Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Myocardial leukocyte transit time and retention during acute endotoxemia Goddard, Christopher Morris 1998

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata

Download

Media
831-ubc_1998-27148X.pdf [ 6.95MB ]
Metadata
JSON: 831-1.0088672.json
JSON-LD: 831-1.0088672-ld.json
RDF/XML (Pretty): 831-1.0088672-rdf.xml
RDF/JSON: 831-1.0088672-rdf.json
Turtle: 831-1.0088672-turtle.txt
N-Triples: 831-1.0088672-rdf-ntriples.txt
Original Record: 831-1.0088672-source.json
Full Text
831-1.0088672-fulltext.txt
Citation
831-1.0088672.ris

Full Text

MYOCARDIAL LEUKOCYTE TRANSIT TIME AND RETENTION DURING ACUTE ENDOTOXEMIA by CHRISTOPHER MORRIS GODDARD B.Med.Sci., The University of Alberta, 1989 M.D., The University of Alberta, 1991 A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY in THE FACULTY OF GRADUATE STUDIES (Department of Medicine, Experimental Medicine) We accept this thesis as conforming to the required standard . THE UNIVERSITY OF BRITISH COLUMBIA November 1997 © Christopher Morris Goddard . In presenting this thesis in partial fulfilment of the requirements for an advanced degree at the University of British Columbia, I agree that the Library shall make it freely available for reference and study. I further agree that permission for extensive copying of this thesis for scholarly purposes may be granted by the head of my department or by his or her representatives. It is understood that copying or publication of this thesis for financial gain shall not be allowed without my written permission. Department The University of British Columbia Vancouver, Canada DE-6 (2/88) ABSTRACT During sepsis, endotoxins are released into the circulation which cause the activation of both cellular and humoral mediators of inflammation. Following activation of the cytohumoral cascade of sepsis the transit time of leukocytes through the microvasculature of the heart is increased. Increased transit time contributes to the retention of increased numbers of activated leukocytes in the heart. Leukocytes are retained primarily within the capillaries of the heart. The neutrophil fraction of leukocytes is slowed and retained to a much greater extent than the lymphocyte fractions. The retention of activated neutrophils is mediated more by the effects of humoral inflammatory mediators on the heart than on the neutrophils themselves. The retention of increased numbers of activated neutrophils within the capillaries of the heart is associated with the development of myocardial edema and myocyte structural changes. These myocardial changes are associated with decreased myocardial contractile function. Reduced myocardial contractility contributes to the pathogenesis of the septic shock state which increases the mortality due to sepsis. Exclusion of activated neutrophils from the capillary circulation of the heart during sepsis protects the myocardium from pathologic changes associated with sepsis and decreases the loss of myocardial contractile function. ii TABLE OF CONTENTS Abstract. ii Table of contents. iii List of tables. vi List of figures. vii Acknowledgments. x CHAPTER I Introduction. 1 1.1 Sepsis and SIRS. 1 1.2 Myocardial dysfunction in sepsis. 4 1.3 Humoral mediators of sepsis. 6 1.4 Cellular mediators of sepsis. 7 1.5 Major Hypotheses. 8 CHAPTER II Leukocyte transit time and retention in the coronary microcirculation during porcine endotoxemia. 9 2.1 Hypothesis. 9 2.2 Experimental design. 9 2.21 Instrumentation. 10 2.22 Experimental protocol. 11 2.23 Measurements. 13 2.24 Data Analysis. 21 2.3 Results. 22 2.4 Discussion. 34 iii 2.5 Summary. 39 CHAPTER III Myocardial morphometry changes during sepsis; correlation with increased leukocyte retention and myocardial contractile failure. 41 3.1 Hypothesis. 41 3.2 Experimental design. 42 3.21 Instrumentation. 42 3.22 Experimental protocol. 48 3.23 Measurements. 49 3.24 Data Analysis. 52 3.3 Results. 52 3.4 Discussion. 58 3.5 Summary. 63 CHAPTER IV Fractional leukocyte transit time and retention during endotoxemia. . 65 4.1 Hypothesis. 65 4.2 Experimental design. 66 4.21 Instrumentation. 66 4.22 Experimental protocol. 70 4.23 Measurements. 71 4.24 Data Analysis. 77 4.3 Results. 78 4.4 Discussion. 88 4.5 Summary. 92 iv CHAPTER V Exclusion of activated leukocytes from the myocardial circulation during sepsis ameliorates myocardial structural changes and contractile failure. 94 5.1 Hypothesis. 94 5.2 Experimental design. 94 5.21 Instrumentation. 94 . 5.22 Experimental protocol. 96 5.23 Measurements. 98 5.24 Data Analysis. 105 5.3 Results. 105 5.4 Discussion. . 1 1 2 5.5 Summary. 118 CHAPTER VI Summary of major conclusions and recommendations for further study. 120 6.1 Summary of major-conclusions. 120 6.2 Recommendations for further study. 121 Bibliography. 124 Chapter I . 1 2 4 Chapter II 129 Chapter III 133 Chapter IV 136 Chapter V 139 Appendix I Symbols and abbreviations used in the text. 144 v LIST OF TABLES CHAPTER I Table 1.1. R.C. Bone et al, Sepsis syndrome: A valid clinical entity. 1 CHAPTER II Table 2.1. Calculation of myocardial capillary transit time of leukocytes. 18 Table 2.2. Additional measured variables. 33 CHAPTER III Table 3.1. Hemodynamic and physiologic parameters for control and endotoxin support rabbits. 45 CHAPTER IV Table 4.1. Coronary leukocyte retention and coronary blood flow. 84 Table 4.2. Total leukocytes per high power field (HPF). . 8 7 vi LIST OF FIGURES CHAPTER II Figure 2.1 Photomicrographs demonstrating morphometric technique. 17 Figure 2.2 Leukocyte counts as a percentage of baseline. 23 Figure 2.31 Transmyocardial leukocyte gradients (control group). 24 Figure 2.32 Transmyocardial leukocyte gradients (endotoxin group). 25 Figure 2.4 Number of leukocytes retained in coronary circulation. 26 Figure 2.51 Myocardial arterial leukocyte clearance (control group). 27 Figure 2.52 Myocardial arterial leukocyte clearance (endotoxin group). 28 Figure 2.6 Myocardial capillary transit times of leukocytes. 30 Figure 2.7 Myocardial capillary transit times of red blood cells. 31 Figure 2.8 Myocardial oxygen extraction ratio. 32 CHAPTER III Figure 3.1 Isolated-supported rabbit heart model 47 Figure 3.2 Plot of left ventricular pressure versus left ventricular volume. 50 Figure 3.3 Volume fraction of myocardial capillaries occupied by leukocytes. 54 Figure 3.4 Volume fraction of abnormal cardiac myocytes. 55 Figure 3.5 Volume fraction of myocardium occupied by interstitium. 56 Figure 3.6 Change in E m a x versus time. 57 CHAPTER IV vii Figure 4.1 Figure 4.2 Figure 4.3 Figure 4.4 Figure 4.5 Figure 4.6 Figure 4.7 Figure 4.8 CHAPTER V Figure 5.1 Figure 5.2 Figure 5.3 Figure. 5.4 Figure 5.5 Figure 5.61 Figure 5.62 Figure 5.7 CHAPTER VI Figure 6.1 Transit time perfusion apparatus. 68 Total leukocyte input distribution (l(t)) and output distribution (O(t)) versus time. . 74 Transit time distribution function (T(t)). . 7 5 CD18 relative mean fluorescence intensity. 79 Coronary transit time of total leukocytes. 81 Coronary transit time of PMN and lymphocytes. 82 Percentage retention of total leukocytes. 83 Percentage retention of PMN and lymphocytes. 86 Dual heart perfusion apparatus. 97 Ventricular pressure plotted versus ventricular volume. 99 Endotoxin concentrations in blood samples pre and post-filter. 102 Effects of leukocyte depletion on myocardial contractility. 106 Effects of endotoxemia and leukocyte filtration on circulating neutrophil count. 107 Myocardial morphometric analysis: Myocardial neutrophil retention. 108 Myocardial morphometric analysis: Fractional area of interstitial edema and myocardial myocyte damage. 109 Expression of L-selectin on neutrophils. 110 Electronmicrographs of control and septic cardiac endothelial cells. 122 viii Control myocardium. Myocardial endothelial cell swelling and cytoskeletal changes. Myocardial endothelial cell/leukocyte interaction: Membrance changes. Myocardial endothelial cell/leukocyte interaction: "Lacunar zone". ix ACKNOWLEDGMENTS I would like to thank Dr. James C. Hogg for inviting me along in the first place and for encouraging and supporting these studies and Dr. Keith R. Walley, my supervisor and teacher, who understood the value of freedom. I would like to thank Mr. Dean English, Ms. Lynne Carter and Ms. Betty Poon for their patience, technical assistance and endless sense of humor. I would like to thank Dr. Barry R. Wiggs for being there when it counted. I would like to thank my colleague Dr. Darryl Knight for his friendship and superlative example. This is for Janet, Kathryn, Bud, Karen and Jennifer who share my journey x CHAPTER I INTRODUCTION 1.1 Sepsis and SIRS The word sepsis is derived from the Greek word for putrefaction (51) or decay referring to either the putrid smell of associated pus or from the wasting syndrome frequently associated with chronic infections. With the advent of modern microbiologic methods it became possible to ascribe the condition of sepsis to the presence of an organism or it's products within a host. Physicians soon became aware, however, that numerous conditions unrelated to infectious disease may manifest symptomatology virtually identical to that of an invading organism. Despite this awareness the term sepsis is frequently misused and this may be associated with inappropriate treatment of the patient. In order to clarify this problem Bone (5) defined the "sepsis syndrome" in 1989 as: Rectal temperature > 101 °F or < 96°F Tachycardia (>90 beat/min) Tachypnea (> 20 breath/min while spontaneously breathing) At least one of the following manifestations of inadequate organ function/perfusion: 1: Alteration in mental status 2. Hypoxemia (Pao2 < 72 torr breathing room air) (overt pulmonary disease not direct cause of hypoxemia) 3. Elevated plasma lactate level 4. Oliguria (urine output < 30 ml or 0.5 ml/kg for at least 1 h) (Table 1.1. from R.C. Bone etal. Sepsis syndrome: A valid clinical entity (5)) 1 When the above criteria are met and there is "clinical evidence of infection" the sepsis syndrome is said to be present. This effectively modifies the definition of sepsis to a condition arising from the presence of an organism or it's products within a host together with the response of the host. This was perhaps the first attempt to combine the numerous clinical manifestations of sepsis into a single diagnostic entity and.to shift the focus of the definition from the activity of the infecting organism to the response of the host. This is an essential distinction and has been instrumental in standardizing multicenter trials of sepsis management (4). However, the operational nature of this definition has lead to several difficulties. For example, variables such as fever, tachycardia and tachypnea may be related to a number of pathophysiologic causes unrelated to sepsis or may be absent when sepsis is present. Furthermore, it is difficult to quantify "altered mental status". Thus this definition does little to enhance understanding of the underlying physiologic alterations manifest during sepsis or endotoxemic states. In, order to address this issue the term "systemic inflammatory response syndrome" or SIRS was introduced. While SIRS is also defined by operational criteria it is a crucial concept linking disparate areas of basic science such as immunology, cardiopulmonary mechanics and autonomic neurobiology with clinical physiology. It is the concept of a "SIRS state" that permits progress to be made into the basic biology of host response to a noxious stimulus. Thus sepsis may be redefined as SIRS occurring secondarily to the presence of bacteria or their products within a host and furthermore an identical SIRS state may arise secondarily to non-infectious causes such as trauma. v; 2 Implicit to SIRS are the concepts of a SIRS "effector" which are the factors mediating host cellular damage or dysfunction and a SIRS "engine" which are the factors causing and perpetuating host inflammatory activity (for example, invasive infection in the case of sepsis). SIRS effectors may be divided two primary divisions, the humoral (1, 7, 38, 55) and the cellular (18, 45). Examples include the adult respiratory distress syndrome (ARDS) (24) and post-ischemic reperfusion injury (10, 30, 32, 35, 44, 43, 58). In ARDS normally quiescent leukocytes become activated and adhere to and damage the endothelium of pulmonary capillaries. Reperfusion of the heart following an ischemic episode leads to the retention of activated leukocytes in post-capillary venules of the heart. Leukocyte retention may mediate damage to potentially viable cardiac myocytes and extend infarct size. SIRS engines include factors such as tumor necrosis factor-alpha (TNF-a) a cytokine induced early in the SIRS state which may mediate activation of host effector response. Proximal cytokine molecules such as TNF-a may induce the formation of other cytokine and non-cytokine molecules which may independently serve as engines leading to cascading tiers of host response. Because of these complex engine-effector relationships operational definitions of SIRS are not capable of distinguishing SIRS states related to infection from other causes. This fact has prompted extensive research into the underlying pathophysiology of SIRS which at the present time is largely undefined. The SIRS state also offers insight into the lack of efficacy of a number of therapeutic trials aimed at sepsis and septic shock (3). SIRS comprises a number of neural and cytohumoral responses of a host to various insults. These responses may be interpreted as adaptive or pathologic based on the host outcome and this distinction is often difficult. Furthermore, while these responses may be synergistic and interrelated they may also represent separate pathways of host response. Thus, interruption of any single pathway may not modify the outcome. For example, the use of anti-TNF-a antibodies do not improve clinical survival in sepsis (37). Therefore mediators of host outcome: variables such as peripheral vascular dysfunction (14, 60, 61) during sepsis will likely prove to be multifactorial rather than reducing to a common underlying pathogenetic factor. A central tenet of this thesis is that independent SIRS pathways may independently lead to an observed pathologic state. This thesis is concerned specifically with one such pathway, specifically the behavior of activated leukocytes in the microcirculation of the heart during sepsis, with the outcome variable being decreased myocardial contractile function; a condition which contributes to the pathogenesis of septic shock and thus to increased mortality during sepsis (4). 1.2 Myocardial dysfunction in sepsis Dysfunction of the cardiovascular system occurs in 40 % of patients with sepsis (4, 21). The mortality rate of sepsis increases from 20 to 30 % in patients without cardiovascular dysfunction (4, 22) to approximately 40 to 70 % in septic shock (4, 48). The cause of this cardiovascular dysfunction is incompletely understood but likely occurs secondary to the effects of components of SIRS on the heart and peripheral circulation. 4 A number of investigators have demonstrated that myocardial dysfunction occurs in animal models of septic shock (36). Parker and colleagues (41) have shown that human survivors of septic shock initially had a markedly decreased mean ejection fraction of 32% despite an increased cardiac output. With resolution of sepsis, left ventricular ejection fraction returned to normal over the course of approximately 10 days (40, 41). Decreased ejection fraction may be accounted for by decreased systolic contractility because afterload is decreased in sepsis (41) and therefore, if systolic contractility were unchanged, ejection fraction would increase. Indeed decreased systolic contractility in sepsis has been confirmed in both human (53) and animal (36, 57) studies of sepsis and has been replicated in a model identical to the porcine model of acute endotoxemia (19, 20) employed in Chapter II below. Decreased systolic function was also observed in humans given an infusion of endotoxin (49). Infusion of endotoxin is a common method employed to experimentally activate SIRS pathways (27, 39) and reliably causes a syndrome of myocardial dysfunction similar to that observed in other models of experimental sepsis such as infected clot implantation (36) and infusion of whole bacteria (28). Thus endotoxin infusion is the method employed in this thesis for modeling myocardial dysfunction in sepsis. However, it is recognized that sepsis induced by infusion of endotoxin versus whole bacterial models may differ (25) and thus individual experimental systems must be evaluated on an individual basis. The mechanisms underlying septic myocardial dysfunction have not been fully elucidated and this is the major subject of this thesis. The vast 5 majority of existing clinical and basic research studies indicate that components of the SIRS state are etiologic in septic myocardial dysfunction. Both humoral and cellular pathways may be involved in septic myocardial dysfunction and the evidence for this is discussed below. 1.3 Humoral mediators of sepsis There is an extensive literature regarding the effects of certain humoral mediators of sepsis on the function of the heart during sepsis (1, 7, 38, 55). One compelling reason for accepting humoral agents as etiologic in septic myocardial dysfunction is the concept of a myocardial depressant substance proposed by Parrillo and colleagues (40) is based on the finding that plasma from septic patients produces contractile dysfunction of isolated cardiac myocytes in-vitro. However, there is evidence to suggest that humoral agents or a myocardial depressant substance do not account entirely for decreased myocardial contractility during sepsis. The exact identification of a circulating factor remains elusive despite 20 years of searching. Confusion exists about whether this factor has a low (500 to 1,000 D [17, 31]) or high molecular weight (10 to 30 kD [16, 40]), whether it is lipid soluble (6) or water soluble and dialyzable (16,17), and the timing of it's expression in plasma (6, 31). A variety of cytokines have been implicated as the depressant factor, including TNF-ct, interleukin-1 (IL-1), IL-2 and IL-6, all of which have been shown to decrease myocardial contractile function in vitro by inducing the production of nitric oxide (NO) (13). Yet others have found that NO does not account for decreased ventricular contractility (8). Furthermore, plasma concentrations of TNF-a (57) and other cytokines do not correlate with the delayed 10-d time course of myocardial dysfunction of sepsis (41), and in vitro antagonism of cytokines has 6 not fully prevented the decrease in myocardial contractility associated with sepsis (19). Furthermore, the application of plasma from endotoxemic animals to the crystalloid-perfused intact isolated heart does not produce decreased myocardial contractility (preliminary results not discussed further in this thesis). Therefore, myocardial contractile dysfunction observed during septic shock may not be completely explained by a simple circulating myocardial depressant factor. 1.4 Cellular mediators of sepsis During the inflammatory cascade of sepsis (SIRS) circulating leukocytes change from a quiescent resting state to an "activated" state in which the leukocyte displays altered rheologic (15, 23, 50, 59), adhesive (11, 34, 52, 54) and functional (9, 42) characteristics which enable leukocytes to mediate damage to either foreign or host tissue cells. For example, leukocytes recruited and activated by cytokines and other inflammatory mediators contribute to the development of ARDS (24), multiple organ failure during septic shock (33, 52), and cardiac ischemia-reperfusion injury (32, 43). There are a number of putative mechanisms whereby activated leukocytes may mediate decreased myocardial contractile function in sepsis (18, 45). Conceivably release of cytokines from macrophages within the myocardium could result in much higher cytokine concentrations within the myocardium with resultant deleterious effects on myocardial contractile function in the absence of high peripheral cytokine levels. Stimulation of the neutrophil oxidative burst dependent on CD11/CD18-mediated adhesion to endothelial cells may release significant quantities of toxic oxygen free radicals (12). 7 Leukocytes may damage endothelial cells or plug capillaries of the heart and other organs altering local oxygen delivery and extraction relationships (26, 29, 46, 56) causing micro-regional ischemia in the absence of global ischemia. However, little is currently known regarding the flow characteristics of activated leukocytes in the coronary microcirculation or whether activated leukocytes contribute to the reduced myocardial contractile function observed during sepsis. 1.5 Major Hypotheses During endotoxemia circulating leukocytes become activated. The transit time of activated leukocytes through the microcirculation of the heart is increased. Increased leukocyte transit time leads to the retention of leukocytes within the microcirculation of the heart. Retained activated leukocytes cause damage to cardiac myocytes. Cardiac myocyte damage in part mediates reduced myocardial contractility in endotoxemic states such as sepsis. Therefore, exclusion of activated leukocytes from the microcirculation of the heart should prevent cardiac myocyte damage and reduced myocardial contractility following infusion of endotoxin. 8 CHAPTER II LEUKOCYTE TRANSIT TIME AND RETENTION IN THE CORONARY MICROCIRCULATION DURING SEPSIS 2.1 Hypothesis. Activated neutrophils lodge in pulmonary capillaries following endotoxin infusion (18). In the lungs, retained neutrophils contribute substantially to tissue damage (6, 18). Conceivably, leukocytes retained in the coronary circulation could contribute to decreased myocardial contractility and microvascular dysfunction, as they do in the lungs. Whether sufficient numbers of leukocytes are delayed for a sufficient time in the coronary microcirculation to contribute to tissue damage is not known. Accordingly, the hypotheses were tested that leukocyte transit is slowed and large numbers of leukocytes are retained in the coronary microcirculation, in an endotoxemic model of hyperdynamic sepsis in pigs. Myocardial leukocyte retention was measured in two ways. First, the transmyocardial leukocyte gradient was calculated as the difference between leukocytes flowing into and out of the coronary circulation for the duration of the experiment. Second, the number and location of leukocytes within the myocardium was quantitated using a morphometric technique adapted from Cruz-Orive and Weibel (7). The myocardial capillary transit time of leukocytes was then calculated using a combination of morphometric measurements and blood volume and flow measurements using radiolabeled markers (21). 2.2 Experimental design. 9 2.21 Instrumentation. Fourteen pigs of either sex weighing 30 ± 5 kg were anesthetized using ketamine (20 mg/kg IM) and thiopental sodium (10 mg/kg IV). Anesthesia was maintained using isoflurane (0.5%) and a constant ketamine infusion (0.1 mg/kg IV). Depth of anesthesia was tested hourly by observation of heart rate changes, blood pressure response, and lacrimation in response to painful stimuli. Paralysis was produced with pancuronium bromide (0.1 mg/kg IV) and maintained by a constant infusion (0.1 mg/kg/h IV). Tracheotomy was performed and the animals were mechanically ventilated with a tidal volume of 12 ml/kg and a respiratory rate adjusted to maintain an arterial CO2 tension (Pac02) °f 35-45 mmHg. A positive end-expiratory pressure of 5 cmH20 was applied to maintain end-expiratory lung volume. The core temperature of the pig was maintained at a normal value of approximately 38.5°C using a heating blanket. An external jugular venous catheter was inserted for infusion of normal saline and drugs. A second external jugular venous line was inserted for infusion of endotoxin or vehicle. An internal carotid artery catheter was inserted for measurements of arterial blood pressure. A second internal carotid arterial catheter (vol 0.8 ml) was inserted and advanced to the proximal aorta to sample blood for measurement of total arterial leukocyte count (S880 Automated Hematology Analyzer, Coulter Counter, Coulter Electronics, Hialeah, FL), pH, arterial oxygen tension, Pacc>2> bicarbonate (ABL 30, Radiometer, Copenhagen), and arterial oxygen saturation (Cooximeter IL 482, Instrumentation Laboratories, Lexington, MA). 