysis. FASEB J (1990) 4:3087–3098.[Abstract]
  63. Langendorff O. Untersuchungen am überlebenden, S.äugerthierherzen. Pflügers Arch Ges Physiol (1895) 61:291–332.[CrossRef]
  64. Döring HJ, Dehnert H. The isolated perfused heart. According to Langendorff. In: BVM-Biomesstechnik V-Verlag March GmbH D-7806, 1988.
  65. Langendorff O. Zur Kenntniss des Blutlaufs in den Kranzgefässen des Herzens. Pflügers Arch Ges Physiol (1899) 78:423.
  66. Gottlieb R, Magnus R. Digitalis und Herzarbeit. Nach Versuchen am überlebenden Warmblüterherzen. Arch Exp Path Pharmakol (1904) 51:30–63.
  67. Heymans J.F, Kochmann F. Une nouvelle méthode de circulation artificielle à travers le coeur isolé de mammifère. Arch Int Pharmacodyn Thérap (1904) 5:379–386.
  68. Humphrey S.M, Gavin J.B, Herdson P.B. The relationship of ischemic contracture to vascular reperfusion in the isolated rat heart. J Mol Cell Cardiol (1980) 12:1397–1406.[CrossRef][ISI][Medline]
  69. Humphrey S.M, Thomson R.W, Gavin J.B. The effect of an isovolumic left ventricle on the coronary vascular competence during reflow after global ischemia in the rat heart. Circ Res (1981) 49:784–791.[Abstract/Free Full Text]
  70. Ganote C.E, Humphrey S.M. Effects of anoxic or oxygenated reperfusion in globally ischemic, isovolumic, perfused rat hearts. Am J Pathol (1985) 120:129–145.[Abstract]
  71. De Leiris J, Harding D.P, Pestre S. The isolated perfused rat heart: a model for studying myocardial hypoxia or ischaemia. Basic Res Cardiol (1984) 79:313–321.[CrossRef][ISI][Medline]
  72. Manning A.S, Hearse D.J, Dennis S.C, Bullock G.R, Coltart D.J. Myocardial ischaemia: an isolated, globally perfused rat heart model for metabolic and pharmacological studies. Eur J Cardiol (1980) 11:1–21.[ISI][Medline]
  73. De Jong J.W, Harm