10 1 A left lateral thoracotomy was performed at the level of the fourth intercostal space. The pericardium was widely opened. A catheter (vol 0.8 ml) was placed via the right internal jugular vein into the coronary sinus and advanced so that the tip lay at the inferior margin of the left atrial appendage. This catheter was used to sample blood for measurement of total leukocyte count and oxygen saturation in the venous outflow of the heart. The left hemiazygous vein was ligated external to the pericardium to prevent contamination of the coronary sinus blood sample by systemic venous blood. The left anterior descending coronary artery (LAD) was dissected free of the epicardium just distal to the level of the bifurcation of the left mainstem coronary artery. An ultrasonic flow probe was then placed around the LAD. LAD flow was measured continuously throughout the experiment using an ultrasonic blood flow transducer (T201, Transonics Systems, Ithaca, NY). A catheter for delivery of radiolabeled microspheres and radiolabeled red blood cells was inserted into the left atrial appendage and sutured into place. 2.22 Experimental protocol. After a 1-h equilibration period, control blood samples were obtained at 1-min intervals for 10 min. During a single sample period, simultaneous 2.5-ml blood samples were taken from the aortic root and coronary sinus to measure transmyocardial leukocyte gradient and oxygen extraction ratio. Samples were drawn at the same rate to negate the effects of variations in the respiratory phase. Before each blood sample, a volume equal to twice the known dead space of both sampling catheters was removed to prevent contamination of the sample with stagnant blood. This aliquot of blood was then reinjected into the pig to preserve total hemoglobin. Approximately 2 ml of each blood sample 11 were transferred to a sterile blood collection tube containing 0.1 ml sodium EDTA as an anticoagulant and thoroughly mixed. Total leukocyte count was measured immediately in all samples to avoid cell clumping. Approximately 0.5 ml of the sample was collected in a heparinized blood gas syringe and immediately analyzed for P02 and arterial oxygen content. After the 10 control blood samples were obtained pigs immediately received either a control infusion of normal saline (n = 7) or an infusion of 50 ug/kg endotoxin (Escherichia coli 0111: B4, Sigma, St. Louis, MO) dissolved in saline (n = 7) over 20 min via the external jugular venous catheter. This dose of endotoxin has previously been demonstrated in an identical porcine model to reproduce the systemic manifestations of the systemic inflammatory response syndrome and to cause reduced myocardial contractile function following infusion of endotoxin (19). Blood sampling was continued at 1-min intervals throughout the 30-min infusion of endotoxin or saline. After the infusion, sampling was continued at 3-min intervals for an additional 45 min. Paired arterial and coronary venous samples were used to calculate the number of leukocytes retained in the coronary circulation during the experiment but were not used in the calculation of myocardial capillary transit time of leukocytes at the end of the experiment. At the end of the experiment (75 min after the start of the endotoxin or QQ vehicle infusion), technetium-99 ( Tc)-labeled red blood cells were injected into the left atrium for measurement of left ventricular myocardial blood volume. After a 10-min equilibration period, strontium-85 ( Sr)-labeled microspheres were injected into the left atrium for measurement of total left ventricular blood flow. The heart was then removed for morphometric analysis <1 min after 12 microsphere injection. Then the volume, flow and morphometric measurements, which were used for the calculation of the myocardial capillary leukocyte transit time, were approximately simultaneously measured at 85 min after the start of the endotoxin or saline infusion. 2.23 Measurements. 2.231 Measurement of left ventricular blood flow. During the experiment it was assumed that LAD flow was a relatively constant fraction of total left ventricular blood flow. Thus total left ventricular blood flow was estimated as ultrasonic probe LAD flow multiplied by the rate of radiolabeled microsphere-measured total left ventricular blood flow divided by the simultaneous ultrasonic probe LAD flow obtained during flow calibration, done at the end of the experiment. During flow calibration, ultrasonic probe LAD flow was recorded, and simultaneous total left ventricular blood flow was measured using Sr-radiolabeled 15-um microspheres. Microspheres were fully mixed in 10 ml normal saline before injection. Microspheres were then rapidly injected over 1 s into the left atrium. A reference blood sample was withdrawn from the aorta at a constant rate of 10 ml/min for 2 min. At the end of the experiment the excised and fixed left ventricle was divided into multiple small samples, and the entire left ventricular myocardium and the reference blood sample were counted for radioactivity. Ventricular blood flow was then calculated as total ventricular counts divided by total reference blood sample counts multiplied by 10 ml/min. These are measurements of flow at the end of the experiment. 13 2.232 Measurement of myocardial blood volume. At the end of the experiment, 16 ml of whole arterial blood was removed QQ and labeled with Tc. The arterial blood was mixed with 4 ml acid-citrate-q q dextrose and radiolabeled with 200 uCi Tc using a Glucoscan kit (NEN, NRP-180, Kirkland, Quebec, Canada). Ten ml of the labeled red blood cells were then injected into the left atrium over 10 s and allowed to circulate for 10 min. A reference peripheral blood sample was then withdrawn. After the heart was fixed in Formalin (section 2.235, Myocardial leukocyte content predicted from arterial leukocyte concentrations) myocardial blood volume was determined as q q Q Q total left ventricular Tc counts divided by total reference blood Tc counts multiplied by the volume of the reference blood sample. Separation of activity p c q q of the Sr microspheres from the Tc-labeled red blood cells below was performed using a computer-based separation of gamma counts. 2.233 Calculation of the number of leukocytes retained in the coronary circulation. During the experiment the transmyocardial leukocyte gradient was computed for each set of blood samples as the difference between simultaneous aortic root leukocyte count and coronary sinus leukocyte count multiplied by the total left ventricular blood flow. The number of leukocytes retained in the coronary circulation was calculated as the sum of the individual transmyocardial leukocyte gradients over each sampling interval throughout the experiment. 2.234 Calculation of myocardial leukocyte clearance. 14 The instantaneous clearance of leukocytes from arterial blood by the myocardium was calculated as the arterial leukocyte concentration minus the coronary sinus leukocyte concentration divided by the arterial leukocyte concentration. Clearance was multiplied by 100 to yield percent myocardial leukocyte clearance. 2.235 Myocardial leukocyte content predicted from arterial leukocyte concentrations. At the end of the experiment, the great vessels and pulmonary veins were clamped to prevent efflux of blood and the heart was removed for fixation in 10% buffered Formalin. A large amount of fixative was injected into the left ventricle by stab puncture, and an equal quantity of left ventricle blood was simultaneously removed to prevent distention of the ventricle. The whole heart was then immersed in a large volume of the same fixative for 8 h. Four hearts from each group were selected randomly for further processing. Each of these hearts was sliced in a transaxial orientation into adjacent 1 -cm tissue discs, and the atria, right ventricle, and epicardial fat were removed from each disc. These tissue discs were then allowed to continue fixing in fresh 10% buffered Formalin for an additional 24 h. After fixation, 10 randomly selected left ventricular tissue samples were removed from each disc, dehydrated, and embedded in glycol methacrylate. The resulting tissue blocks were sectioned at 2 um. Serial sections were stained with methenamine silver (myocardial interstitium and capillaries) and toluidine blue (myocytes and leukocytes) and coded before analysis. 15 Quantitative histological assessment of the left ventricular myocardium was performed by an investigator blinded as to treatment group using a computer-assisted point-counting technique (1) based on that of Cruz-Orive and Weibel (7). Counting was performed at three levels of magnification. Level 1 (low power, x 40) was used to determine the fraction of tissue occupied by large (diameter > 100 u,m) and medium-sized (10 um) blood vessels and the fraction that was "myocardium." Level 2 (medium power, x 100) was used to determine the fraction of myocardium from level 1 that was capillaries (diameter > 10 urn), interstitium, and myocytes. Level 3 (high power x 400) was used to determine the fraction of the total capillary volume that was leukocytes [Fig. 2.1]. Ten random samples of each heart were assessed from each animal, and ten random fields of each sample were counted by a single experienced investigator using a 90-point field grid at low and medium power. Fifteen random fields per sample were counted with a 90-point grid at high power. In all cases this produced a relative standard error of volume fraction of <10%. The fraction of capillary volume occupied by leukocytes divided by the known volume of a spherical leukocyte (taken as 1.74 x 10~1^ ml) (10) yielded the observed number of leukocytes per unit capillary volume [G in Table 2.1]. The calculated morphometric capillary blood volume multiplied by the measured peripheral arterial leukocyte concentration just before harvesting the heart yields the predicted number of leukocytes per unit of capillary volume, if there was no difference between arterial leukocyte concentration and myocardial capillary leukocyte concentration. This theoretical leukocyte concentration in the myocardial capillaries is defined as the "number of leukocytes predicted from the arterial leukocyte concentration." The observed number of leukocytes per milliliter of capillary blood volume divided by the 16 FIGURE 2.1 Photomicrographs demonstrating morphometric technique Fig . 2.1. Photomicrographs taken from a single animal from endotoxin group that demonstrate morphometr ic technique of Cruz-Or ive and Weibel (7), in which object p h a s e of p reced ing lower power b e c o m e s reference p h a s e for object being counted at that level. A: arrow, large blood vesse l ; sca le bar, 100 um. B: arrowheads, capi l lar ies; sca le bar, 40 um. C : arrowheads, capi l lar ies; arrow, polymorphonuclear leukocyte visible within a capi l lary (endotoxin group); sca le bar, 10 Lim. 17 TABLE 2.1 Calculation of myocardial capillary transit time of leukocytes. Quantity Description Endotoxin Control , A Left ventricular blood flow 8.1 +2.1 6.3 ±2.0 B Myocardial blood volume 3.5 ±1.0 3.4 ± 1.0 C Fraction of myocardial blood volume that is capillaries 0.78 + 0.06 0.80 ± 0.04 D = (B/A) x C Myocardial capillary transit time of red blood cells 0.36 ±0.13 0.45 ±0.13 E Fraction of capillary volume that is leukocytes 0.053 ± 0.009 0.036 ±0.013 F = B x C x E Volume of leukocytes in myocardial capillaries 0.148 ±0.058 0.095 ±0.040 G = E/1.74x10" 1 0* Number of leukocytes per ml of myocardial capillary blood 307 ± 51 x 106 206±78x 106 H Arterial leukocyte concentration 3.1 ±1.5x 106 18.3 ± 6.4 x 1 0 6 l = G/H Observed concentration of leukocytes in myocardial capillaries/number predicted from arterial leukocyte concentration 111 ±37 11 ±2 J = A x H x 1 . 7 4 x 1 0 - 1 0 * Volume flow of leukocytes through myocardial capillaries 4.1 ±1.3x 10"3 21.6±13.3x10" 3 K=F/J Myocardial capillary transit time of leukocytes 39.1 ± 20.6 5.0 ±1.4 Table 2.1. Values are means ± SD; n = 4 pigs for both endotoxin and control groups. Equations do not give exact results because the reported numbers are averages. A, microspheres (ml/s/100 g); B, labeled red blood cells (ml/100 g); C, morphometry; D, in s; E, morphometry; F, in ml/100 g; G, in number/ml; H, Coulter counter (number/ml); I, dimensionless; J, in ml/s; and K, in s. * Leukocyte volume was taken as 1.74 x 10 " 1 0 ml (10). 18 number predicted from arterial leukocyte concentrations is shown as I in Table 2.1. Observed-to-predicted ratios >1 indicate increased concentration of leukocytes in the myocardial capillaries relative to arterial blood. 2.236 Calculation of myocardial capillary transit time of leukocytes. The myocardial capillary transit time of leukocytes at the end of the experiment was calculated by dividing the volume of leukocytes in myocardial capillaries by the volume flow of leukocytes through this capillary bed (21) [K in Table 2.1]. The volume of leukocytes in myocardial capillaries was obtained by multiplying the myocardial capillary blood volume by the fraction of capillary volume that is leukocytes [F in 2.1]. The volume flow of leukocytes through myocardial capillaries was calculated as total myocardial blood flow from radiolabeled microsphere measurements multiplied by the arterial leukocyte count multiplied by leukocyte volume [J in Table 2.1]. 2.237 Calculation of myocardial capillary transit time of red blood cells. Capillary transit time of red blood cells at the end of the experiment was calculated as the left ventricular myocardial blood volume measured using radiolabeled red blood cells (ml/100 g wet wt) divided by the total left ventricular blood flow measured using radiolabeled microspheres (ml/s/100 g wet wt) multiplied by that fraction of total coronary blood volume contained in capillaries (1), 2.238 Measurement of leukocyte activation. 19 A number of investigators have shown that the infusion of endotoxin in vivo leads to' the "activation" of circulating leukocytes (2, 3, 36). To determine if leukocyte activation occurred in this experiment, two additional pigs were studied that had undergone anesthesia, instrumentation, and blood sampling protocols identical to those of the main experimental groups. The upregulation of CD18, a marker of cell activation (36), on polymorphonuclear neutrophils was quantified before and 75 min after the infusion of normal saline (n = 1) or 50 Lig/kg endotoxin. The expression of CD18 on neutrophils was determined by immunofluorescent flow cytometric analysis. Three milliliters of whole blood were aspirated from the internal carotid artery immediately before infusion of saline or endotoxin and at 75 min after infusion. The blood samples were anticoagulated with EDTA. Neutrophils were prepared for analysis using a commercially available kit (Coulter Clone, Coulter Electronics). Briefly, 100 uJ of EDTA blood was incubated with 200 ul phosphate-buffered saline and 0.4 ug mouse anti-CD18 antibody for 10 min at room temperature. Specimens were washed with phosphate-buffered saline, the supernatant was removed, and the sample was incubated for 10 min in fluorescein isothiocyanate (FITC)-conjugated goat anti-mouse immunoglobulin G (IgG) (Sigma). For each blood sample a negative control was prepared using nonimmune mouse FITC-conjugated lgG2a (Becton Dickinson). The cells were then washed twice using phosphate-buffered saline, and the red blood cells were lysed (Immunolyse, Coulter Clone, Coulter Electronics). The remaining leukocytes were fixed with 1% paraformaldehyde and stored at 4°C. Flow cytometry was performed on the specimens within 24 h of preparation using a Profile EPIC II flow cytometer (Coulter Electronics). Analysis gates for the neutrophils were established using the distinctive forward and side-scatter profiles. A total of 3,000 gated cells 20 were evaluated per specimen, and mean fluorescence intensity for CD18 was measured. The infusion of endotoxin led to a 61% increase in mean fluorescence intensity for CD 18 over the preinfusion baseline, whereas there was no increase after saline infusion. This suggests that endotoxin infusion in this experiment is activating the majority of the circulating leukocytes. 2.239 Calculation of myocardial oxygen extraction ratio. Blood oxygen content was calculated as C02 = (S02 x Hb x 1.36) + (P02 x 0.03), where S02 is the percent saturation of oxygen, Hb is blood hemoglobin content (g/l), 1.36 is the oxygen-carrying capacity of hemoglobin (ml 02/g hemoglobin), P02 is the blood partial pressure of oxygen (mmHg), and 0.03 is the coefficient of oxygen solubility (ml/l). Myocardial oxygen extraction ratio was calculated as arterial minus coronary venous oxygen content, divided by arterial oxygen content. 2.24 Data analysis. An unpaired Student's Mest was used to test for differences between the control group and the endotoxin group with respect to 1) total number of leukocytes retained in the coronary circulation, 2) the ratio of observed to predicted myocardial leukocyte content, and 3) myocardial capillary transit time of leukocytes and red blood cells. A two-way repeated measure analysis of variance was used to test for differences in myocardial oxygen extraction ratio 21 between the control and endotoxin groups. P < 0.05 was chosen as statistically significant. Data are expressed as means + SD throughout. 2.3 Results. During infusion of endotoxin, total peripheral leukocyte counts in the aortic root and coronary sinus decreased in the endotoxin group [Fig. 2.2]. During the 10-min control period, transmyocardial leukocyte gradients [Fig. 2.31, 2.32] and numbers of leukocytes retained in the coronary circulation [Fig. 2.4] were not significantly different from zero or between endotoxin (-0.26 ±0.12 x 10 9 leukocytes) and control (-0.22 + 0.08 x 109 leukocytes) groups. During the total experimental observation period, more leukocytes were retained in the endotoxin group (2.1 ± 0.8 x 109 leukocytes) than in the control group (0.05 ± 0.06 x 109 leukocytes) (p < 0.05). The overall mean retention of leukocytes in the control group was not different from zero. Myocardial clearance of leukocytes [Fig. 2.51, 2.52] was consistently positive for the first hour after the start of the endotoxin infusion but was not different from zero in the control group. The ratio of observed myocardial capillary leukocytes to the number of leukocytes predicted from the arterial leukocyte concentration was 111 ± 37 in the endotoxin group. This was significantly different from the control value of 11 ± 2 (P < 0.002). In the endotoxin group, the vast majority of leukocytes were retained in capillaries [Fig. 2.1]. Leukocytes were only rarely observed within vessels larger than capillaries. No extravascular leukocytes were observed. Myocardial capillary transit times of leukocytes were significantly 22 FIGURE 2.2 Leukocyte counts as a percentage of baseline Fig. 2.2. Simultaneous leukocyte counts as a percentage of baseline from aortic root (black squares) and coronary sinus (open circles) of control and endotoxin groups. Values are means; n = 7 animals in each group. 23 FIGURE 2.31 Transmyocardial leukocyte gradient (control group) Fig. 2.31. Transmyocardial leukocyte gradients in control group. Bars represent means of all animals at each time; n = 7 animals in each group. During saline infusion transmyocardial leukocyte gradients fluctuate evenly from positive to negative indicating no retention of leukocytes within coronary circulation. WBC, white blood cells. 24 FIGURE 2.32 Transmyocardial leukocyte gradient (endotoxin group) UJ Q < DC O _ ^ <§> S i _lc&-<2 Q x < CD ° 5 O ~ >-CO z < DC 300 -, 200 100 H 0 -100 --200 --300 --400 Li lUiu. ENDOTOXIN INFUSION •HIH.I..I. n -10 0 10 20 30 75 TIME (MINUTES) Fig. 2.32. Transmyocardial leukocyte gradients in endotoxin group. Bars represent means of all animals at each time; n = 7 animals in each group. During endotoxin infusion transmyocardial leukocyte gradient is consistently positive, indicating retention of leukocytes within coronary circulation. WBC, white blood cells. 25 FIGURE 2.4 Number of leukocytes retained in coronary circulation 3 i o i— 5 O CC o >-cc < z 0-3 o ° o ~ LU<JT x o z x < h-LU CC CO 111 o o LU 2 H 0 -1 J • ENDOTOXIN • CONTROL -10-0 0-30 30-75 -10-0 0-30 30-75 TIME (MINUTES) Fig. 2.4. Number of leukocytes retained in coronary circulation for time intervals -10 - 0, 0 - 30, and 30 - 75 min. Bars are means ± SD; n = 7 animals in each group. There is no significant difference between endotoxin and control groups before infusion of endotoxin or vehicle, respectively. By 30 and 75 min after start of infusion, a large number of leukocytes are retained in coronary circulation in endotoxin group (* p < 0.05) compared with control group. Total number of leukocytes retained in control group was not significantly different from zero at any time. 26 FIGURE 2.51 Myocardial arterial leukocyte clearance (control group) 10 n LU o z < cc < LU _l o LU o o ZD LU < CC LU I-cc < _l < Q CC < O O >-0 •10 TIME (MINUTES) Fig. 2.51. Myocardial arterial leukocyte clearance in control group. Values are means; n = 7 animals in each group. 27 FIGURE 2.52 Myocardial arterial leukocyte clearance (endotoxin group) 10 n LU o z < DC < LU i o LU I— > o o ZD UJ < DC LU H rr < < Q DC < O O > TIME (MINUTES) Fig. 2.52. Myocardial arterial leukocyte clearance in endotoxin group. Values are means; n = 7 animals in each group. 28 prolonged at 39.1 ± 20.6 s (range 28.4 - 67.9 s) in the endotoxin group compared with 5.0 + 1.4 s (range 3.4 - 6.7 s) in the control group (p < 0.008) [Fig. 2.6]. Myocardial capillary transit time of red blood cells was 0.36 ± 0.13 s (range 0.23 - 0.52 s) in the endotoxin group, which was not significantly different from the control group value of 0.45 ± 0.13 s (range 0.26 - 0.57 s) [Fig. 2.7]. These observations indicate that endotoxin infusion has no significant effect on the time taken by red blood cells to pass through the left ventricular microvasculature but selectively slowed transit of leukocytes. Oxygen saturation of hemoglobin in the aortic root of both the endotoxin and control groups was maintained throughout the experiment at 99 - 100% with supplementary oxygen. There was no significant change in the coronary sinus oxygen saturation in the control group from the preinfusion value of 29.1 ± 4.8% at any time during the experiment. Thus the myocardial oxygen extraction ratio in the control group showed no significant change from the preinfusion value of 71 ± 8% at any time during the experiment. Coronary sinus hemoglobin oxygen saturation increased in the endotoxin group from the preinfusion value of 30.6 ± 5.0 to 63.8 ± 8.9% at 75 min following the start of the infusion (p < 0.002). The myocardial oxygen extraction ratio decreased significantly from a preinfusion value of 69.6 ± 4.9 to 37.6 ± 8.6% (< 0.05) at 75 min after the start of infusion in the endotoxin group [Fig. 2.8]. There was no significant change in the mean arterial pressure, heart rate, arterial hemoglobin, or core body temperature from the preinfusion values of either the control or endotoxin groups [Table 2.2]. There were no significant differences in mean arterial pressure or heart rate between control or endotoxin groups at any time. There were comparable but nonsignificant decreases in 29 Fig. 2.6. Myocardial capillary transit times of leukocytes. Bars are means ± SD; n = 4 animals in each group. Leukocytes in endotoxin group have significantly prolonged myocardial capillary transit times (* p < 0.05) compared with leukocytes in control group. 30 FIGURE 2.7 Myocardial capillary transit times of red blood cells CO Q z o o LU CO CO _l _l LU o Q o o _l CQ Q LU DC H CO z < cr i— >-or cu < 0.7 i 0.6 H 0.5 0.4 H 0.3 H 0.2 H 0.1 1 0.0 • ENDOTOXIN • CONTROL Fig. 2.7. Myocardial capillary transit times of red blood cells (RBCs). Bars represent means ± SD; n = 4 animals in each group. 31 FIGURE 2.8 Myocardial oxygen extraction ratio 100 -i o < DC f-X L U Z L U CD >-X o _ l • < Q DC < O o >-ENDOTOXIN CONTROL -10 0 10 20 30 40 50 60 70 80 TIME (MINUTES) Fig. 2.8. Myocardial oxygen extraction ratio. Values are means ± SD; n = 7 animals in each group. At 75 min after start of endotoxin infusion, myocardial oxygen extraction ratio decreased significantly to 45 ± 19 % from a preinfusion value of 68 ± 7 % (* p < 0.05). There was no significant change over time in myocardial extraction ratio of control group from a preinfusion value of 71 ± 8 %. 32 TABLE 2.2 Additional measured variables Mean Arterial Pressure, mmHg Group Preinfusion 30 min 75 min Endotoxin Control 94 ± 14 93 + 12 94 ±15 95+11 . 88 ± 19 93 ± 17 Heart Rate, beats/min Group Preinfusion 30 min 75 min Endotoxin Control 79 ±7 81 ±9 91 ± 11 82 ±4 98 ±7 88 ± 14 Body Temperature, °C Group Preinfusion 30 min 75 min Endotoxin Control 38.7 ± 0.1 38.7 ± 0.2 39.0 ± 0.4 38.5 ±0.1 39.1 ±0.3 38.5 ± 0.3 Hemoglobin, g/dl Group Preinfusion 30 min 75 min Endotoxin Control 8.3 ±0.7 8.2 ±0.8 8.2 ±0.6 8.1 ± 0.8 7.6 ± 0.7 7.7 ± 0.7 Table 2.2. Values are means ± SD; n = 7 pigs. 33 hemoglobin in both the endotoxin and control groups. 2.4 Discussion. This study demonstrates that during the infusion of endotoxin in a porcine model of sepsis, leukocyte transit times through myocardial capillaries are prolonged from 5 to 40 s, leading to the retention of >2 x 109 leukocytes in the coronary microcirculation/100 g of myocardium. Quantitative histological analysis shows that leukocytes are retained primarily within the myocardial capillary bed in concentrations approximately 100 times that of peripheral blood. On average, one leukocyte is retained within a myocardial volume of approximately (37 LUTI3) which is the volume of approximately eight cardiac myocytes (30). It is postulated, based on this degree of leukocyte slowing and retention, that leukocytes may contribute to myocardial dysfunction of sepsis (25, 27, 28, 34), just as similar degrees of leukocyte slowing and retention contribute to pulmonary dysfunction in sepsis and in the acute respiratory distress syndrome (6,22). Measurements of myocardial capillary transit time of red blood cells have recently been reported (1), however, the transit time of leukocytes in the myocardial capillary bed is unknown. The data used to calculate myocardial capillary transit time of leukocytes [Table 2.1] also allowed calculation of myocardial capillary transit time of red blood cells (1). The myocardial capillary transit time of red blood cells measured in pigs (0.4 s) is comparable to previous measurements in the rabbit using a similar technique (0.86 s, range 0.22 - 2.58 s) (1), suggesting that these measurements are reasonable. Capillary transit time of leukocytes through hamster cremaster muscle is not different from the 34 red blood cell transit time of approximately 1 s (13). Yet the mean capillary transit time of leukocytes through the pulmonary circulation is approximately 190 s (range 3.3 - 935 s) (21), which is about 60 times longer than the pulmonary capillary transit time of red blood cells (3.0 s, range 0.03 - 14.5 s) (21). Between these two extremes, it was found that the myocardial capillary transit time of leukocytes is 5 s, or approximately 10 times longer than simultaneously measured capillary transit time of red blood cells. Several observations suggest that leukocytes were activated following endotoxin infusion and that activation of leukocytes increased their capillary transit time. In previous in vitro (9) and in vivo studies in a range of animal models (2, 3, 35), endotoxin has been shown to activate leukocytes. Our measurements of CD18 expression following 50 u.g/kg endotoxin in pigs suggest that endotoxin significantly activated the leukocytes in this experiment. Once activated, leukocyte transit times through capillary-sized pores in vitro increases 10-fold (15). Activation of leukocytes results in temporary plugging of skeletal muscle capillaries (35), indicating increased leukocyte transit time. However, the extent of the increase in capillary transit time of leukocytes in vivo has not been fully elucidated, particularly in the coronary circulation. Similar to the capillary-pore measurements, it was found that myocardial capillary transit time of leukocytes increased by an order of magnitude, from approximately 5 to 40 s. Barroso-Aranda et al. (2) suggested that endotoxin causes myocardial capillary plugging by leukocytes. This may have been due to irreversible retention of leukocytes within myocardial capillaries without a change in capillary transit time of those leukocytes that were not retained or, alternatively, due to prolongation of myocardial capillary transit time of leukocytes. For 35 example, a doubling of transit time will double the number of leukocytes seen in the capillaries at a specific instant in time. These results suggest that prolongation of myocardial capillary transit times could account for these findings. However, these measurements were made 75 min after the start of endotoxin infusion, so irreversible retention of leukocytes may become a more important explanation for capillary plugging later on. A potentially important difference between the current study and that of Barroso-Aranda et al. (2) is that the current study used a lower dose of endotoxin (50 ug/kg vs. 9 mg/kg) in a different species (pig vs. rat). These lower endotoxin doses were nonlethal, induced minimal hemo-dynamic changes, and may be more relevant to clinical conditions. The potential causes of slowed myocardial capillary transit of leukocytes following endotoxin exposure include decreased leukocyte deformability and increased leukocyte-endothelial cell adherence (19). Potentially confounding effects of endotoxin, including increased coronary blood flow and increased catecholamines, most likely did not contribute to slowing or retention of. leukocytes, because increased blood flow decreases leukocyte retention in lungs (31) and catecholamines decrease leukocyte margination in the systemic circulation (4). During activation, there is a sharp decrease in the deformability of the neutrophil (24). Loss of deformability results in greatly increased neutrophil transit times through the pulmonary circulation (19, 23) due to the impedance of neutrophils as they attempt to negotiate pulmonary capillaries with mean diameters (6.03 ± 1.38 um) (10) narrower than the widest diameter of a neutrophil (6.6 ± 0.6 um) (10). Similarly, leukocyte transit could be slowed through the heart, where the capillary diameters (5.6 ± 1.3 urn) (17) are smaller on average than pulmonary capillary diameters. Transit times may also be 3 6 increased due to adherence of leukocytes to vascular endothelium, due to increased expression of adhesion molecules on leukocytes, primarily the integrin molecule CD18, and upregulation of corresponding ligands on systemic endothelial cells, primarily the immunoglobulin intracellular adhesion molecule-1 and the selectin E-selectin. Both alterations likely contribute to the increase in myocardial capillary transit time and retention of leukocytes above arterial concentrations in this study. There are two primary locations for the sequestration of leukocytes in a vascular bed; these are the capillaries and the postcapillary venules (11). The results of the present study show that leukocytes were retained primarily within capillaries and not within postcapillary venules. This observation differs from current paradigms, in which leukocyte retention is thought to occur in postcapillary venules and to be associated with the shedding of L-selectin from the surface of the circulating leukocyte (22). Leukocytes, once slowed and retained in the myocardial capillary bed, could contribute to impaired ventricular function in several ways. Humoral substances, including tumor necrosis factor-alpha (TNF-a) (34), interleukin-2 (IL-2) (25), and IL-6 (25), that are released by leukocytes could adversely affect the heart in sepsis (32). In cell culture, exposure to TNF-a concentrations 900-3,200 U/ml caused cardiac myocyte dysfunction (14). Although circulating TNF-a levels in sepsis may not achieve this concentration, local production of TNF-a or other toxic humoral substances by an activated leukocyte (20) could increase the intracardiac concentrations above the concentration in peripheral blood. For example, TNF-a production by activated neutrophils may represent a significant source of TNF-a within endotoxin-exposed lung tissue (37), 37 Neutrophils may also represent a significant source of nitric oxide and neutrophil chemoattractants (38). High local concentrations of specific cytokines and nitric oxide (5) may result in suppression of cell function, damage, or death. Cytokine-mediated neutrophil oxidative burst may be dependent on preceding CD11/CD18-mediated adhesion (12, 26), and the oxidative burst may damage cells. Direct cell damage may therefore be enhanced by prolonged contact between endothelial cells and neutrophils. The myocardial oxygen extraction ratio is one indicator of capillary gas exchange function. Normally, this very constant ratio is approximately 70-75% (33). In septic humans (8) and in the current study of endotoxin infusion, the myocardial oxygen extraction ratio is greatly decreased. In this experiment, increased coronary blood flow, possibly due to a number of vasoactive substances related to endotoxin infusion (29), accounts for some of the decrease in oxygen extraction ratio. In addition, it is speculated that slowed and retained leukocytes may also contribute to the observed decrease in oxygen extraction either by obstruction of the capillary bed or by release of vasoactive mediators. Because retained leukocytes were predominantly located in capillaries [Fig. 2.1], they could have caused focal mechanical obstruction of microvessels and impaired capillary gas exchange. A number of limitations to this data are important to recognize. The vast majority of leukocytes counted were neutrophils based on size, nuclear morphology, and cytoplasmic characteristics. The known volume of a neutrophil (10) was therefore used to calculate capillary leukocyte transit time [Table 2.1]. A second limitation is that, because single leukocytes almost completely filled individual myocardial capillaries, capillaries were counted as 38 either containing a leukocyte or not. This slightly overestimates the volume fraction of the capillary bed occupied by leukocytes. A third limitation is that it was assumed that blood flow through the LAD is a constant proportion of the total blood flow through the left ventricle. This seems to be a reasonable assumption, even in a model of sepsis, because significant changes between vascular territories within the same ventricle have not been demonstrated (16). Finally, the method used for calculating capillary transit time depends on steady-state conditions with no ongoing sequestration or loss of cells at the time of measurement. This condition would not have been met had these measurements been made during the 30-min period of endotoxin infusion but, after 75 min, ongoing sequestration of leukocytes had largely ceased [Fig. 2.2 and Fig. 2.31, 2.32]. Therefore transit times were calculated at this point in the experiment by fixing the heart for morphometric analysis immediately after myocardial blood volume and flow were measured. Diapedesis of leukocytes into tissue would tend to increase the estimate of capillary transit time of leukocytes. However, leukocytes in the tissues were not observed at 75 min after the start of the endotoxin infusion, which suggests that the number of leukocytes leaving the vascular space was likely small compared with the 25 x 10 6 (Ax H from Table 2.1) leukocytes/s being delivered by the coronary arteries. 2.5 Summary. In summary, these results show that the myocardial capillary transit of leukocytes is prolonged, leading to the retention of large numbers of leukocytes within the capillaries of the heart in an endotoxin-induced model of hyperdynamic sepsis. It is speculated that the activity of these leukocytes is 39 important to the early pathophysiologic events leading to reduced ventricular contractility and vascular dysfunction in sepsis. 40 CHAPTER III MYOCARDIAL MORPHOMETRIC CHANGES DURING SEPSIS; CORRELATION WITH INCREASED LEUKOCYTE RETENTION AND MYOCARDIAL CONTRACTILE FAILURE 3.1 Hypothesis A circulating myocardial depressant factor present in the plasma of septic humans (23) and in animal models of sepsis (9) can cause a rapidly reversible decrease in contractility of isolated myocardial tissue in vitro. Inflammatory mediators that have been considered as myocardial depressant factors, including TNF-a, IL-2, and IL-6 (7), and others, have half-lives that are in the order of several hours in vivo (10). However, the observed reduction in contractility is not rapidly reversible or short lived, lasting for approximately 10 days in human septic shock (22, 24) and in whole animal models of septic shock (20). This suggests that a reversible circulating myocardial depressant factor may not be the full explanation. It is hypothesised that blood-borne factors may also cause damage and death of myocytes and other myocardial structural changes, leading to decreased ventricular contractility. Decreased contractility has been shown to occur within 4-6 h of a septic stimulus (32), but it is not known whether myocardial structural changes occur early enough to account for this. A significant problem in studies of cardiac function is that indirect effects of septic shock on coronary perfusion pressure (28), preload (16), afterload (16), heart rate (19, 33), and the mechanical interaction of the heart with surrounding structures (15) may also lead to morphometric changes (31) and 41 decreased ventricular contractility or confound even load-insensitive measures of ventricular contractility (15). Whole animal studies that accurately model septic shock are unavoidably associated with changes in one or more of these confounding effects. As a result, the extent to which blood-borne elements account for the decrease in left ventricular contractility in vivo has not been fully elucidated independently of confounding effects. Conversely, in vitro studies of specific mediators, in which contractility can be measured independently of confounding effects, do not fully model the complex and dynamic septic inflammatory cascade. To address the hypothesis in a whole animal model of sepsis while controlling coronary perfusion pressure, preload, afterload, and heart rate and eliminating mechanical interaction of the heart with surrounding structures, a preparation consisting of an isolated rabbit heart attached to a modified Langendorff column that was perfused by a support rabbit was used. The support rabbit was rendered endotoxemic, and myocardial structural changes and contractility were assessed in the isolated heart. Coupling a whole animal model of sepsis to an isolated heart preparation allowed the determination of the role of blood-borne factors in causing myocardial structural changes and decreased ventricular contractility. 3.2 Experimental design. 3.21 Instrumentation. 3.211 Surgical preparation of the support rabbit. 42 Thirteen 3.5 ± 0.5 kg New Zealand White rabbits of either sex were anesthetized initially with an intramuscular injection of a mixture of ketamine (40 mg/kg; MTC Pharmaceuticals, Cambridge, ON, Canada) and xylazine (5 mg/kg; Chemagro Limited, Etobicoke, ON, Canada). To maintain deep surgical anesthesia for the duration of the experiment, a-chloralose (40 mg/kg; Sigma, St. Louis, MO) was then injected via a temporary venipuncture in the left marginal ear vein. The criteria for a surgical anesthetic plane was the absence of lacrimation and no change in heart rate or blood pressure after a painful stimulus of pressure was applied to a hind toe. Depth of anesthesia was tested hourly and prior to any intervention. No support animals required supplementary anesthesia for the duration of the experiment. A midline ventral incision was made in the neck and a tracheotomy tube was inserted. Rabbits were ventilated with air and supplemental oxygen, using a Harvard ventilator (Harvard Apparatus Canada, Saint-Laurent, Quebec) to maintain Po2 at approximately 400 mmHg and Pco2 at approximately 35 mmHg. Polyethylene catheters (ID 1.67 mm, OD 2.42 mm; Intramedic, Becton Dickinson, Parsippany, NJ) were inserted into the right carotid artery and the left external jugular vein to perfuse and drain the extracorporeal circuit of the isolated heart respectively. A three-way stopcock allowed the infusion of fluids and endotoxin via the jugular venous catheter. A polyethylene catheter (ID 1.14 mm, OD 1.57 mm; Intramedic, Becton Dickinson) was inserted into the abdominal aorta via the left femoral artery to monitor aortic blood pressure. Normal saline with heparin (7 lU/ml; Organon Teknika, Toronto, ON, Canada) was continuously infused at 12 ml/kg/h via the jugular venous catheter to maintain anticoagulation of the support rabbit. A rectal temperature probe was inserted. The core body temperature of the rabbit was maintained at 38.7°C (normal rabbit core body temperature) by-using a heating blanket. After approximately 15 min 43 equilibration, the animal received an intravenous bolus of heparin (Organon Teknika) of 1,000 lU/kg. After an additional 15 min period of stabilisation the support rabbit was connected to the Langendorff column and the extracorporeal circuit was established. 3.212 Hemodynamic and pH support of the support rabbit. Normal saline was infused periodically in 5-ml boluses to maintain the mean arterial blood pressure above approximately 90 mmHg [Table 3.1]. Slightly more normal saline was infused into the support rabbits of the endotoxin group (80 ± 15 ml) than in the control group (45 ± 10 ml). The arterial pH of the support rabbit was maintained at approximately 7.4 by periodic IV infusions of 7.5% sodium bicarbonate via the jugular venous catheter. There was no significant difference in the total volume of sodium bicarbonate infused into the support rabbits of the endotoxin (12 ± 4 ml) and control groups (9 ± 3 ml). 3.213 Surgical preparation of the isolated heart Thirteen 2.5 ± 0.5 kg rabbits were anesthetized using a mixture of ketamine (40 mg/kg) and xylazine (5 mg/kg) intramuscularly. Midline sternotomy and pericardotomy were performed. The heart was rapidly excised, weighed and affixed by the aorta to the perfusion column so that it was now perfused by carotid arterial blood from the support rabbit. The pulmonary artery was incised at the base of the right ventricular outflow tract to allow unobstructed drainage from the right ventricle. A cannula (ID 0.58 mm, OD 0.965 mm; Intramedic, Becton Dickinson) was inserted into the isolated heart 44 TABLE 3.1 Hemodynamic and physiologic parameters for support rabbits Time, h 0 1 2 3 4 5 Heart rate Endotoxin Control 271 ± 27 264 ± 24 257 ± 19 244 ± 22 245 ± 1 3 243 ± 21 233 ± 30 240 ± 17 230 ± 28 * 225 ± 17* 218 ± 2 8 * 225 ±17 * MAP Endotoxin Control 114 ± 2 5 110 + 23 112 ± 2 3 110±25 99 ±31 106 ± 2 0 103 ±29 98 ±21 99 ±24 94 ± 14 85 ±31 90 ± 14 [WBC] Endotoxin Control 7.3 ± 4.1 7.0 ± 2.3 2.5 ± 1 . 3 * 7.1 ± 1.8 1.9 ± 0 . 5 * 8.5 ±1.4 1.9 ±0 .6 * 8.2 + 2.4 1.9 ±0.6 * 9.0 ±1.8 2.1 ± 0 . 8 * 9.8 ±3.1 Hemoglobin Endotoxin Control 8.7 ± 1.2 7.9 ± 1.1 8.4 ± 1.7 7.6 ± 1.0 7.7 ± 2.1 7.4 ± 1.3 7.4 ± 1.9* 7.3 ± 0.8 * 7.0 ± 1 . 9 * 7.0 ± 0.9 * 6.7 ± 1.7* 7.0 ± 0.9 * Lactate Endotoxin Control 1.7 ±0.6 2.4 ± 0.6 2.6 ± 1.2 3.1 ± 0 . 8 * 5.6 ± 2.4 4.1 ±1 .0 * 8.3 + 3.6 5.7 + 1.4 * 10.5 ±4.4 7.9 + 1.9* 13.6 ± 4 . 4 * 10.2 + 3.2* Table 3.1. Values are means ± SD; n = 6 control rabbits and n =7 endotoxin rabbits. * P < 0.05 compared with time 0. Heart rate, beats/min. MAP, mean arterial pressure, mmHg. [WBC], leukocyte concentration, cells x 109. Hemoglobin, g/dl. Lactate, mmol. 45 via an apical ventricular stab incision to drain the thebesian venous flow. A 5-mm incision was made in the left atrium and a 7-F single-lumen pressure transducer catheter (Millar Instruments, Houston, TX) surrounded by a latex balloon was inserted into the left ventricle (28) and fixed in place using an external ligature surrounding the left atrium. The latex balloon did not develop measurable transmural pressure until it had been inflated to a volume greater than 3 ml. This ensured that at balloon inflation volumes less than 3 ml the measured ventricular pressure was due to ventricular forces and not due to the balloon. Pacing electrodes were attached to the surface of the right ventricle and left atrial appendage. The isolated heart was paced at 150 beats/min. The ventricular balloon was inflated with normal saline via the Millar catheter until a left ventricular end-diastolic pressure (LVEDP) of 4 mmHg was attained. At no time was the balloon inflated to volumes greater than 0.6 ml. The isolated, supported heart was then allowed to beat isovolemically for an equilibration period of 1 h. 3.214 Langendorff column and extracorporeal circuit A modified Langendorff column and extracorporeal circuit were utilised after Sandhu et al. (25) [Fig. 3.1]. Arterial blood from the support rabbit was pumped, using a roller pump via the carotid arterial catheter, to a 37°C water-jacketed, glass heating chamber. From the chamber, blood flowed to an open perfusion column that was attached to the proximal aorta of the isolated heart to retrogradely perfuse the coronary arteries of the isolated heart. The perfusion column height was raised to produce a pressure of 75 mmHg at the level of the aortic valve. The isolated heart was contained in a heated 37°C water-jacketed, glass reservoir. Blood overflowing from the perfusion column and venous blood 46 FIGURE 3.1 Isolated-supported rabbit heart model ENDOTOXIN "J" (1000ug/kg I.V8) Fig. 3.1. Isolated-supported rabbit heart model. Blood is pumped from carotid artery of support rabbit to an open perfusion column at a height of 75 mmHg. Coronary arteries of isolated heart are retrogradely perfused via the aorta. Venous blood and blood spilling from open perfusion column is pumped back to support rabbit via a 40 urn blood filter. In endotoxin group, support rabbit is rendered endotoxemic and myocardial function is assessed over time in the isolated heart. Emax. the slope of end-systolic pressure-volume relationship. 47 from the isolated heart were pumped using a roller pump through a 40-u.m blood filter (SQ40S Blood Transfusion Filter, Pall Biomedical Products, East Hills, NY) back to the support rabbit via the jugular venous catheter. The total volume of the extracorporeal circuit was approximately 35 ml. Before connection of the support rabbit, the extracorporeal circuit was filled with Plasma-Lyte A (Baxter, Toronto, ON, Canada) solution containing 150 lU/ml heparin. In addition, 20 ml of 6% Dextran 70 in normal saline (Macrodex, Pharmacja, Mississauga, ON, Canada) were added to the reservoir as blood was initially displaced from the donor rabbit. The support rabbit was allowed to equilibrate for approximately 45 min before being connected to the isolated heart. 3.22 Experimental protocol. After the 1-h equilibration period of the isolated hearts, a baseline E m a x measurement was made. The support rabbits then received either a control infusion of normal saline (n = 6) or an infusion of endotoxin (Escherichia coli 0111 :B4, Sigma) 1 mg/kg (n = 7) over 30 min via the left femoral venous catheter. Measurements of all variables were then repeated at 1 -h intervals for a total of 5 h after the start of the infusion. At the conclusion of the protocol, the blood perfusion circuit was interrupted and oxygenated normal saline at 37°C was simultaneously infused into the perfusion column at 75 mmHg to wash all red blood cells from the myocardial circulation (approximately 4 min). Glutaraldehyde, 2.5% in phosphate buffer, was then added to the saline perfusion circuit and the heart perfusion was fixed for 15 min. The isolated heart was then removed from the apparatus and transferred to a container of the same fixative for 48 h prior to processing. 48 3.23 Measurements. 3.231 Measurement of left ventricular contractility Left ventricular contractility for each time point was determined as the slope of the end-systolic pressure-volume relationship (ESPVR), E m a x . E m a x is a measure of left ventricular contractility, which is relatively independent of ventricular preload and afterload conditions (16). E m a x has previously been shown to be a reliable measure of ventricular contractility in a blood-perfused isolated heart model (34). The ESPVR was determined by periodically inflating the intraventricular balloon at a constant rate of 0.8 ml/min using a syringe pump. A constant balloon inflation rate allowed time measurements made from the start of balloon filling to be directly converted to intraventricular volume measurements. The maximum volume of balloon inflation was predetermined as the volume at which cardiac dysrhythmia occurred (approximately 0.6 ml). At no time did maximum balloon inflation volume approach the maximum volume of the balloon (3 ml). During inflation, left ventricular pressure measured with the Millar transducer was simultaneously sampled at 250 Hz and stored in digital format. After inflation to maximum volume, the balloon was deflated without delay at the same rate of 0.8 ml/min. A typical plot of the ESPVR is shown [Fig. 3.2]. The slope of the best fit line of the ESPVR is E m a x . 3.232 Other measurements At each measurement set arterial blood samples were drawn from the support rabbit for determination Of Po 2, Pco2, pH, lactate, hemoglobin, and total leukocyte count. The heart rate of the support rabbit was also noted. 49 FIGURE 3.2 Plot of left ventricular pressure versus left ventricular volume cn I E £ LU DC ZD CO CO LU DC 0_ 70 - , 60 50 40 H 30 -\ 20 -J 10 H 0 200 250 300 350 400 VOLUME (ul) Fig. 3.2. Typical plot of left ventricular pressure vs. left ventricular volume. These data were obtained from 1 experimental set by increasing the volume of the isolated heart intraventricular balloon. A line plotted through left ventricular peak systolic pressure points represents end-systolic pressure-volume relationship (ESPVR). Slope of ESPVR is Em ax. a measure of contractility relatively independent of ventricular loading conditions. 50 3.234 Myocardial morphometric analysis. Myocardial morphometric analysis was carried out using the technique of Cruz-Orive and Weibel (3) described in Chapter II (section 2.235, Myocardial leukocyte content predicted from arterial leukocyte concentrations). This method is reiterated here in order to illustrate minor methodologic differences. The glutaraldehyde-fixed left ventricle was sliced in a transaxial orientation into adjacent 3-mm tissue disks (1). Four tissue samples were removed from each disk. Ten tissue samples were selected at random from the sum of all samples and embedded in glycol methacrylate. The resulting tissue blocks were then sectioned at 2 u.m and stained with toluidine blue 0. Left ventricular morphometry was assessed quantitatively by point counting using a multilevel technique similar to that described by Cruz-Orive and Weibel (3). Point counting was carried out at two levels of magnification: level 1 at 40x magnification to distinguish large (>10 urn diameter) blood vessels from myocardium, and level 2 at 100x magnification to distinguish small blood vessels (<10 urn diameter), including capillaries, interstitium, leukocytes and structurally abnormal myocytes from normal myocardium. Capillaries were distinguished from other small (<10 u,m diameter) blood vessels by the absence of vessel wall components other than endothelium. Where significant doubt existed, higher magnification was temporarily used or the point was discarded and an additional point was assessed. This technique generates the relative volume fraction of the object counted. By this means, the volume fractions of 1) capillaries occupied by leukocytes, 2) structurally abnormal myocytes, and 3) myocardial interstitium relative to total myocardial volume were determined. Only those myocyte structural abnormalities visible by light microscopy at 10Ox 51 magnification were quantified. Myocyte structural changes defined as abnormal consisted of myocyte swelling, nuclear swelling and hypochromasia, cytoplasmic vacuolation, zonal contraction banding, and contraction band necrosis. Five random fields per tissue sample at both magnifications were counted with a 90-point grid and yielded a relative standard error of volume fraction of < 10%. 3.24 Data analysis. Differences in morphometric measures were tested between the control and endotoxin groups using unpaired t-tests. Differences in E m a x of the isolated hearts overtime were tested between the control and endotoxin groups using a two-way analysis of variance with one-way being repeated measures. When a significant difference was found, specific differences were.tested between the control and endotoxin groups using a sequentially rejective Bonferroni test procedure (11) to correct for multiple comparisons. Changes over time within each group were tested by using the first-order polynomial for a difference from zero. A p < 0.05 was chosen as statistically significant. Results are reported as means ± SD throughout. 3.3 Results At the end of the experiment the volume fraction of the myocardium occupied by capillaries was not significantly different between groups (11.1 ± 3.3% in the endotoxin group and 10.5 ± 1.6% in the control group). The volume fraction of the myocardium occupied by capillaries containing leukocytes was 52 significantly greater in the endotoxin-treated group (1.74 ± 0.63%, p < 0.05) than in the control group (0.32 ± 0.10%). Therefore, the volume fraction of myocardial capillaries occupied by leukocytes was significantly greater in the endotoxin group (15.7 ± 3.5%, p < 0.05) vs. the control group (3.0 ± 0.7%) [Fig. 3.3]. The majority of leukocytes counted were morphometrically consistent with polymorphonuclear neutrophils and were located within myocardial capillaries. The volume fraction of structurally abnormal myocytes was significantly greater in the endotoxin-treated group (7.6 ± 3.6%, p < 0.05) than in the control group (0.8 ± 0.4%) [Fig. 3.4]. The observed myocyte abnormalities consisted of focal diffuse areas of myocyte swelling, nuclear swelling and hypochromasia, cytoplasmic vacuolation, zonal contraction banding, and contraction band necrosis. The volume fraction of myocardial interstitium was significantly greater in the endotoxin group (23.2 ± 5.2%, p < 0.05) than, in the control group (14.3 ±2.1%) [Fig. 3.5]. During the experiment there were no significant differences in E m a x between the two groups at baseline or for the first 3 h after the start of endotoxin infusion. By 4 and 5 h, E m a x of isolated hearts whose support rabbits had received endotoxin had decreased'significantly by 15 ± 5 (p < 0.05) and 17 ± 7% (p < 0.03), respectively. There was no change in E m a x in the control group from baseline values (0 ± 2%) [Fig. 3.6]. There was a significant decrease in the heart rate of the support rabbits from both the endotoxin and control groups by 4 h after infusion. At no time point was there a significant difference in heart rate between groups [Table 3.1]. 53 FIGURE 3.3 Volume fraction of myocardial capillaries occupied by leukocytes 20 -, CO LU o o LU _J CD O O CO LU a. < O LL O O < CC 10 A ENDOTOXIN CONTROL Fig. 3.3. Volume fraction of myocardial capillaries occupied by leukocytes. More myocardial capillaries in endotoxin group were occupied by leukocytes (* p < 0.05) than in control group. 54 Fig. C.4. More cardiac myocytes were abnormal in endotoxin group (* p < 0.05) than in control group. Observed myocyte abnormalities consisted of focal diffuse areas of myocyte swelling, nuclear swelling and hypochromasia, cytoplasmic vacuolation, zonal contraction banding, and contraction band necrosis. 55 FIGURE 3.5 Volume fraction of myocardium occupied by interstitium 40 i CO DC LU O < CC LL LU _1 o > 30 A 20 A 10 A 0 ENDOTOXIN CONTROL Fig. 3.5. Morphometric volume fraction of total myocardium that was interstitium is shown. Interstitial space was greater in the endotoxin group (* p < 0.05) than in control group. 56 FIGURE 3.6 Change in E m a x versus time E UJ < 20 n 10 0 4 -10 4 -20 4 -30 CONTROL ENDOTOXIN INFUSION - i 1 r 0 0.5 1 3 4 TIME (HOURS) Fig. 3.6. Change in Emax vs. time (h after the start of infusion, bar) is shown for endotoxin and control groups. At 4 and 5 h after start of infusion, hearts from endotoxin group show a reduction • n Emax (* P < 0.05), whereas hearts from control group show no change in E m a x -57 Although the mean arterial pressure of the support rabbits from both the endotoxin and control groups tended to decrease over time, the decrease was not statistically significant. At no time was there a difference in mean arterial pressure between groups [Table 3.1]. The total leukocyte concentration of the endotoxin group support rabbits fell significantly from 7.3 ± 4.1 X 109 to 2.1 ± 0.8 X 109 cells (p < 0.05), indicating trapping of leukocytes in the lungs and peripheral tissues, whereas there was no significant change in the total leukocyte concentration of the control group [Table 3.1]. Both the endotoxin and control group support rabbits experienced significant drops in arterial hemoglobin by 3 h postinfusion. However, there were no significant differences in arterial hemoglobin values between groups at any time during the experiment [Table 3:1]. Both the endotoxin and the control groups experienced significant absolute increases in arterial lactate levels [Table 3.1], but there was no significant difference in lactate levels between the groups at any time during the experiment. 3.4 Discussion These results demonstrate that blood-borne factors cause myocardial morphometric changes within 5 h of the onset of endotoxin infusion. These changes consist of an increased number of leukocytes in myocardial capillaries, an increased volume of structurally abnormal myocytes, and an expanded myocardial interstitial volume. These myocardial structural changes were associated with decreased ventricular contractility in the isolated, supported rabbit heart. Decreased contractility was independent of any changes in myocardial perfusion pressure, preload, afterload, heart rate, and ventricular interaction because these factors were controlled in the isolated, supported 58 heart where E m a x was measured. Taken together, these results suggest that myocyte damage and other myocardial structural changes may contribute to decreased ventricular contractility and may explain the prolonged duration of decreased contractility during sepsis. A significant accumulation of leukocytes within myocardial capillaries was observed following endotoxin exposure [Fig. 3.3]. It has previously been shown that leukocyte transit through myocardial capillaries is greatly slowed and large numbers of leukocytes are retained in myocardial capillaries within 30 min of the start of an endotoxin infusion in pigs (8). Barroso-Aranda et al. (2) have similarly suggested that endotoxin causes myocardial capillary plugging by activated leukocytes. Leukocytes are also retained in capillaries during myocardial ischemia-reperfusion injury (5) and after activation with complement fragment C5a (18) and contribute to tissue damage (5, 18). On the basis of these observations, it is postulated that, after endotoxin infusion, leukocytes are initially retained in myocardial capillaries and then contribute to later myocyte damage and interstitial edema. Retained or slowed leukocytes could potentially cause myocyte damage in a number of ways. Monocytes and polymorphonuclear neutrophils are capable of releasing TNF-a, interleukins, products of the lipoxygenase pathway, and other substances capable of reducing contractile function or damaging the myocyte (13, 35). It has further been shown that the non-specific oxidative burst of activated polymorphonuclear neutrophils is capable of causing myocyte damage (6). In addition, activated leukocytes block capillaries (2, 8, 18). This could lead to local tissue hypoxia and an inadequate supply of metabolites. Tissue hypoxia due to blockage of capillary microvasculature, disruption of 59 mitochondria, or impairment of metabolic pathways could lead to decreased ventricular contractility (30). Subsequent death or disruption of ventricular myocytes could lead to the increased volume fraction of structurally abnormal myocytes. The myocardial structural changes observed may then lead to decreased ventricular contractility, much as similar changes do during ischemia-reperfusion injury (5). In the endotoxin group, the volume of the myocardial interstitial space was increased, indicating myocardial edema. Potential causes of myocardial edema include increased capillary permeability due to the inflammatory effects of cytokines, leukocytes, and oxygen free radicals. Myocardial edema also occurs during ischemia-reperfusion and has been postulated to affect the systolic function of the left ventricle (12). Myocardial edema may also represent a relative diffusion barrier for metabolic substrates. It is possible that the infusion of normal saline into the support rabbits may have contributed to myocardial edema. However, the total volume of saline infused into the endotoxin group was not significantly different from the volume infused into the control group. Furthermore, despite a significant decrease in hemoglobin by 3 h post-infusion in both groups, at no time was there a significant difference in hemoglobin between groups. This suggests that expansion of the interstitial space by edema fluid was mainly due to endotoxin infusion and could account for part of the observed decrease in ventricular contractility (12). Coupling a whole animal model of sepsis to an isolated heart preparation provides new information on the role of blood^borne factors in causing decreased ventricular contractility. The reason is that, in humans and in whole animal preparations that reasonably model septic shock, accurate 60 measurement of ventricular contractility is difficult. Common clinical indexes, including ejection fraction and fractional shortening, are significantly altered by changes in afterload and, to a lesser extent, by changes in preload (16). Other measures, including maximum velocity of shortening and maximum rate of change of pressure with respect to time, are more influenced by changes in preload (16). Indexes such as E m a x , that are relatively preload and afterload insensitive, are difficult to measure in humans and, even then, are significantly altered by changes in heart rate (19) and by mechanical interaction of the heart with surrounding structures (15). Completely preventing the indirect effects of hypotension, heart rate changes, and other influences on ventricular contractility is impossible (14, 21, 26, 32). Conversely, all of these variables can be controlled in isolated heart preparations. Two major problems with isolated heart preparations are 1) the septic inflammatory response is complex so that it is not fully modelled by addition of specific mediators in vitro and 2) isolated heart preparations are unstable and remain viable for <3 h (11), which is insufficient time to study decreased contractility that occurs 4-6 h after a septic stimulus (32). When an isolated heart is perfused with blood from a support rabbit, these problems are overcome because the isolated heart is exposed to the complete in vivo milieu. Furthermore, the preparation is extremely stable so that contractility at the end of the 5-h control experiments was unchanged and the SD was low (0 ± 2% change from baseline), which indicates minimal variability. The decrease in contractility observed here is only due to blood-borne factors. Preventing changes in coronary perfusion pressure (28), preload and afterload (31), heart rate (33), and interaction of the left ventricle with surrounding structures (15) was important because they can, by themselves, 61 change ventricular function. Decreased coronary perfusion pressure leading to tissue hypoxia (28) decreases contractility, making all whole animal models of septic shock difficult to interpret. Even when decreased coronary perfusion pressure does not result in evidence of frank global ischemia (4), decreased, myocardial function can be observed (17). Acute changes in preload and afterload alone do not substantially alter values of load-insensitive indexes of contractility, including E m a x used in this study (16). However, prolonged (6-8 h) reduction in preload and afterload results in a substantial reduction in ventricular function (31). Acute increases in heart rate generally increase ventricular contractility (19), yet a prolonged increase in heart rate is used as a method of inducing decreased ventricular contractility (33). Interaction of the left ventricle with surrounding structures affects ventricular function (15). This can be avoided to some extent in open-chest studies (32). However, elevated pulmonary artery pressures produce significant right-to-left septal shift, resulting in altered measures of left ventricular systolic and diastolic function (15) even in open-chest studies. Thus, even in recent studies using load-independent indexes of ventricular contractility, evidence for changes in cardiac'function due to blood-borne factors is confounded by decreases in coronary perfusion pressure, preload, afterload, increased heart rate, and pulmonary hypertension and other causes of ventricular interaction. Measurements of E m a x have not been reported during human sepsis, to our knowledge, so that our observation of 17 ± 7% decrease in E m a x is not directly comparable. However, E m a x is approximately linearly related to changes in ejection fraction over this range when preload, afterload, and heart rate are maintained constant, (32) as they were in this study. Parker et al. (22) 62 reported an 11% decrease in ejection fraction in septic humans with severe sepsis (35% mortality rate). A similar reduction in ejection fraction occurred 5 h after endotoxin administration in humans (26), which was associated with a 23% decrease in the ratio of peak systolic pressure to end-systolic volume. Thus the observed 17% decrease in E m a x appears to be very similar in magnitude to the decrease in contractile function observed during human sepsis. ^ In this model blood-borne factors include those in the plasma and those in the cellular component of blood (polymorphonuclear neutrophils, monocytes, etc.). Studies by Parrillo and co-workers (23) have demonstrated the presence of a soluble myocardial depressant substance in the plasma of septic patients. Such a factor could account for some of the reduction in contractility observed in this model. It is possible that the persistent effect of an otherwise reversible myocardial depressant factor could cause the observed myocardial damage. Myocardial damage could perpetuate reduced contractility even after the factor was cleared from the circulation, thus accounting for the prolonged (10 day) myocardial depression in sepsis. A significant increase in lactate levels was observed in the support animals in this model. It is possible that the cellular effects of increased lactate could adversely affect the ventricular myocytes (29). However, there was no significant difference in lactate observed between endotoxin and control groups. 6 3 3.5 Summary It is concluded that blood-borne factors, including both plasma and cellular elements, cause myocardial structural changes associated with reduced myocardial contractility in an intact isolated, supported rabbit heart model of experimental endotoxemia. Septic myocardial dysfunction may represent the net effect of both rapidly reversible effects mediated by circulating myocardial depressant factors and much more prolonged myocardial structural changes. 64 CHAPTER IV FRACTIONAL LEUKOCYTE TRANSIT TIME AND RETENTION DURING SEPSIS 4.1 Hypothesis Leukocytes are retained in the coronary circulation of animals during experimental endotoxemia (7, 1). Retention of leukocytes is associated with irreversible damage to a portion of the myocardium during sepsis and may lead to myocardial contractile dysfunction (8). Exclusion of leukocytes from the circulation of a heart otherwise exposed to the products of the septic cytohumoral cascade protects the heart from both myocardial damage and contractile dysfunction (9). Early leukocyte retention in animal models of sepsis is primarily within the capillary bed, and not in the post-capillary venules where receptor-ligand mediated adherence is the main mechanism of retention. Conceivably, neutrophils that become rigid upon activation lodge in smaller coronary capillaries so that systemic activation of leukocytes is the primary event leading to leukocyte retention in the heart during early sepsis. Alternatively, activation of coronary capillary endothelium may be more important. Whether early leukocyte retention in the myocardium is dependent on leukocyte activation, activation of the coronary endothelium, or a combination of both is not known. To address this issue an isolated-supported rabbit heart model of sepsis that has previously been shown to replicate both myocardial depression and leukocyte retention observed during whole animal sepsis was utilised (8). By measuring multiple consecutive leukocyte concentrations in blood flowing into 65 (I(t)) and flowing out (O(t)) of the coronary circulation at the time of a step change in arterial leukocyte concentration, the leukocyte transit time can be determined as the mean of the transfer function (T(t)) where I (t) • T(t) = O(t) By measuring differential leukocyte fraction's of leukocyte distributions the relative contributions of neutrophils (PMN) and lymphocytes to the total leukocyte coronary transit time were determined (see Fig. 4.2 and Fig. 4.3, section 4.233, Calculation of transit times). 4.2 Experimental design 4.21 Instrumentation. 4.211 Surgical preparation of the support rabbit. Twenty-four 2.5 ± 0.5 kg female New Zealand White rabbits were surgically prepared as described in Chapter III (section 3.211 Surgical preparation of the support rabbit). 4.212 Hemodynamic and pH support of the support rabbit. Normal saline was infused periodically in 5 ml boluses to maintain the mean arterial blood pressure above approximately 90 mmHg. Slightly more normal saline was infused into support rabbits receiving endotoxin than in support rabbits receiving vehicle only (40 ± 10 ml, n=12 versus 30 ±5 ml, n=12 respectively), however, this was not significantly different. The arterial pH of the support rabbit was maintained at approximately 7.4 by periodic infusions of 7.5 66 % sodium bicarbonate via the jugular venous catheter. There was no significant difference in the total volume of sodium bicarbonate infused into the support rabbits which received endotoxin (5 ± 3 ml) versus rabbits which received saline (4 ± 1 ml). 4.213 Surgical preparation of the isolated heart. The surgical preparation of the isolated heart was similar to that described in Chapter III (section 3.213 Surgical preparation of the isolated heart). The essential differences are iterated below. A left ventricular balloon was not inserted for measurement of myocardial contractility and a Thebesian drain was not inserted for reasons stated below. Twenty-four 2.1 ± 0.4 kg rabbits were utilised. A Thebesian drain was not inserted owing to the very small proportion of the total myocardial blood flow (< 5 %) accounted for by the Thebesian circulation. The absence of a Thebesian drain also prevented contamination of collected fractions by blood potentially flowing at a different rate than the main coronary circulation. The isolated heart was paced at 150 beats/min and allowed to equilibrate for 15 min in order to ensure stability of rhythm and total coronary blood flow. 4.214 Langendorff column and extra corporeal circuit. A modified Langendorff column and extra corporeal circuit was utilised as described in Chapter III (section 3.214 Langendorff column and extra corporeal circuit) [Fig. 4.1]. Arterial blood from the support rabbit was pumped via the carotid arterial catheter using a roller pump (Masterflex, Cole-Parmer 67 FIGURE 4.1 Transit time perfusion apparatus G R O U P G R O U P V E N O U S B L O O D Fig. 4.1. Perfusion apparatus. Apparatus depicted is a variation of that shown in Chapter III, Fig. 3.1. Note the addition of two in-series leukocyte filters to exclude all leukocytes from the support rabbit from the isolated heart. Flow pathways depicted with dotted lines are operative during the period of leukocyte infusion only. 68 Instrument Co., Chicago, IL) through two leukocyte filters (Pall Biomedical Products Co., East Hills, NY) placed in series in order to exclude all support rabbit leukocytes from the circulation of the isolated heart (< 1.0 x 10^ leukocytes/I). Total coronary blood flow was continuously measured using an in-line ultrasonic flow transducer (Transonic Systems Inc., Ithaca, NY). 4.215 Preparation of leukocytes for perfusion. Twenty-four 2.0 ± 0.5 kg rabbits were anesthetized using a mixture of ketamine (40 mg/kg) and xylazine (5 mg/kg) intramuscularly. A polyethylene catheter (ID 1.67 mm, OD 2.42 mm) was inserted into the right carotid artery to harvest blood for perfusion of leukocytes into the isolated heart. An identical catheter was inserted into the right external jugular vein for the infusion of endotoxin or vehicle. Supplemental oxygen was administered via face mask to the donor rabbit to maintain a PQ2 equal to the blood of the support rabbit perfusing the isolated heart. The body temperature of the donor rabbit was maintained at 38.7°C using a heating blanket. The donor rabbit was then anti-coagulated with an IV bolus of heparin 1000 lU/kg. Fifteen minutes later either endotoxin (E. coli 0111 :B4,100 ug/kg IV over 30 min, Sigma, St. Louis, MO) or vehicle was infused. Four minutes after the start of endotoxin infusion blood was pumped at 2.0 ml/min using a syringe pump (Harvard Apparatus Canada, Saint-Laurent, Quebec) via the carotid artery catheter into a sealed heparinized (heparin 500 ul of 1:1000) 35 ml syringe which was gently agitated continuously to prevent settling of formed blood elements. Test aliquots from leukocyte containing syringe blood and support rabbit blood were collected immediately prior to the perfusion run for analysis of total leukocyte count and hematocrit 69 (S880 Automated Hematology Analyser, Coulter Counter, Coulter Electronics, Hialeah, FL) , leukocyte differential analysis, and PQ2-4.22 Experimental protocol The isolated heart was perfused with carotid arterial blood (passed through 2 leukocyte filters) from the support rabbit for 2 h after the start of an endotoxin infusion (100 u.g/kg IV over 30 min) or a similar vehicle (control) infusion. Then perfusion was immediately switched to blood from the blood donor rabbit in the 35 ml syringe, less than 30 s after this donor blood had been collected. The flow rate and perfusion pressure was matched to the pre-donor blood values. When the complete leukocyte containing sample (20 of 35 ml) had been infused the leukocyte-free support rabbit perfusion circuit was restored without delay or alteration in perfusion pressure or flow in order to allow leukocytes still transiting the heart to exit the system. A fraction collector was used to collect coronary venous effluent in continuous 0.5 ml aliquots in individual glass collection vials containing 40 u.l EDTA solution. Six baseline samples of venous effluent during support rabbit perfusion were collected. Venous aliquot sampling was continued during and after donor blood infusion until leukocyte concentrations in venous blood had decreased back to baseline. Comparison of differences between support rabbit and leukocyte containing blood hematocrit allowed calculation of the degree of mixing of both blood types in any given syringe. In all experiments mixing of blood in venous effluent had ceased by 3 s after the step change in perfusion indicating that 70 there was a minimum of intracoronary blood mixing. Comparison of total leukocyte count and hematocrit of blood samples from the infusion syringe before and after the infusion run showed no significant differences indicating homogeneity of the leukocyte containing blood throughout the infusion interval. Four experimental groups were studied. Group 1 (n=6): control, support rabbit and blood donor rabbit received vehicle only (L-/H-). Group 2 (n=6): blood donor rabbit received endotoxin. The support rabbit received vehicle only (L+/H-). Group 3 (n=6): the support rabbit received endotoxin 100 ug/kg IV over 30 min via the jugular venous catheter two hours prior to infusion of the leukocyte containing blood. The blood donor rabbit received vehicle only (L-/H+). Group 4 (n=6): the blood donor rabbit received endotoxin as in Group 2 and the support rabbit received endotoxin as in Group 3 (L+/H+) (L, leukocytes, H, Heart, + activated, - non-activated). At the completion of the protocol the support rabbit circulation was interrupted and oxygenated normal saline at 37°C was infused at identical flow and pressure until the venous effluent was free of red blood cells (approximately 2 min). Glutaraldehyde 2.5% in phosphate-buffered saline was then added to the perfusion circuit and the heart was perfusion fixed for 15 min. The aorta of the isolated heart was then ligated and the heart was immersed in a large volume of 2.5% glutaraldehyde in phosphate-buffered saline to fix for an additional 48 h. 4.23 Measurements. 4.231 Measurement of leukocyte activation. 71 The infusion of endotoxin in-vivo leads to the activation of circulating leukocytes (1, 4, 33). To confirm that leukocyte activation occurred in this experiment baseline expression of CD18 (33) on polymorphonuclear neutrophils was determined using immunofluorescent flow cytometric analysis of blood samples taken from donor rabbits prior to operation (baseline). A blood sample, following infusion of endotoxin or vehicle into the donor rabbit, was taken directly from the syringe sample to determine the degree of activation of leukocytes prior to infusion through the isolated heart. Ten sequential blood samples were then obtained from the blood aliquots following transit through the isolated heart and the resultant mean value for CD18 activation was used to describe the activation of blood following infusion through the isolated heart as per the method described in Chapter II (section 2.238, Measurement of leukocyte activation). Data was expressed as mean fluorescence intensity (MFI) and normalised to the MFI of leukocytes in blood from donor rabbits prior to operation (baseline). 4.232 Calculation of total and differential leukocyte content Measurements of total leukocyte concentration (S880 Automated Hematology Analyser, Coulter Counter, Coulter Electronics, Hialeah, FL) were made on each 0.5 ml aliquot of blood collected following transit through the coronary circulation of the isolated heart and on a sample aliquot from the syringe prior to and following infusion of endotoxin or vehicle. Similar samples were analysed from each blood donor rabbit prior to infusion of endotoxin or vehicle. In order to determine the relative content of various leukocyte populations a thin blood smear was prepared from each aliquot and from the 72 syringe sample. Blood smears were stained with Quick Wright's (Cameo Quick Stain II, Cambridge Diagnostic Products Inc., Fort Lauderdale, FL). Differential leukocyte count was quantified by counting 100 cells/smear at 40 x magnification. Multiplication of the individual leukocyte fraction of an aliquot by the total number of leukocytes in the same aliquot yielded the total number of each leukocyte fraction in that aliquot. In this way the total number of lymphocytes and neutrophils (PMN) in each aliquot were quantified. Following activation by endotoxin, sample aliquots of blood contained only negligible numbers of monocytes and so these cells were not quantified. 4.233 Calculation of transit times Leukocyte number was plotted versus time of the venous aliquots [Fig. 4.2]. Since the input distribution function l(t) [Fig. 4.2] is a step function (going from a leukocyte concentration ratio of zero to the constant value in the leukocyte containing syringe) the transit time distribution (T(t)) [Fig. 4.3] is simply the time derivative of the output distribution function O(t) [Fig. 4.2] where l(t). T(t) = O(t) The mean to the transit time distribution is the mean leukocyte transit time for the apparatus and the heart combined. To determine the transit time of the apparatus, five calibration experiments were performed in identical fashion but with exclusion of the isolated heart over a range of flow rates. By performing an identical transit time analysis on the resultant aliquots the mean time for leukocytes to transit the perfusion apparatus was calculated. A regression analysis on the relationship between flow and perfusion apparatus transit time gave an r2 value of 0.98 indicating that apparatus transit time could be closely 73 FIGURE 4.2 Total leukocyte input distribution (l(t)) and output distribution (O(t)) versus time Fig. 4.2. The number of leukocytes/0.5 ml aliquot sample from one experiment are plotted versus time (open square). The best fit curve of this raw data (solid line) is the output distribution function 0(t) for the experiment. The input distribution function l(t)SR during support rabbit perfusion (open triangle) (< 0 s and > 60 s) is equivalent to zero as leukocytes are filtered from the system. The input distribution l(t)BS during blood sample (20 ml) infusion (open circle) (time 0 to 60 s) is equivalent to the number of leukocytes in each 0.5 ml aliquot of donor blood. 74 FIGURE 4.3 Transit time distribution function (T(t)) 0.08 0 30 60 90 TIME (SECONDS) Fig. 4.3. The time derivative of the output distribution function O(t) for the sample data shown in Figure 4.2 is plotted here. This is the transit time distribution T(t)) for the data shown where l(t) • T(t) = O(t). The mean of the resultant transit time distribution is the mean leukocyte transit time for the heart and the system combined. The transit time for the system alone is derived using an identical analysis from separate experiments in which blood flow is identical but the isolated heart is excluded from the system. Since blood flow and system transit time are highly correlated (r2 = 0.98) the system transit time may be subtracted from the total transit time to yield the coronary transit time of leukocytes. 75 predicted by the flow rate. Therefore the flow dependent transit time of the perfusion apparatus was subtracted from the total transit time for each experimental measurement to yield mean coronary transit time of leukocytes. To calculate the individual coronary transit times of lymphocytes and PMN the mean values for the number of lymphocytes and PMN per aliquot from all experiments in each group were plotted versus time and mean lymphocyte and PMN coronary transit times were calculated for each group as described above. These values were then normalised to the mean total leukocyte coronary transit time of the control group (group 1, L-/H-) in order to compare the relative contribution of lymphocytes and PMN to the mean total coronary leukocyte transit time of the experimental groups (groups 2, L+/H-, 3, L-/H+ and 4, L+/H+). 4.234 Calculation of leukocyte retention The total number of leukocytes entering the heart was calculated as the total volume infused (20 ml for each heart) multiplied by the total leukocyte concentration of the sample (cells x 106/ml) to yield total number of leukocytes infused. The total number of leukocytes exiting the heart was calculated as the volume of all vials containing leukocytes (0.5 ml for each vial) multiplied by the total leukocyte concentration in each vial. The total number of leukocytes retained in the heart was calculated as the difference between the total number of leukocytes entering and exiting the heart. The fraction of total leukocytes retained was calculated as the total number retained divided by the total number entering the heart. 7 6 In order to determine whether or not the experimental apparatus was a site of significant leukocyte retention five separate experiments were performed in which normal saline at 37°C was perfused through the extra corporeal circuit and no isolated heart was attached to the aortic cannula. These experiments were otherwise identical to the other experiments. Utilising an analysis method identical to that described above there was no significant retention of leukocytes by the experimental apparatus. The numbers of lymphocytes and PMN in each blood aliquot was determined as the percentage of each cell type multiplied by the total number of leukocytes in that aliquot in order to determine the relative retention of lymphocytes and PMN. 4.235 Myocardial morphometric analysis In order to verify leukocyte retention by the coronary circulation, morphometric analysis of the fixed isolated hearts was performed as described in Chapter III (section 3.234, Myocardial morphometric analysis). A single tissue disc was selected at random from each heart. Tissue sections were stained with Hematoxylin and Eosin. Samples were examined at 100 x magnification only. Random fields were assessed by enumerating all leukocytes within the field. Ten random fields were assessed for each tissue section. The mean density of leukocytes was calculated as the number of leukocytes counted per tissue section divided by 10 fields. 4.24 Data Analysis 77 A two-way analysis of variance was used to test for differences among groups in leukocyte transit time and leukocyte retention. When a significant difference was found unpaired t-tests, corrected for multiple comparisons using a sequentially rejective Bonferroni test procedure, was used. A paired Student's t-test was used to compare leukocyte retention in all groups against zero retention. Values obtained for mean fluorescence intensity of the donor blood sample were compared between groups and with the resultant mean fluorescence intensity of the output samples using a two-way analysis of variance. P < 0.05 was chosen as statistically significant Data are expressed as means ± SD throughout. 4.3 Results Leukocyte activation data is summarised [Fig. 4.4]. There was no significant increase in MFI (versus pre-operative baseline) of leukocytes in blood from donor rabbits receiving vehicle only (groups 1, L-/H- and 3, L-/H+). This indicates that the collection and handling of the leukocyte containing blood and the injection of vehicle did not lead to activation of leukocytes. There was a significant increase of MFI of leukocytes in blood from blood donor rabbits receiving endotoxin (groups 2, L+/H- and 4, L+/H+) indicating activation of these leukocytes. The MFI of leukocytes exiting hearts whose blood donor and support rabbits had received vehicle only (group 1, L-/H-) was analysed to determine whether non-specific factors related to tubing, connectors, admixture of blood types, or the collection of samples lead to non-specific activation of leukocytes. There was no significant activation of leukocytes following system and coronary transit in group 1 (L-/H-). Analysis of non-preactivated leukocytes 78 FIGURE 4.4 CD18 relative mean fluorescence intensity UJ > UJ cc 2 -i 1 ^ 0 -1 -> • PRE-INFUSION POST-INFUSION 2 (L+/H-) 3 (L-/H+) 4 (L+/H+) GROUP Fig. 4.4. Relative change in CD18 mean fluorescence intensity (MFI) versus baseline (CD18 MR of donor rabbits prior to operation) prior to infusion through the isolated heart (open squares) and following infusion through the isolated heart (black squares). Values shown are the mean ± SD (n = 6/group). Infusion of endotoxin into donor rabbits (groups 2, L+/H- and 4, L+/H+) caused significant increase in MFI indicating activation of leukocytes. Transit of non-activated leukocytes through hearts with activated endothelium (group 3, L-/H+) caused significant increase in MFI. L, leukocytes activated (+) or non-activated (-), H, heart activated (+) or non-activated (-). * Indicates statistically significant (p < 0.05) versus baseline. ** indicates statistically significant (p < 0.05) versus pre-isolated heart value for same group. 79 exiting isolated hearts whose support rabbits had received endotoxin (group 3, L-/H+) showed a small but significant increase in MFI. There was no significant change in MFI of activated leukocytes transiting the coronary circulation of hearts whose support rabbits had received vehicle or endotoxin (groups 2, L+/H-and 4, L+/H+respectively). There was no significant difference in total leukocyte coronary transit time between groups 1 (L-/H-) and 2 (L+/H-) the non-activated coronary endothelium groups or between groups 3 (L-/H+) and 4 (L+/H+), the activated coronary endothelium'groups [Fig. 4.5]. There was a significant increase in total leukocyte coronary transit time in the activated coronary endothelium groups versus the non-activated coronary endothelium groups (p < 0.05) indicating that total leukocyte coronary transit time was independent of the state of activation of leukocytes entering the coronary circulation. There was no change in lymphocyte coronary transit time for any group relative to the total coronary leukocyte transit time of the control group (group 1, L-/H-) [Fig. 4.6]. There was a significant increase in the coronary transit time of PMN for group 3 (L-/H+) and group 4 (L+/H+) (p < 0.05) but not for group 2 (L+/H-) versus the total coronary leukocyte transit time of the control group (group 1, L-/H-) [Fig. 4.6] indicating that the observed increase in total leukocyte.coronary transit time of group 3 (L-/H+) and group 4 (L+/H+) versus control (group 1, L-/H-) was accounted for by the increase in PMN coronary transit time. There was no significant difference in total absolute [Table 4.1] or percentage [Fig. 4.7] leukocyte retention by the coronary circulation between 80 FIGURE 4.5 Coronary transit time of total leukocytes 30 n 20 10 0 -I- J I I - J II l M M I II F-g-s-s-s-i-i-s| 1 2 3 4 (L-/H-) (L+/H-) (L-/H+) (L+/H+) GROUP Fig. 4.5. Coronary transit time of total leukocytes. Values shown are the mean ± SD (n = 6 / group). The coronary transit time of leukocytes was significantly increased following activation of the coronary endothelium by endotoxin (groups 3, L-/H+, and 4, L+/H+). Activation of perfusing leukocytes alone (group 2, L+/H-) did not lead to a significant increase in coronary transit time versus control (group 1, L-/H-). L, leukocytes activated (+) or non-activated (-), H, heart activated (+) or non-activated (-). * Indicates statistically significant (p < 0.05) versus control (group 1, L-/H-). 81 FIGURE 4.6 Coronary transit time of PMN and lymphocytes 20 CO Q z O o LU CO 10 • PMN • LYMPHOCYTES (L-/H-) 2 (L+/H-) 3 (L-/H+) 4 (L+/H+) GROUP Fig. 4.6. Coronary transit time of PMN (black square) and lymphocytes (open square). Values shown are the mean ± SD (n = 6 / group). L, leukocytes activated (+) or non-activated (-), H, heart activated (+) or non-activated (-). * Indicates statistically significant (p < 0.05) versus baseline. 82 FIGURE 4.7 Percentage retention of total leukocytes 40 -, 30 UJ o cc UJ 20 10 0 (L-/H-) 2 (L+/H-) 3 (L-/H+) 4 (L+/H+) GROUP Fig. 4.7. Percentage of total leukocytes retained by the coronary microcirculation. Values are the mean ± SD (n=6/group). A significant percentage of total leukocytes perfused were retained by the heart following activation of the coronary endothelium by endotoxin (groups 3, L-/H+, and 4, L+/H+). Activation of the perfusing leukocytes alone did not lead to significant percentage retention of leukocytes (group 2, L+/H-). L, leukocytes activated (+) or non-activated (-), H, heart activated (+), or not (-). * indicates statistically significant (p < 0.05) versus control (group 1, L-/H-). 83 TABLE 4.1 Coronary leukocyte retention and coronary blood flow Group n Retention (cells x 106) Blood Flow (ml/min/g) Endothelium non-activated 1 (L-/H-) 6 TOT PMN 11.6 ±4.1 7.4 ±3.8 1.1 ±0.1 groups 2 (L+/H-) 6 TOT PMN 9.3 ± 4.6 5.5 ±2.4 1.2 ±0.2 Endothelium activated 3 (L-/H+) 6 TOT PMN * 25.3 ± 6.8 * 1.7.1 ± 5.1 1.1 ±0.2 groups 4 (L+/H+) 6 TOT PMN * 24.8 ±8.4 * 14.1 ±5.8 1.1 ±0.2 Table 4.1. Total and differential absolute coronary retention of leukocytes and mean coronary blood flow.are shown. Data are mean + SD, n = 6 animals per group. TOT, total leukocytes, PMN, polymorphonuclear neutrophil, LYM, lymphocyte,!, leukocytes activated (+) or non-activated (-), H, heart activated (+) or non-activated (-). * Indicates statistically significant p < 0.05 versus control (group 1). 84 groups 1 (L-/H-) and 2 (L+/H-), the non-activated coronary endothelium groups or between groups 3 (L-/H+) and 4 (L+/H+), the activated coronary endothelium groups. However, there was a significant increase in the total number of leukocytes retained by the coronary circulation when the coronary endothelium was activated (groups 3 (L-/H+) and 4 (L+/H+)) versus non-activated (groups 1 (L-/H-) and 2 (L+/H-)). There was no difference between groups 1 (L-/H-) and 3 (L-/H+) versus groups 2 (L+/H-) and 4 (L+/H+) indicating that the tendency of the coronary circulation to retain leukocytes was not affected by the state of activation of leukocytes entering the coronary circulation. There was no significant absolute [Table 4.1] or percentage [Fig. 4.8] retention of lymphocytes by any group. There was significant absolute [Table 4.1] and percentage [Fig. 4.8] retention of PMN by groups 3 (L-/H+) and 4 (L+/H+) versus groups 1 (L-/H-) and 2 (L+/H-) indicating that the increased total leukocyte retention observed in groups 3 (L-/H+) and 4 (L+/H+) was accounted for by PMN. There was no difference in total coronary blood flow between groups [Table 4.1]. There was no change in coronary perfusion pressure from the pre-set column pressure of 75 mmHg at any time during the experiment. Morphometric quantification of myocardial leukocytes [Table 4.2] showed no increase in density of leukocytes (leukocytes/high power field) retention in group 1 (L-/H-) versus group 2 (L+/H-) or in groups 3 (L-/H+) versus group 4 (L+/H+) (p < 0.05). However, groups 3 (L-/H+) and 4 (L+/H+), endothelium activated groups, had significantly increased myocardial leukocyte density versus the endothelium non-activated groups 1 (L-/H-) and 2 (L+/H-) consistent with the calculated increased total leukocyte retention. The majority of retained 85 FIGURE 4.8 Percentage retention of PMN and lymphocytes 80 -j 70 -1 2 3 4 (L-/H-) (L+/H-) (L-/H+) (L+/H+) GROUP Fig. 4.8. Percentage of total PMN (black squares) and total lymphocytes (open squares) retained by the coronary circulation. Values shown are the mean ± SD (n=6/group). A significant percentage of total PMN are retained following activation of the coronary endothelium whether the perfusing leukocytes are activated (group 4, L+/H+) or not (group 3; L-/H+). There was no significant percentage retention of PMN when the endothelium was non-activated (group 2, L+/H-). There was no significant retention of lymphocytes in any group versus control (group 1, L-/H-). L, leukocytes activated (+) or non-activated (-), H, heart activated (+) or non-activated (-). * indicates statistically significant (p < 0.05) versus control (group 1, L-/H+). 86 TABLE 4.2 . Total leukocytes per high power field (HPF) Group n Leukocytes/HPF 1 6 0.17 ±0.04 (L-/H-) Endothelium non-activated groups 2 6 • 0.47 ±0.11 (L+/H-) "• 3 6 * 1.00 ±0.38 (L-/H+) Endothelium activated groups 4 6 * 1.32 ±0.65 (L+/H+) Table 4.2. Total leukocytes per high power field (HPF, 100 x magnification) of left ventricular sections are shown. Data are mean ± SD, n = 6 animals per group. The majority of leukocytes counted were PMN. L, Leukocytes activated (+) or non-activated (-), H, activated (+) or non-activated (-). * Indicates statistically significant p < 0.05 versus control (group 1). 87 leukocytes were PMN. Leukocytes were retained primarily within coronary capillaries at this time point. 4.4 Discussion This study demonstrates that perfusion of the coronary circulation with blood from an endotoxin treated animal is more important than activation of leukocytes in causing the retention of leukocytes within the coronary circulation. PMN are primarily responsible for delayed leukocyte transit times and increased leukocyte retention in this model of sepsis and lymphocyte transit time is not significantly affected. Several mechanisms have been suggested by which PMN may become entrapped in the coronary circulation. These are briefly summarised as factors related to bulk circulatory dynamics, leukocyte specific factors and endothelium specific factors. A change in circulatory dynamics due to local mediators, such as nitric oxide (15,17) or TNF-a (30), or alterations in sympathetic neurotransmission (34) leading to a reduction in coronary perfusion pressure or blood flow may lead to increased retention of activated and stiffened leukocytes. Differences in total coronary blood flow or coronary perfusion pressure were not observed between groups. This suggests that the retention of leukocytes was independent of total coronary blood flow or perfusion pressure. However, it is possible that alterations in microcirculatory dynamics due to changes in red blood cell deformability (2), platelet aggregation (11), or other factors not affecting measures of bulk flow may have caused local capillary hypoperfusion with resultant leukocyte retention. Furthermore, areas of the coronary 88 microcirculation with reduced flow may permit settling of leukocytes on the endothelial surface and thereby facilitate bonding of adhesion molecules. A number of leukocyte related factors may contribute to leukocyte retention during sepsis. TNF-a produced by activated monocytes can suppress PMN chemotaxis, increase adhesion of PMN to gelatin, increase surface CD11b and decrease the fluid state of the PMN membrane (27). Decreased fluidity of leukocytes may impede or prevent capillary transit to post capillary venules leading to entrapment. This rheologic "stiffening" effect of activation of leukocytes on leukocyte retention in the lung has been well described (12). Various models of leukocyte retention by the coronary circulation- have suggested that leukocyte activation and stiffening may lead to retention (19). However, in the present study, no difference in retention of leukocytes was observed between hearts infused with leukocytes which were activated (group 2, L+/H-) or non-activated (group 1, L-/H-). This suggests that Theologically mediated events were relatively insignificant to retention in this model. Receptor mediated adhesion of leukocytes to vascular endothelial cells has been demonstrated (6). However, leukocyte CD18 upregulation alone (group 2, L+/H-), in this model did not lead to increased leukocyte retention. These results suggest that activation of the coronary endothelium is an important cause of leukocyte retention. Evidence suggests that specific receptor mediated adhesion does contribute to increased leukocyte retention by the coronary circulation during whole animal sepsis. The present study supports the hypothesis that adhesion receptor ligand upregulation on the surface of coronary endothelial cells coexists with the pathologic production of factors capable of modifying leukocytes within the environment of the coronary 89 circulation itself. A number of endothelial mediated factors may contribute to the retention of leukocytes. During sepsis endotoxin activates endothelial cells via both direct (via soluble CD14-endotoxin complexes) and indirect (via monocyte bound CD14-endotoxin complexes) pathways (25), the indirect pathway being quantitatively more significant. Monocytes bound to endotoxin via membrane associated CD14 receptors are stimulated to secrete TNF-a (10). TNF-a may interact with endothelial cell TNF-a receptors leading to the production of IL-1 (23). IL-1 increases surface expression of endothelial cell adhesion molecules (5) and induces the secretion of inflammatory cytokines such as IL-6 (5). Endothelial ceil adhesion molecule expression may be complex and require a progression of pathologic stimuli to manifest a pro-adhesive surface (29). Increased production of tissue factor by activated endothelial cells leads to the local deposition of thrombin (26) which may further disrupt local microcirculatory dynamics leading to leukocyte retention. Local thrombin deposition may lead to increased production of PAF by endothelial cells (31) which may prime PMN responses to activating stimuli. Analysis of the leukocyte effluent from experiments in which only the heart was activated (group 3, L-/H+) demonstrated an increase in CDi8 MFI after transiting the coronary circulation consistent with activation of leukocytes within the heart itself. This and other experimental evidence supports the speculation that the evolution of the sepsis syndrome involves an early cytohumoral cascade which causes alterations in the function of various organs (3,21) with the affected tissue becoming the source of pathologic mediators capable of mediating local damage or modifying local immune responses. This may serve to localise and enhance leukocyte mediated inflammatory responses at the site of tissue inflammation. Indeed, juxtaposition of activated leukocytes 90 and activated endothelial cells may be essential for leukocyte mediated endothelial cell damage (22). Thus, myocardial function, manifesting at the organ level, declines in proportion to the volume of dysfunctional or devitalised tissue. It seems likely from this experiment that a combination of both local or distal activation of leukocytes and activation of coronary endothelial cells is required for leukocyte retention during sepsis. Furthermore it seems likely that a number of independent mechanisms may lead to increased transit time and retention of leukocytes in whole animal models of septic shock. Specific indices of endothelial activation were not measured in this experiment as the purpose of the experiment was to identify the relative effects of heart and leukocyte factors on coronary leukocyte transit time and retention. However, evidence from this study reasonably supports the assumption that pathologic changes related to sepsis had occurred in the isolated hearts whose support rabbits had been exposed to endotoxin. The early cytohumoral cascade of sepsis is capable of inducing the local production of chemokines by the "septic" myocardium (20). Local chemokine production may mediate subsequent septic changes in the myocardium or reinforce inflammatory localisation to areas of damaged myocardium. Upregulation of endothelial receptors mediates or strengthens leukocyte-endothelial cell adhesion (28). Abnormal endothelial cell membrane transport function occurring during sepsis (14) may lead to endothelial cell swelling. Endothelial cell swelling (16) in association with intracapillary fibrin deposition (26) may cause or contribute to the observed increased heterogeneity of microvascular blood flow (13,32) during sepsis. This may lead to passive entrapment of leukocytes in non-flowing microvascular segments. 91 This study utilised established methods for determining the transit time of an indicator (leukocytes) in a single inlet-single outlet system (18). This model is particularly appropriate as it permits recovery and quantitation of the entire injectate sample thus minimising sampling error. The transit time values obtained in the control group (group 1, L-/H-) are similar to reported values for a whole animal model of sepsis (7) supporting this model as an accurate simulation of whole animal sepsis. This model furthermore, offers certain advantages over existing techniques. Techniques utilising forced cellular washout and morphometric analysis may yield differences between various leukocyte fractions with respect to relative retention but are unable to generate differential leukocyte transit times. Morphometric techniques exist for the calculation of transit time in the whole animal but are time consuming, require separate analysis of red blood cell flow and blood volume and are subject to time varying changes in total coronary blood flow and blood volume. Methods utilising radiolabeling of leukocytes cause the activation of leukocytes (24) preventing distinction between transit time of activated and non-activated leukocytes. Furthermore, radiolabeling methods directly study only a very small proportion of the total number of leukocytes transiting the circulation in whole animal models. 4.5 Summary In summary, this study demonstrates that, during sepsis, a crucial factor in the retention of leukocytes by the coronary circulation is the activation of the coronary endothelium by products of the cytohumoral cascade. Retention of leukocytes is relatively independent of coronary blood flow and perfusion pressure. Modification of transiting leukocytes by factors present within 92 endotoxin activated hearts may lead to coronary leukocyte retention. The timing of leukocyte retention by the coronary circulation, during sepsis, is consistent with an etiologic role of leukocytes in the reduced myocardial function observed in whole animal models of sepsis. 93 CHAPTER V EXCLUSION OF ACTIVATED LEUKOCYTES FROM THE MYOCARDIAL CIRCULATION DURING SEPSIS AMELIORATES MYOCARDIAL STRUCTURAL CHANGES AND CONTRACTILE FAILURE 5.1 Hypothesis In the current study it was asked whether leukocytes cause damage and contractile dysfunction within the first 6 h in ah animal preparation in which changes in the contractility of two isolated rabbit hearts were measured directly and simultaneously while the hearts were perfused with blood from a single endotoxemic rabbit. One isolated heart was perfused with blood that had leukocytes removed by filtration, and the second was perfused with unfiltered blood. The use with this isolated heart preparation of an index of contractility that incorporates afterload (18), eliminates the effects on measures of ventricular contractility of the changes in heart rate (26), coronary perfusion pressure (33), preload, and afterload (18) that occur during septic shock, and which confound measures of ventricular contractility in vivo. Both hearts were perfused with blood from the same endotoxemic support rabbit, modelling the complex plasma milieu of sepsis. 5.2 Experimental design 5.21 Instrumentation. 5.211 Surgical Preparation of the Support Rabbit 94 Twelve female New Zealand White rabbits were surgically prepared as described in Chapter III, section 3.211. Because of the longer (6 h) protocol supplementary anesthesia in the form of intermittent injections of a-chloralose was required. There was no significant difference in the total volume of a-chloralose administered to the support rabbits. 5.212 Hemodynamic and pH support of the support rabbit. Blood pressure monitoring and hourly arterial blood sampling was done via a left femoral arterial polyethylene catheter. Heparinized (150 Ill/ml) crystalloid solution (Plasma-Lyte; Baxter Corp., Toronto, ON) was infused at 10 ml/kg/h via a polyethylene catheter in the left femoral vein. Sodium bicarbonate (7.5%) was infused periodically to maintain arterial pH above 7.35. During the experimental period, arterial bicarbonate averaged 17.9 ± 1.3 mmoi/l (mean ± SEM), and arterial pH averaged 7.40 ± 0.03. Intermittent boluses of crystalloid were infused to maintain the mean arterial blood pressure above 60 mmHg. 5.213 Surgical Preparation of the Isolated Hearts The surgical preparation of the isolated hearts was identical to that described in Chapter III, section 3.213 (Surgical preparation of the isolated heart). Left ventricular balloons were inserted in both isolated hearts in identical fashion. Both hearts were paced simultaneously at 150 beats/min. The ventricular balloons were inflated with normal saline until a left ventricular end-diastolic pressure of 5 mmHg was achieved. One heart was randomly assigned to receive leukocyte-filtered blood and the second heart received unfiltered blood. 95 5.214 Langendorff column and extracorporeal circuit. The preparation of the Langendorff column and extracorporeal perfusion circuit was identical to that described in Chapter III, section 3.214 (Langendorff column and extracorporeal circuit). In this experiment two identical perfusion circuits were utilised simultaneously [Fig. 5.1]. 5.22 Experimental Protocol The preparation was allowed to stabilise for 60 min, after which blood perfusing one of the hearts was diverted through a leukocyte filter (Sepacell 4C2468; Baxter-Fenwal Division, Mississauga, ON). This filter is packed with 1.8-u.m polyester fibres coated with a copolymer of hydroxylethyl methacrylate and diethyl-aminoethyl-methacrylate, and has a volume of 14 ml. Platelet function is not altered during filtration (2). Filtration has no effect on blood pH, potassium, sodium, or calcium, concentration (2). The filter had been previously primed with 20 ml of heparinized arterial blood from the support rabbit, obtained during the stabilisation period. Endotoxin (Escherichia coli 0111 :B4; Sigma Chemical Co., St. Louis, MO) 1 mg/kg was then administered intravenously to the support rabbit over a period of 30 min. Measurements were continued for a period of 6 h after the endotoxin infusion had started. At the conclusion of the experiment, the blood perfusion circuit was interrupted and normal saline was infused into the isolated hearts to flush out red blood cells from the myocardial circulation (approximately 5 min). Glutaraldehyde 2.5% in phosphate buffer was then added to the saline perfusion circuit and the hearts were perfusion fixed for 10 min. The isolated 96 FIGURE 5.1 Dual heart perfusion apparatus ENDOTOXIN " " J " (1000ug/kg I.V.) SALINE 75 mmHg SUPPORT RABBIT Fig 5.1. Apparatus shown is identical to that depicted in Fig. 3.1 but modified so that two isolated hearts (A and B) are perfused simultaneously by the same support rabbit. Note the addition of a leukocyte specific filter to the arterial inflow line of one of the isolated hearts (B). A single syringe pump not depicted here (Fig. 3.1) simultaneously controls volume inflow to the intraventricular balloons and thus the ESPVR (Fig. 5.2). 97 hearts were then removed from the apparatus and transferred to a container of the same fixative. ,5.23 Measurements. 5.231 Left ventricular contractility Contractility for each heart was measured at 30-min intervals, using the slope of the end-systolic pressure-volume relationship (Em a x) and also to corroborate these findings, the maximum rate of change of intraventricular pressure (dP/dtmax). E m a x is a.very specific measure of changes in contractility, largely independent of the effects of preload and afterload (18). To calculate Emax the intraventricular balloons were filled with saline at a constant rate over 1 min from a volume of 200 to 600 uJ, using a syringe pump. A constant balloon inflation rate allowed time measurements to be converted directly to intraventricular volume measurements. The maximum balloon inflation volume was much less than the balloon's unstressed volume. During inflation of the balloon, left ventricular pressure was sampled at 100 Hz. The slope of the line that best fit the end-systolic pressure-volume points generated for each isolated heart was calculated through linear regression. In all cases the end-systolic pressure-volume relationships were quite linear over the range of measurement (r2>0.96) [Fig. 5.2]. The maximum slope of the pressure-volume curves for three to five consecutive contractions was then determined for each isolated heart and recorded. For calculation of dP/dtmax, a sensitive measure of changes in contractility (18), intraventricular pressure was measured at 100 Hz at an end-diastolic balloon volume of 200 ul. The Lagrange five-point formula was then used to determine dP/dtm a x from these data. 98 FIGURE 5.2 Ventricular pressure plotted versus ventricular volume cn m E rr CO CO LU DC Q_ 70 - , 60 — 50 - \ 40 30 -A 20 - \ 10 -A ISOLATED HEART A ISOLATED HEART B 200 250 VOLUME A (ul) 300 200 250 VOLUME B (Ml) 300 Fig. 5.2. Ventricular pressure plotted versus ventricular volume during a typical balloon inflation in an isolated heart. The best-fit line (r2 = 0.99) drawn through end-systolic points is shown. The. slope of this line, Emax> is a very specific measure of changes in ventricular contractility, and is largely independent of changes in preload and afterload (18). Data shown is identical to that shown in Chapter III, Fig. 3.2. 99 5.232 Other measurements Arterial Po2, Pco2, pH, and bicarbonate were measured at 30-min intervals with a blood gas analyser (ABL30; Radiometer, Copenhagen, Denmark). Arterial lactate concentration (YSI 2300 Glucose-Lactate analyser; YSI Instruments, Yellow Springs, UT; normal range less than 1.8 mmol/l) hemoglobin, white blood cell (WBC) count, and platelet count (Coulter Electronics, Inc., Hialeah, FL) were measured at 60 minute intervals. 5.233 Effects of the Leukocyte Filter The isolated-supported rabbit heart system maintains stable function for the observation period used in this study (10). However, it was important to exclude the possibility that the leukocyte filter alone had effects on contractility, and that following endotoxin infusion, the leukocyte filter did not bind endotoxin and therefore protect the downstream heart from endotoxin effects. In order to exclude the possibility that factors related to the leukocyte filter could lead to altered myocardial function, four additional experiments were performed in which a single isolated heart was perfused in identical fashion to that described previously, but with non-endotoxemic perfusing arterial blood passed through a leukocyte filter. There was no significant change in E m a x from baseline (14.1 ± 1.6 mmHg/100 ul) to 6h (1.43 + 3.5 mmHg/100 ul) following saline infusion. In order to exclude the possibility that the leukocyte filter bound endotoxin, four additional experiments were performed as described previously except that the isolated hearts were excluded. Prior to infusion of endotoxin, simultaneous blood samples (2.0 ml) were drawn from sampling ports located 100 at the inlet and cutlet of the leukocyte filter, Endotoxin was then infused into the support rabbit as described earlier. Blood sampling was continued in identical fashion at 30 min and 60 min from the start of endotoxin infusion. Sampling was then continued at 2-h intervals until the conclusion of the experiment. Blood samples were collected in sterile, pyrogen-free centrifuge tubes (2.5 ml total volume) and anticoagulated with heparin (1,000 lU/ml solution) .50 u,l/ml (Organon Teknika). Samples were immediately centrifuged for 5 min at 1,600 rpm. The plasma supernatant was then pipetted, using sterile technique, and the plasma sample was transferred to sterile, pyrogen-free centrifuge tubes. Plasma samples were immediately frozen in liquid nitrogen and maintained at -70°C until being assayed for endotoxin. Plasma samples were assayed for endotoxin with Limulus amebocyte lysate (LAL) assay system (E-Toxate; Sigma) as previously described (12). In order to remove LAL inhibitors, plasma samples were diluted 1:10 in endotoxin-free water and heated for 5 min at 37°C (12). Semiqua.ntification of endotoxin concentration was obtained by performing multiple serial dilutions of each LAL inhibitor-free sample, and the concentrations are reported as endotoxin units/ml (EU/ml). There was no detectable endotoxin in plasma samples prior to the infusion of endotoxin (<0.6 EU/ml). Peak endotoxin levels in all cases were obtained at 30 min after the start of endotoxin infusion. There was no significant difference in endotoxin concentration in simultaneous plasma samples obtained before and after passage through the leukocyte filter [Fig. 5.3], and blood flow was not different before and after passage through the leukocyte filter. 101 FIGURE 5.3 Endotoxin concentrations in blood samples pre and post-filter 8 CO o T— X UJ-z o h-< DC LU O z . o o z X o f— O Q z LU 6 A 4 A 2 A 0 INFUSION 0 0.5 1 PRE-FILTER POST-FILTER - i : r 2 3 TIME (HOURS) —r-4 Fig. 5.3. Endotoxin concentrations in blood samples (n = 4 per point) taken before and after leukocyte filtration at time points after endotoxin infusion. There is no significant difference in concentrations before and after filtration. 102 5.234 Determination of L-selectin Expression on PMN, In order to determine whether PMN were activated and to verify the CD18-based measurements of PMN activation described in Chapter II, section 2.238 and Chapter IV, section 4.231 a different measure of PMN activation was utilised and this is described below. During activation, L-selectin is shed from PMN, so that a decrease in immunofluorescence staining for L-selectin is an indicator of PMN activation (37). For these measurements, five separate sets of experiments were performed with the support animal preparation only. Endotoxin administration and leukocyte filtration occurred at identical time points, as described previously. Blood samples for immunofluorescence staining were collected from arterial blood entering the filter at baseline and at 1, 3 and 5 h after endotoxin administration. A whole-blood method for preparing specimens was used. The cells were prepared from ethylene diamine tetraacetic acid (EDTA)-treated blood, using a commercially available kit (Coulter Clone; Coulter Electronics). For each sample, 200 ul of EDTA-treated blood was incubated first with 200 uJ of phosphate-buffered saline (PBS) and 20 uL of DREG 200 antibody (200 Lig/ml; courtesy Dr. Butcher,) for 10 min at room temperature. For each sample, nonimmune mouse IgG was used as a negative control (1-5381; Sigma). The cells were then washed and centrifuged at 1,600 rpm for 5 min and the supernatant was discarded. The cells were then incubated with 20 u.l of fluorescein isothiocyanate (FITC)-conjugated, whole-molecule anti mouse IgG (F2012; Sigma) for 10 min at room temperature in the dark. The cells were 103 washed and centrifuged at 1,600 rpm for 5 min, and the supernatant was discarded. The red blood cells were then lysed (Immuno-lyse Coulter Clone; Coulter Electronics), washed, and centrifuged at 1,600 rpm for 5 min. The remaining leukocytes were fixed with 1% paraformaldehyde and stored at 4°C. Flow cytometry was performed on the specimens within 24 h (Profile EPIC II; Coulter Electronics). Analysis gates for the PMN subset of leukocytes were established with the distinctive forward- and side-scatter profiles. Owing to the small concentration of neutrophils in filtered blood, a total of 1,000 gated cells were evaluated per specimen and the result expressed as MFI. 5.235 Myocardial morphometric analysis Morphometric analysis of the fixed isolated hearts was performed as described in Chapter III (section 3.234, Myocardial morphometric analysis). Tissue sections (3-u.m thick) were stained with hematoxylin and eosin (H&E). The number of random fields needed to count 100 leukocytes in each group at 400 x magnification was recorded. The average number of leukocytes per 400 x field was then determined as a quantitative assessment of myocardial leukocyte content. The fractional area of interstitial edema and fractional area of structurally abnormal myocytes (defined in Chapter III, section 3.234 as myocyte swelling and hypochromasia, cytoplasmic vacuolation, zonal contraction banding and, contraction band necrosis) were measured as described in Chapter III, section 3.234 using digitally captured images analysed using Bioview software (Infrascan, Richmond, B.C., Canada). The use of digital image analysis increased the speed at which morphometric analysis was performed but did not change the technique from that previously described (Chapter III, section 3.234). 104 5.24 Data Analysis The principal null hypothesis that there was no difference in contractility between the hearts perfused with leukocyte-filtered and leukocyte-unfiltered blood following the administration of endotoxin was tested with a two-way analysis of variance (ANOVA), with one way being repeated measures over time. When a significant difference was found, specific differences within groups were identified through paired t-tests, and specific differences between groups were identified with unpaired t-tests. Corrections for multiple comparisons were made with a sequentially rejective Bonferroni test procedure. All data are expressed as mean ± SEM. A value of p < 0.05 was regarded as statistically significant. 5.3 Results There was no difference in E m a x between groups at baseline [Fig. 5.4]. In the hearts perfused with unfiltered blood, E m a x decreased to 81 + 6% of baseline at 6 h after the start of endotoxin infusion (p<0.05). In the hearts perfused with filtered blood there was no change in E m a x from baseline (103 + 4% of baseline value at 6 h, p = NS). The change in E m a x was significantly different between groups (p < 0.05, ANOVA). Similarly, there was no difference in dP/6Xmax between groups at baseline, but after the infusion of endotoxin, dP/dtmax in hearts perfused with unfiltered blood decreased to 74 ± 9% of baseline (p < 0.05). In hearts perfused with leukocyte-filtered blood, dP/dtmax did not change significantly. The change in dP/dtmax was significantly different between groups (p < 0.05 < ANOVA). 105 FIGURE 5.4 Effects of leukocyte depletion on myocardial contractility 20 -, 10 0 4 X 03 E LU -10 4 -20 A -30 FILTERED UNFILTERED INFUSION - r -2 - r — 3 4 5 0 0.5 1 TIME (HOURS) Fig. 5.4. Effects of leukocyte depletion on myocardial contractility, as measured with Emax. following endotoxin administration, in hearts perfused with leukocyte-filtered and unfiltered blood (n = 12). Following the administration of endotoxin, E m a x was not significantly decreased in hearts perfused with leukocyte-filtered blood. There was a significant decrease in E m a x in hearts perfused with unfiltered blood (* p < 0.05) compared with hearts perfused with leukocyte-filtered blood, ANOVA. 106 FIGURE 5.5 Effects of endotoxemia and leukocyte filtration on circulating neutrophil count 10 -, o Z 3 O o -J X a. O cc H 3 UJ 0.1 J 0.01 0.001 FILTERED UNFILTERED INFUSION •1 0 1 2 3 4 TIME (HOURS) Fig. 5.5. The effects of endotoxemia and leukocyte filtration on the circulating neutrophil count. Results are shown as the absolute neutrophil count for baseline and 2 and 4 h after endotoxin infusion. Filtered coronary perfusate had significantly fewer neutrophils than unfiltered coronary perfusate (p < 0.05). After the administration of endotoxin, there was a significant decrease in the absolute neutrophil count in the support rabbit's circulation. * p < 0.05 compared with baseline. 107 FIGURE 5.61 Myocardial morphometric analysis: Myocardial neutrophil retention 1.2 n | 1.0 -LL Fig. 5.61. Myocardial neutrophil counts per 400x field for hearts perfused with unfiltered and filtered blood. Filtering resulted in substantially fewer myocardial neutrophils (f p < 0.01) present within myocardial capillaries by 6h after the infusion of endotoxin into the support rabbits. 108 FIGURE 5.62 Myocardial morphometric analysis: Fractional area of interstitial edema and myocardial myocyte damage 30 -. o t-o < DC U_ < L U DC < UNFILTERED • FILTERED 20 H 10 H 0.0 INTERSTITIAL EDEMA MYOCYTE DAMAGE Fig. 5.62. Fractional area of myocardial interstitial edema, and fractional area of myocyte necrosis for hearts perfused with unfiltered and filtered blood. Filtering resulted in less interstitial edema (* p <0.05), and a trend toward less myocardial damage (zonal contraction bands, contraction band necrosis, and vacuolated cytoplasm) (* p < 0.05) by 6h after the infusion of endotoxin into the support rabbits. 109 FIGURE 5.7 Expression of L-selectin on neutrophils 12.5 n 10 7.5 A 5 A 2.5 0 INFUSION L-SELECTIN EXPRESSION ON PMN ENTERING LEUKOCYTE FILTER —r-4 ~i 5 0 0.5 1 2 3 TIME (HOURS) Fig. 5.7. Expression of L-selectin on neutrophils in arterial blood entering the filter at baseline and 1, 3, and 5 h after endotoxin administration. For each time period 1,000 neutrophils were evaluated. Results are expressed as mean fluorescence intensity. Neutrophil expression of L-selectin significantly decreased from baseline * p < 0.01. 1.10 There was a significant difference in the cellular component of the coronary perfusate of each heart. The leukocyte count in arterial blood was 5.7 ± 0.6 x 109/l at baseline. Following filtration and after administration of endotoxin, there was a significant decrease in leukocyte count in filtered blood to 0.50 ± 0.05 x 109/l. Leukocyte differential counts demonstrated that absolute neutrophil counts remained decreased (p < 0.05) in filtered as compared with unfiltered blood for the duration of the experiment [Fig. 5.5]. Consistent with these results were the findings that at the end of the experiment, the predominant subset of leukocytes in the myocardium consisted of neutrophils, and that the number of neutrophils in the myocardium was reduced by 77% (p < 0.05) in the hearts perfused with filtered as compared with the hearts perfused with unfiltered blood [Fig. 5.61]. In addition, morphometric measurements showed that the area fraction of interstitium was reduced by 49% (p < 0.05) and that the area fraction of myocardial damage was reduced by 40% (p>< 0.13) in the hearts perfused with filtered blood as compared with the hearts perfused with unfiltered blood [Fig. 5.Q2). After the administration of endotoxin, there was a significant decrease in the expression of L-selectin on neutrophils from support rabbit arterial blood [Fig. 5.7]. Mean fluorescence intensity decreased by 55 ± 18% (p < 0.01) from baseline to after endotoxin infusion, and remained at this low level for the duration of the experiment. Infusion of endotoxin resulted in a septic response in the support rabbit, including a significant increase in arterial lactate concentration from 2.1 ± 0.4 mmol/l at baseline to 13.4 ± 1.0 mmol/l at 6 h (p < 0.01), and a decrease in 111 blood pressure from 114 ±4 mm Hg to 79 ± 5 mm Hg (p < 0.01) over the same period. Over the period of measurement, arterial hemoglobin decreased from 84 ± 6 g/L to 63 ± 2 g/L (p < 0.01), and the arterial platelet count decreased from 177 ± 27 X 109/l to 45 ± 26 X 109/l (p < 0.01). There was no difference between the coronary hemoglobin or platelet concentration in filtered and unfiltered blood. 5.4 Discussion In this model of endotoxin-mediated septic shock, the reduction of leukocytes perfusing the coronary circulation as a consequence of filtration preserved systolic contractility of the isolated heart. However, contractility decreased significantly in the heart perfused with leukocyte-containing blood. Both hearts were exposed to the same metabolic milieu of a single donor rabbit. Left ventricular preload, afterload, coronary perfusion pressure, and heart rate were constant and equal in both isolated hearts throughput the study. Thus, the observed differences in. contractile function between the hearts perfused with filtered and unfiltered blood cannot be explained by hemodynamic changes in the support rabbit or by noncellular changes in the coronary perfusate. These differences in contractility were found with both E m a x , a very specific end-systolic index of left ventricular contractility (18), and (dP/dtmax), a very sensitive early-contraction-phase index (18). The predominant subset of leukocytes found within the isolated heart perfused with unfiltered blood consisted of PMN which, in circulating bipod, were activated. PMN accumulation in the heart was associated with increased myocardial interstitial edema and myocyte necrosis. These observations suggest that leukocytes may be important in causing the early decrease in left ventricular contractility during sepsis. 112 The decrease in contractility observed in the current study is similar in magnitude to the decrease observed in previous animal models, in humans following endotoxin administration, and in human septic shock. A 17 ± 6% decrease in E m a x was observed which is similar to that observed in our earlier study of myocardial contractile dysfunction of a single perfused heart following endotoxin infusion (Chapter III). E m a x is approximately linearly related to changes in ejection fraction when preload, afterload, and. heart rate are kept constant (36), as in the present study. Thus, the decrease in contractility observed in the study is reasonably comparable to an 11% decrease in ejection, fraction observed during human septic shock (29) and a 23% decrease in the ratio of peak systolic pressure to end-systolic volume observed 5 h after endotoxin administration in humans (32). An isolated heart preparation was used to measure changes in contractility because in vivo studies of myocardial contractile function are confounded by changes in heart rate (18), coronary perfusion pressure (26), preload, and afterload (33) that occur during septic shock. For example, the addition in vivo of the inhibitors of the inflammatory-mediator pathway may directly alter contractility, but will also produce secondary effects on coronary perfusion pressure, preload, and afterload, making an interpretation of the inhibitor's primary effect complex. An important limitation of conventional crystalloid-perfused isolated heart preparations is that they remain functionally stable for only 1 to 3 h, which is less than the 4 to 5 h before the earliest decrease occurs in myocardial function following endotoxin administration in vivo (36, 28). Therefore, isolated hearts in our study were perfused with blood from a support rabbit, which results in a stable and constant contractile state in control rabbits for more than 8 h (10), 113 It is difficult to extrapolate from previous in vitro studies of myocytes with myocardial contractile dysfunction to whole-animal models or the human condition of septic shock, because normally perfused and functioning myocytes in vivo may respond differently and are exposed to conditions more complex than those reported in in-vitro myocyte studies. For example, isolated myocyte studies demonstrate that TNF-a and IL-1 reversibly (within 20 min) decrease myocyte contractility via NO production (8). In contrast, studies in vivo report that TNF-a does not decrease contractility via NO (4). Other studies demonstrate that the time course of sepsis-induced contractile dysfunction in animal models (36) and in humans (29) occurs over hours and days, and does not correlate with the brief increase in serum TNF-a. To overcome the problems of in vivo and in vitro studies, a preparation was developed in which changes in contractility of two isolated rabbit hearts were measured directly and simultaneously over a 6 h period while both hearts were perfused with blood from a single endotoxemic rabbit. Thus the complex features of septic shock are reproduced in vivo and contractility is measured in vitro, where confounding influences on contractile function are removed. Leukocytes, and particularly neutrophils, are retained in the heart early in the systemic inflammatory response (1, 9) secondary to rheologic changes or via endothelial adhesion. As previously noted, Barroso-Aranda (1) observed an increase in the number of intramyocardial capillaries obstructed by leukocytes in rats following endotoxin infusion. In this regard, it has previously been demonstrated that large numbers of leukocytes are retained in the coronary capillary bed and that leukocyte transit time across the coronary circulation is increased approximately 10-fold during endotoxemia in pigs (9). The increase 114 in intramyocardial leukocytes following endotoxin administration is associated with a decrease in E m a x (10). Therefore, in septic shock, activated leukocytes are retained in the myocardium. Potentially, they could then contribute to impaired myocardial contractility. The new observation that the leukocyte filter used in the present study prevented early myocardial contractile dysfunction during sepsis is consistent with the findings in studies evaluating other organ-system dysfunction (lung, gut) during septic shock, and in studies of cardiac ischemia-reperfusion injury. Activated neutrophils produce tissue damage through both oxidative and nonoxidative . enzymatic mechanisms (31), and monocytes/macrophages elaborate multiple mediators, including cytokines implicated in contractile dysfunction (8). During septic shock activated neutrophils lead to an increase in vascular permeability and are involved in the development of organ dysfunction (25, 34). Anti-CD18-specific antibodies have improved survival in dogs challenged with TNF-a, and attenuate both neutropenia and alveolar-capillary membrane injury during septic shock (6, 17). Leukocytes also mediate tissue damage and dysfunction in ischemia-reperfusion injury (30, 22). In dogs, removal of leukocytes from the coronary circulation by filtration reduced myocardial ischemia-reperfusion injury following a period of coronary occlusion (30). The blockade of coronary endothelial adhesion by neutrophils produced by administering antibodies to either CD11b, CD18, L-selectin, or intercellular adhesion molecule-1 (ICAM-t) has been shown to attenuate myocardial ischemia-reperfusion injury and dysfunction (22, 23) suggesting that neutrophil adhesion (38) is required to mediate injury. Therefore, it seemed reasonable to 115 postulate that leukocytes would also be involved in septic myocardial dysfunction. It was found that the PMN subset of leukocytes is the most prominent within the myocardium in this model of sepsis. Expression of L-selectin on circulating neutrophils decreased during our study. During leukocyte activation, L-selectin is shed from the cell surface, making it a useful indicator of neutrophil activation (10). L-selectin is important for the initial adherence and rolling of circulating leukocytes to activated endothelial surfaces (20). A second explanation for the decrease in L-selectin expression in our study is that the filter that used may have sequestered cells expressing high levels of L-selectin. Interestingly, as the study progressed, there was a slight increase in L-selectin expression on circulating neutrophils [Fig. 5.7]. This possibly represents bone marrow release, since newly released neutrophils express higher levels of L-selectin than older circulating cells (21). The platelet count also decreased dramatically during endotoxemia in this study. Platelet aggregation and adhesion are known to lead to coronary occlusion and secondary myocardial dysfunction. However, there was no difference in the platelet count in filtered and unfiltered blood. In addition, coronary blood flow is increased in septic shock (5), and therefore global hypoperfusion does not appear to be responsible for changes in contractility during septic shock. Thus, the contribution of platelet aggregation and plugging of the coronary circulation to the observed differences between the hearts in the present study is likely to be small. However, these results do not exclude a possible role for platelets in sepsis-induced myocardial dysfunction. Previous studies have shown that inhibiting various mediators of inflammation, including TNF-a (13), platelet activating factor (PAF) (15), and NO 116 (14) reduces myocardial dysfunction in models of sepsis. One explanation for this effect is that these substances act as circulating myocardial depressant factors. In vitro studies support the. notion that circulating factors in plasma decrease myocardial contractility (19, 27,11, 8). However, the dose-dependent and reversible myocardial depression of TNF-a and other cytokines via NO generation in vitro (8) is not completely consistent with findings in in-vivo studies (4) and with the observation that the time course of ventricular dysfunction during sepsis (2) is not correlated with the time course of expression of TNF-a and other cytokines. These findings, which show that the leukocyte filter used in our study prevented the decrease in ventricular contractility commonly observed during sepsis, do not contradict these previous studies supporting the concept of myocardial depressant factors (3, 8). Instead, this new finding may explain some of the controversy surrounding this topic, and suggests an additional mechanism of action of circulating inflammatory mediators. That is, TNF-a and other cytokines expressed early in sepsis may mediate part of the decrease in myocardial contractility through one or more of their multiple downstream effects. It is suggested that downstream effects of TNF-a and other inflammatory mediators include the activation of myocardial cells and endothelium to generate chemotactic cytokines (24) and express adhesion molecules (38), and the activation of leukocytes, which become stiff and increase their surface expression of adhesion molecules (16). As a result, TNF-a and other cytokines result in increased accumulation of activated leukocytes in the heart (9). Conceivably, intramyocardial leukocytes could produce high local concentrations of cytokines, reactive oxygen intermediates, or nonoxidative damaging enzymes, or lead to the release of NO all of which may damage myocardial cells and decrease myocardial contractility. Myocardial damage could conceivably contribute to prolonged periods of 117 decreased myocardial contractility even after TNF-a (and other mediator) levels decrease, and possibly after increased numbers of leukocytes are cleared from the heart. The results of the current study strongly suggest a role for leukocytes in the myocardial dysfunction of sepsis. It is conceivable that the leukocyte filter used in the study prevented a decrease in contractility in other ways. The possibility that the leukocyte filter removes endotoxin was examined and no significant difference was found. However, it is conceivable that it removes other, unmeasured inflammatory mediators. When leukocyte concentrations are increased in several sepsis models by the administration of granulocyte colony-stimulating factor (G-CSF), survival and some measures of cardiac function are improved (7). In contrast, the administration of granulocyte-macrophage colony-stimulating factor (GM-CSF) in sepsis worsens survival. These results may be explained by the associated decrease in endotoxin and TNF-a concentrations with G-CSF (7) and the increase in their concentrations with GM-CSF (35). Whatever the mechanism of effect of G-CSF and GM-CSF, these results illustrate that the role of leukocytes in sepsis-induced myocardial dysfunction is not fully understood. The current experiment describes changes in contractility over a period of 6 h. The relationship of these early results to the changes in cardiac function over days during sepsis in humans (29) is also uncertain. 118 5.5 Summary In summary, activated leukocytes are retained in the coronary circulation of rabbits treated with endotoxin, and are associated with increased myocardial interstitial edema and myocardial damage. When perfusing blood is passed through a leukocyte filter, these effects are diminished and decreased myocardial contractility is prevented after endotoxin infusion. 119 CHAPTER VI SUMMARY OF MAJOR CONCLUSIONS AND RECOMMENDATIONS FOR FURTHER STUDY 6.1 Summary of major conclusions In Chapter II it was demonstrated that following activation of the cytohumoral cascade of sepsis the transit time of leukocytes through the microvasculature of the heart is increased leading to the retention of large numbers of activated leukocytes within the capillaries of the heart. This observation prompted the development of a novel model of septic myocardial dysfunction in which an isolated rabbit heart was perfused with blood from a second donor rabbit; the donor rabbit was infused with endotoxin and contractile function was measured in the isolated heart. This simultaneously permitted exposure of a normal heart to the complete cytohumoral cascade of sepsis but also allowed extraneous variables such as preload and afterload to be controlled or eliminated. This model was employed in Chapter III to demonstrate that the retention of large numbers of activated leukocytes within the microcirculation of the heart precedes the development of decreased myocardial contractile function and that decreased myocardial contractile function occurs in association with the development of myocardial edema and myocyte structural changes. Such structural changes of the myocardium have been associated with decreased myocardial contractile function in other experimental models and may contribute to endotoxemic or septic myocardial failure. 120 In Chapter IV it was demonstrated that the neutrophil fraction of leukocytes is slowed and retained to a much greater extent than the lymphocyte fractions. The retention of activated neutrophils is mediated by the effects of humoral inflammatory mediators on the heart and not,on the neutrophils themselves. Furthermore it was suggested that activation of quiescent neutrophils may occur locally within the microcirculation of an activated organ such as the heart. Finally, in Chapter V, it was demonstrated that the exclusion of activated neutrophils from the capillary circulation of the heart during acute endotoxemia protects the myocardium from the associated pathologic structural changes demonstrated in Chapter III and that such protection decreases the loss of myocardial contractile function. 6.2 Recommendations for further study The above studies demonstrate that the entrapment of neutrophils within the microcirculation of the heart is mediated by the effects of humoral mediators of SIRS on the heart. A number of putative mechanisms were proposed whereby such mediators may lead to myocardial neutrophil retention, as discussed above. However, these have not been definitively studied as part of this thesis. At the present time the effects of acute endotoxemia on the microstructure of the coronary vasculature are being studied. In particular the contribution of myocardial endothelial cell swelling and cytoskeletal changes to the retention of activated leukocytes is being studied using electron microscopy and quantitative morphometric analysis [Fig. 6.1, panel B]. In addition a model is being developed in which receptor ligands present on the surface of endothelial cells following activation by humoral mediators of sepsis are 121 FIGURE 6.1 Electronmicrographs of control and septic cardiac endothelial cells Fig. 6.1, Panel A: Control myocardium. Panel B: Electronmicrograph demonstrating myocardial endothelial cell swelling (black arrow heads) and cytoskeletal changes (arrow). Panel C: Electronmicrograph demonstrating endothelial cell membrane changes in apposition to a poly-morphonuclear neutrophil (PMN) (white arrow head) versus normal appearing endothelial cell membrane (blackarrow head). Panel D: Electronmicrograph demonstrating the putative "lacunar zone" between a cardiac endothelial cell and a PMN (white arrow heads), a PMN granule has been dicharged into this zone (white arrow). Panels B through D are derived from isolated-sup-ported rabbit hearts exposed to endotoxemic blood for four hours. Magnification bars, panels A through C are 10 um, panel D is 5 urn. 122 quantified and their contribution to neutrophil retention studied using specific receptor blocking agents. Another outstanding question is the specific effect of activated neutrophils on cells of the heart which either directly or indirectly lead to the observed structural changes. This is being studied in two ways. First, electron microscopy is being used to examine the nature of the boundary zone between the surface of the neutrophil and that of the endothelial cell, the so called "lacunar" zone. This zone may contain cytokine, metabolite or oxidative mediators in excess of that measured in the peripheral circulation such that normal defense systems may be overwhelmed. Endothelial cell plasma membrane changes or damage may therefore be concentrated in this area [Fig. 6.1, panels C and D]. Secondly, a model of isolated myocardial cell function in which cardiac myocytes or endothelial cells are co-cultured with various leukocyte populations is being developed. It is hoped that it will be possible to separate the effects of activated leukocyte species into two basic groups. In particular, it is hoped that effects related to the release of toxic species into the vicinity of myocardial cells may be separated from effects related to the flow of information, via either soluble humoral or fixed receptor mediators, from leukocyte to myocardial cell or vice versa. To facilitate this study a model utilizing cytoplast technology which depletes whole leukocytes of their organelles while allowing precise control of the number and type of cell surface receptors is being developed. This will be developed with the ultimate aim of "reprogramming" the SIRS state and preventing or reversing septic myocardial dysfunction. 123 BIBLIOGRAPHY CHAPTER I 1 Arai, K., F. Lee, A. Miyajima, S. Miyatake, N. Arai and T. Yakata. Cytokines: coordinators of immune and inflammatory responses. Annu. Rev. Biochem. 59: 783-836, 1990. 2 Bagge, U., B. Amundson and C. Lauritzen. White blood cell rheology: an intravital study in man. Adv. Microcirc.: 1-17,1977. 3 Bone, R.C.. Why sepsis trials fail. JAMA 276(7): 565-566, 1996. 4 Bone, R.C., C.J. Fisher, T.P. Clemmer et al. The methylprednisolone severe sepsis study group: a controlled clinical trial of high-dose methylprednisolone in the treatment of severe sepsis and septic shock. N. Engl. J. Med. 317: 653-658, 1987. 5 Bone, R.C., C.J. Fisher, T.P. Clemmer, G.J. Slotman, CA. Metz, and R A Balk. Sepsis syndrome: a valid clinical entity. Crit. Care Med. 17: 389-393, 1989.. 6 Carli, A., M.C. Auclair, C. Benassayag and E. Nunez. Evidence for an early lipid soluble eardiodepressant factor in rat serum after a sublethal dose of endotoxin. Circ. Shock 8: 301-312, 1981. 7 Christman, J.W., A.P. Wheeler and G.R. Bernard. Cytokines and sepsis: what are the therapeutic implications. J, Crit. Care 6(3): 172-182, 1991. 8 Decking, U.K.M., C.W. Flesche, A. Godecke and J. Schrader. Endotoxin-induced contractile dysfunction in guinea pig hearts is not mediated by nitric oxide. Am. J. Physiol. 268: H2460-H2565, 1995. 9 Downey, G.P., T. Fukushima, L. Fialkow and T.K. Waddell. Intracellular signaling in neutrophil priming and activation. Seminars in cell biology 6(6): 345-356, 1995. 10 Engler, R.L, G.W. Schmid-Schonbein and R.S. Pavelec. Leukocyte capillary plugging in myocardial ischemia and reperfusion in the dog. Am. J. Pathol. 111: 98-111, 1983. 11 Entman, M.L., K. Youker, SB . Shappell, C. Siegel, R. Rothlein, W.J. Dreyer, F.. Scnmalstieg and C.W. Smith. Neutrophil adherence to isolated 124 adult canine myocytes: evidence for a CD18-dependent mechanism. J . Clin. Invest. 85: 1497-1506, 1990. 12 Entman, M.L, K. Youker, T. Shoji, G. Kukielka, S.B. Shappell, A.A. Taylor and C.W. Smith. Neutrophil induced oxidative injury of cardiac myocytes. J. Clin. Invest. 90: 1335-1345, 1990. 13 Finkel, M.S., .V. Oddis, T.D. Jacob, S.C. Watkins, B.G. Hattler and R.L Simmons. Negative inotropic effects of cytokines on the heart mediated by nitric oxide. Science 257: 387-389, 1992. 14 Fox, G.A., N.A. Paterson and D'.G. McCormack. Cyclooxygenase inhibition and vascular reactivity in a rat model of hyperdynamic sepsis. J. Cardiovasc. Pharmacol. 28(1): 30-35, 1996. ; 15 Frank, R.S.. Time-dependent alteration in the deformability of human neutrophils in response to chemotactic activation. Blood 76: 2606-2612, 1990. 16 Gomez, A., R. Wang, H. Unruh, R.B. Light, D. Bose, T. Chau, E. Corron, and S. Mink. Hemofiltration reverses left ventricular dysfunction during sepsis in dogs. Anesthesiology 73:671-685,1990. 17 Hallstrom, S., C. Vogl, H. Redl and G. Schlag. Net inotropic plasma activity in canine hypovolemic traumatic shock: low molecular weight plasma fraction after prolonged hypotension depresses cardiac muscle function in vitro. Circ. Shock 30: 129-144, 1990. 18 Harlan, J.M. Consequences of leukocyte-vessel wall interactions in inflammatory and immune reactions. Seminars in Thrombosis and Hemostasis 13(4): 434-444, 1987. . 19 Herbertson, M.J., H.A. Werner, C M . Goddard, JA. Russel, A. Wheeler, R. Coxon and K.R. Walley. Anti-tumor necrosis factor-a prevents decreased ventricular contractility in endotoxemic pigs. Am. J. Respir. Crit. Care Med. 152: 480-488. 20 Herbertson, M.J., H.A. Werrier, J.A. Russel, K. Iversen and K.R. Walley. Myocardial oxygen extraction ratio is decreased during endotoxemia in pigs. J. Appl. Physiol. 79: 479-486, 1995. 21 Hess, M.L, A. HastNIo, L.J. Greenfield. Spectrum of cardiovascular function during Gram-negative sepsis. Prog. Cardiovasc. Dis. 23: 279-298, 1981. 125 22 Hinshaw, L, P. Peduzzi, E. Young, C. Sprung, C/Shatney, J. Sheagran. Effect of high-dose glucocorticoid therapy on mortality in patients with clinical signs of systemic sepsis. N. Engl. Med. 317: 659T665, 1987. 23 Hoffstein, ST., R.S. Friedman and G. Weissman! Degranulation, membrane addition, and shape change during chemotactic factor-induced aggregation of human neutrophils. J. Cell Biol. 95: 234-241, 1982. 24 Hogg, J.C., and C M . Doerschuk. Leukocyte traffic in the lung. Annu. Rev. Physiol. 57: 97-114. 25 Hotchkiss, R.S. and I.E. Karl. Calcium: a regulator of the inflammatory response in endotoxemia and sepsis. New Horizons 4(1): 58-71, 1996. 26 Humer, M.F., P.T. Phang, B.P. Friesen, M.F. Allard, C M . Goddard and K.R. Walley. Heterogeneity of gut capillary transit times and impaired gut oxygen extraction in endotoxemic pigs. J. Appl. Physiol. 81: 895-904, 1996. 27 Hung, J.H. and W.Y.W. Lew. Temporal sequence of endotoxin-induced systolic and diastolic myocardial depression in rabbits. Am. J. Physiol. 265 (Heart Circ. Physiol. 34): H810-H819, 1993. 28 Krosl, P., J. Pretorius, H. Red I and G. Schlag. Myocardial function in septic sheep. Shock 1(5): 325-334, 1994. 29 Lam, C , K. Tyml, C. Martin and W. Sibbald. Microvascular perfusion is impaired in a rat model of normotensive sepsis. J. Clin. Invest. 94: 2077-2083, 1994. 30 Lefer, D.J.. Role of leukocyte adhesion in myocardial ischemia-reperfusion injury. Physiology and pathophysiology of leukocyte adhesion. New York, Oxford University Press: 393-407,1995. 31 Lefer, A. and T. Inge. Differentiation of a myocardial depressant factor present in shock plasma from known plasma peptides and salts. Proc. Soc. Exp. Biol. 142: 422-433, 1973. 32 Ma, A.L., P.S. Tsao and A. Lefer. Antibody to CD18 exerts endothelial and cardiac protective effects in myocardial ischemia and reperfusion. J. Clin. Invest. 88:1237-1243, 1991. 33 Matsuda, T., I. Rubenstein, R.A Robbins, S. Koyama, W.J. Joyner and S.I. Rennard. Role of neutrophils in endotbxin-mediated microvascular injury in hamsters. J. Appl. Physiol. 71:307-313, 1991. 126 34 McEver, R.P.. Leukocyte interactions mediated by selectins. Thrombosis and hemostasis 65(3): 223-228, 1991. 35 Myers, M.L, C. Webb, M. Moffat, D. Mclver and R. Del Maestro. Activated neutrophils impair rabbit heart recovery after hypothermic global ischemia. Ann. Thorac. Surg. 53: 247-252,1992. 36 Natanson, C , M.P. Fink, H.K. Ballantyne, T.J. MacVittie, J.J. Conklin and J.E. Parillo. Gram-negative bacteremia produces both severe systolic and diastolic cardiac dysfunction in a canine model that simulates human septic shock. J. Clin. Invest. 78: 259-270,1986. 37 Natanson, C , W.D. Hoffman, A.F. Suffredini, P.Q. Eihacker and R.L. Danner. Selected treatment strategies for septic shock based on proposed mechanisms of pathogenesis. Ann; Int. Med. 120(9): 771-783, 1994. 38 Nathan, C. and M. Sporn. Cytokines in context. J. Cell Biol. 113(5): 981-986, 1991. 39 Noshima, S., H. Noda, D.N. Herndon; LD. Traber and D.L Traber. Left ventricular performance during continuous endotoxin-induced hyperdynamic endotoxemia in sheep. J. Appl. Physiol. 74(4): 1528-1533, 1993. 40 Parillo, J.E., C. Burch, J.H. Shelhamer, M.M. Parker, C. Natanson and W. Schuette. A circulating myocardial depressant substance in humans with .septic shock. J. Clin. Invest. 76:1539-1553, 1985. 41 Parker, M.M., J.H. Shelhamer, S.L. Bacharach, M.V: Green, C. Natanson and T.M. Frederick. Profound but reversible myocardial depression in patients with septic shock. Ann. Intern. Med. 100: 483-490,1984. 42 Redl, H., G. Schlag, S. Bahrami, J. Davies, M. Jochum and A. Bengtsson. Experimental and clinical evidence of leukocyte activation in trauma and sepsis. Prog. Clin. Biol. Res, 388: 221-245,1994. 43 Sheridan, F.M., I.M. Dauber, I.F. McMurtry, E.J. Lesnefsky and L.D. Horwitz. Role of leukocytes in coronary vascular endothelial injury due to ischemia and reperfusion. Circ. Res. 69:1566-1574, 1991. 44 Simpson, P.J., R.F. Todd m, j.c. Fantone, J.K. Mickelson, J.D. Griffin and B.R. Lucchesi. Reduction of experimental canine myocardial reperfusion injury by a monoclonal antibody, ariti-Mo1, anti-CD11b) that inhibits leukocyte adhesion. J. Clin. Invest. 81: 624-629,1988. 127 45 Smith, J.A.. Neutrophils, host defense, and inflammation: a double-edged sword. J. Leukoc. Biol. 56: 672-686,1994. 46 Song, H and K. Tyml. Evidence for sensing and integration of biological signals by the capillary network. Am. J. Physiol. 265: H1235-H1242, 1993. 47 Springer, T.A.. Adhesion receptors of the immune system. Nature 346: 425-434, 1990. 48 Sprung, C L , V.C Panagiota, E.H. Marcial, M. Pierce, M.A. Gelbard and W.M. Long. The effects of high dose corticosteroids in patients with septic shock: A prospective, controlled study. N. Engl. Med. 311: 1137-1143, 1984. 49 Suffredini, A.F., R.E. Fromm, M.M. Parker, M. Brenner, .A. Kovacs and RA Wesley. The cardiovascular response of normal humans to the administration of endotoxin. N. Engl. Med. 321: 280-287, 1989. 50 Sung, K.-L.P., . Dong, G.W. Schmid-Schonbein, S. Chien and R. Skalak. Leukocyte relaxation properties. Biophys. J . 54: 331-336, 1988. 51 Taber's Cycylopedic Medical Dictionary 16 Ed.. F.A. Davis Company, 1989. 52 Thomas, J.R., J.M. Harlan, C L . Rice and .K. Winn. Role of leukocyte CD11/CD18 complex in endotoxic and septic shock in rabbits. J. Appl. Physiol. 73(4): 1510-1516, 1992. 53 Thomas, F., J. Smith, J. Orme Jr., T. Clemmer, A. Hagan and G. Elliott. Reversible segmental myocardial dysfunction in septic shock. Crit. Care Med. 14: 587-588, 1986. 54 Tonnensen, M.G.. Neutrophil-endothelial cell interactions: mechanisms of neutrophil adherence to vascular endothelium. J. Invest. Dermatol. 93: 53s-58s, 1989. 55 Tracey, K.J. and S.F. Lowry. The role of cytokine mediators in septic shock. Adv. Surg. 23: 21-56,1990. 56 Walley, K.R.. Heterogeneity of oxygen delivery impairs oxygen extraction by peripheral tissues: Theory. J. Appl. Physiol. 81: 885-894,1996. 57 Walley, K.R., P.C. Hebert, Y.Wakai, P. Wilcox. J. Road and J. Cooper. Decrease in left ventricular contractility after tumor necrosis factor-a infusion in dogs. J. Appl. Physiol. 76:1060-1067, 1994. 128 58 Williams, F.M., P.D. Collins, M. Tanniere-Zeller and T.J. Williams. The relationship between neutrophils and increased microvascular permeability in a model of myocardial ischaemia and reperfusion in the rabbit. Br. J. Pharmacol. 100: 729-734,1990. 59 Worthen, G.S., B. Schwab III, E.L. Elson and G.P. Downey. Mechanics of stimulated neutrophils: cell stiffening induces retention in capillaries. Science Wash. DC 245: 183-186, 1989. 60 Wurster, S.H., P. Wang, R.E. Dean and I.H. Chaudry. Vascular smooth muscle contractile function is impaired during early and late stages of sepsis. J. Surg. Res. 56(6): 556-561, 1994. 61 Yaghi, A., N.A. Paterson and D.G. McCormack. Vascular reactivity in sepsis: importance of controls and role of nitric oxide. Am. J. Resp. Crit. Care Med. 151(3 pt 1): 706-712, 1995, CHAPTER II 1. Allard, M.F., C.T. Kamimura, D.R. English, S.L Henning and B.R. Wiggs. Regional myocardial capillary erythrocyte transit time in the normal resting heart. Circ. Res. 72: 187-193, 1993. 2. Barroso-Aranda, J., G.W. Schmid-Schoenbein, B.W. Zweifach and J.C. Mafhison. Polymorphonuclear neutrophil contribution to induced tolerance to bacterial lipopolysaccharide, Circ. Res. 69; 1196-1206, 1991. 3. Bengtsson, A., H. Redl, E, Paul, G. Schlag, T.E. Mollnes and J. Davies. Complement and leukocyte activation in septic baboons. Circ. Shock 83-88, 1993. 4. Bierman, H.R., K.H. Kelly, F.L Cordes, R.L Byron, J. Polhemus and S. Rappoport. The release of leukocytes and platelets from the pulmonary circulation by epinephrine. Blood, 7: 683-692, 1952. 5. Brady, A.J.B., P.A. Poole-Wilson, S.E. Harding and J.B. Warren. Nitric oxide production within cardiac myocytes reduces their contractility in endotoxemia. Am. J. Physiol. 263 (Heart Circ. Physiol. 32): H1963-H1966, 1992. 6. Brigham, K.L. and B. Meyrick. Endotoxin and lung injury. Am. Rev. Resp. Dis. 133:: 913-927, 1986. 129 7. Cruz-Orive, L.M. and Weibel. Sampling design for stereology. J. Microsc. Oxf. 122: 235-257, 1981. 8. Cunnion, R.E., G.L Schaer, M.M. Parker, C. Natanson and J.E. Parillo. The coronary circulation in human septic shock. Circulation 73: 637-644, 1986. 9. Dahinden, C , C. Galanos arid J. Fehr. Granulocyte activation. by endotoxin. I. Correlation between adherence and other granulocyte functions, and role of endotoxin structure on biologic activity. J. Immunol. 130: 857-862, 1983. 10. Doerschuk, CM., N. Beyers, H.O. Coxson, B. Wiggs and J.C Hogg. Comparison of neutrophil and capillary diameters and their relation to neutrophil sequestration in the lung. J. Appl. Physiol. 74: 3040-3045, 1993. 11. Downey, G.P., G.S. Worthen, P.M. Henson and D.M. Hyde. Neutrophil sequestration and migration in localized pulmonary inflammation. Am. Rev. Resp. Dis. 147: 168-176, 1993. 12. Entman, M.L., K. Youker, T. Shoji, G. Kukielka, S.B. Shapell, AA. Taylor and C.W. Smith. Neutrophil induced oxidative injury of cardiac myocytes. J. Clin, Invest. 90: 1335-1345, 1992. 13. Eppihimer, M.j; and H.H. Lipowsky. Leukocyte sequestration in the microvasculature in normal and low flow states. Am. J. Physiol. 267 (Heart Circ. Physiol. 36): H1122-H1134, 1994. 14. Finkel, M.S., CV. Oddis, T.D. Jacob, S.C Watkins, B.G. Hatter and R.L Simmons. Negative inotropic effects of cytokines on the heart mediated by nitric oxide. Science Wash. DC 257: 387-389, 1992. 15. Frank, R.S. Time-dependent alterations in the deformability of human neutrophils in response to chemotactic activation. Blood 76: 2606-2612, 1990 16. Goldfarb, R.D., L.M. Nightingale, P. Kish, P.B. Weber, and D.J. Loegering. Left ventricular function during lethal and sublethal endotoxemia in swine. Am. J. Physiol. 251 (Heart Circ. Physiol. 20): H364-H373, 1986. 17. Grayson, J., J.W. Davidson, A. Fjtzgerald-Finch and C. Scott. The functional morphology of the coronary microcirculation of the dog. Microvasc. Res. 8: 20-43,1976. 18. Haslett, C , G.S. Worthen, P.C. Giclas, D.C Morrison, J.E. Henson and P.M. Henson. The pulmonary vasculature sequestration of neutrophils in 130 endotoxemia is initiated by an effect of endotoxin on the neutrophil of the rabbit. Am. Rev. Resp. Dis. 136: 9-18,1987. 19. Hogg, J.C. Neutrophil kinetics and lung injury. Physiol. Rev. 67: 1249-1295, 1987. 20. Hogg, J.C. The interaction between polymorphonuclear cells and pulmonary endothelium. In: Pulmonary Vascular Physiology and Pathophysiology, edited by E.K. Weir, and J.T. Reeves. New York: Marcel Dekker, 1989, Chapt. 17, p. 630. 21. Hogg, J.C, H.O. Coxson, M.L. Brumwell, N. Beyers, C M . Doerschuk, W. MacNee and B.R. Wiggs. Erythrocyte and polymorphonuclear cell transit time and concentration in human pulmonary capillaries. J. Appl. Physiol. 77: 1795-1980, 1994. 22. Hogg, J.C. and C M . Doerschuk. Leukocyte traffic in the lung. Annu. Rev. Physiol. 57: 97-114, 1995. 23. Lien, D.C, P.M. Henson, R.L Capen, J.E. Henson, W.L Hanson, W.W. Wagner and G.S. Worthen. Neutrophil kinetics in the pulmonary microcirculation during acute inflammation. Lab. Invest. 65: 145-159, 1991. 24. Nash, G.B., J.G. Jones, J. Mikita, B. Christopher and J.A. Dormandy. Effects of preparative procedures and of cell activation on flow of white cells through micropore filters. Br. J: Haematol. 70:171-176, 1988. 25. Natanson, C. Studies using a canine model to investigate the cardiovascular abnormality of and potential therapeutic strategies for septic shock. Clin. Res. 38: 206-214,1990. 26. Nathan, C , S. Srimal, C. Farber, E. Sanchez, L. Kabbask, A. Asch, J. Gailit and S.D. Wright. Cytokine-induced respiratory burst of human neutrophils-dependence on extracellular matrix proteins and CD11/CD18 integrins. J. Cell. Biol. 109, 1341-1349, 1989. 27. Parillo, J.E., C. Burch, J.H. Shelhamer, MM. Parker, C Natanson and W. Schuette. A circulating myocardial depressant substance in humans with septic shock. J. Clin. Invest. 76:1539-1553,1985. 28. Parker, M.M., J.H. Shelhamer, S.L. Bacharach, M.V. Green, C. Natanson, T.M. Frederick, B.A. Damske and J.E. Parillo. Profound but reversible myocardial depression in patients with septic shock. Ann. Intern. Med. 100: 483-490, 1984. 131 29. Raper, R.F., W.J. Sibbald, J. Hobson, A. Neal and H. Chung. Changes in myocardial blood flow rated during hyperdynamic sepsis with induced changes in arterial perfusing pressures and metabolic need. Crit. Care Med. 21: 1192-1199, 1993. 30. Sommer, J.R. and R.B. Jennings. Ultrastructure of cardiac muscle. In: The Heart and Cardiovascular System: Scientific Foundations (2nd ed.), edited by H.A. Fozzard, E. Haber, R.B. Jennings, A.M. Katz, and H.E. Morgan. New York: Raven, 1991, p. 22. 31. Thommasen, H.V., B.A. Martin, B.R. Wiggs, M. Quiroga, E.M. Baile and J.C. Hogg. Effect of pulmonary blood flow on leukocyte uptake and release by the dog lung. J. Appl. Physiol. 56: 966-974, 1984. 32. Tracey, K.J. and S.F. Lowry. The role of cytokine mediators in septic shock. Adv. Surg. 28: 21-56, 1990. 33. Walley, K.R., C.J. Becker, R.A. Hogan, R. Teplinsky and L.D.H. Wood. Progressive hypoxemia limits left ventricular oxygen consumption and contractility. Circ. Res. 63: 849-859, 1988. 34. Walley, K.R., P.C. Hebert, Y. Wakai, P.G. Wilcox, J.D. Road and D.J. Cooper. Decrease in left ventricular contractility after tumor necrosis factor-a infusion in dogs. J. Appl. Physiol. 76:1060-1067, 1994. 35. Warnke, K.C. and T.C. Skalak. Leukocyte plugging in vivo in skeletal muscle arteriolar trees. Am. J. Physiol. 262 (Heart Circ. Physiol. 31): H1149-H1155, 1992. 36. Witthaut, R., A. Farhood, C.W. Smith and H. Jaeschke. Complement and tumor necrosis factor-alpha contribute to Mac-1 (CD11b/CD18) up-regulation and systemic neutrophil activation during endotoxemia in vivo. J. Leukocyte Biol. 55: 105-111, 1994. 37. Xing, Z., H. Kirpalani, D. Torry, M. Jordana and J. Gauldie. Polymorphonuclear leukocytes as a significant source of tumor necrosis factor-a in endotoxin challenged lung tissue. Am. J. Pathol. 143: 1009-1015, 1994. 38. Yui, Y., R. Hattori, K. Kosuga, H. Eizawa, K. Hiki, S. Ohkawa, K. Ohnishi, S. Terao and C. Kawai. Calmodulin-independent nitric oxide synthase from rat polymorphonuclear neutrophils. J. Biol. Chem. 266: 3369-3371, 1991. 132 CHAPTER III 1. Allard, M. F., CT. Kamimura, D.R. English, S.L. Henning, and B.R. Wiggs. Regional myocardial capillary erythrocyte transit time in the normal resting heart. Circ. Res. 72: ,187-193, 1993. 2. Barroso-Aranda, J., G.W. Schmid-Schonbein, B.W. Zweifach and J .C Mathison. Polymorphonuclear neutrophil contribution to induced tolerance to bacterial lipopolysaccharide. Circ. Res. 69: 1196-1206, 1991. 3. Cruz-Orive, L.M. and E.R. Weibel. Sampling designs for stereology. J: Microsc. 122: 235-257, 1981. 4. Dhainaut, J., M. Huyghebaert, J.F. Monsallier, G. Lefevre, J. Dall'Ava-Santucci, F. Branet, D.. Villemant, A. Carli and D. Raichvarg. Coronary hemodynamics and myocardial metabolism of lactate, free fatty acids, glucose, and ketones in patients with septic shock. Circulation 75: 533-541, 1987. 5. Entman, M.L, L. Michael, R.D. Rossen, W.J. Dreyer, D.C Anderson, A.A. Taylor and C.W. Smith. Inflammation in the course of early myocardial ischemia. FASEB J. 5: 2529-2537, 1991. 6. Entman, M.L, K. Youker, T. Shoji, G. Kukielka, S.B. Shapell, AA. Taylor and C.W. Smith. Neutrophil induced oxidative injury of cardiac myocytes. J.Clin. Invest. 90: 1335-1345, 1992. 7. Finkel, M.S., CV. Oddis, T.D. Jacob, S.C Watkins, B.G. Hattler and R.L Simmons. Negative inotropic effects of cytokines on the heart mediated by nitric oxide. Science Wash. DC 257: 387-389, 1992. 8. Goddard, C M , M.F. Allard, J.C. Hogg, M.J. Herbertson and K.R. Walley. Prolonged leukocyte transit time in coronary microcirculation of endotoxemic pigs. Am. J. Physiol. 269 (Heart Circ. Physiol. 38): H1389-H1397,1995. 9. Gomez, A., R. Wang, H. Unruh, R.B. Light, D. Bose, T. Chau, E. Correa and S. Mink. Hemofiltration reverses left ventricular dysfunction during sepsis in dogs. Anesthesiology 73: 671-685, 1990. 10. Hesse, D.G., K.J. Tracey, Y. Fong, K.R. Manogue, M.A. Palladino, A. Cerami, G.T. Shires and S.F. Lowry. Cytokine appearance in human endotoxemia and primate bacteremia. Surg. Gynecol. ObsteL 166:147-153,1988. 133 11. Holland, B.S. and M. DiPonzio Copenhaver. An improved sequentially rejective Bonferroni test procedure. Biometrics 43: 417-423-1987. 12. Hsu, D.T., Z.C. Weng, A.C. Nicolosi, P.W. Detwiler, R. Sciacca and H.M. Spotnitz. Quantitative effects of myocardial edema on the left ventricular pressure-volume relation. Influence of cardioplegia osmolarity over two hours of ischemic arrest. J . Thorac. Cardiovasc. Surg. 106: 651-657, 1993. 13. Hung, J. and W.Y.W. Lew. Cellular mechanisms of endotoxin-induced myocardial depression in rabbits. Circ. Res. 73: 125-134, 1993. 14. Hung, J. and W.Y. W. Lew. Temporal sequence of endotoxin-induced systolic and diastolic myocardial depression in rabbits. Am. J. Physiol. 265 (Heart Circ. Physiol. 34): H810-H819, 1993. 15. Janicki, J.S., S.G. Shroff and K.T. Weber. Ventricular interdependence. In: Heart-Lung Interactions in Health and Disease, edited by S. M. Scharf and S. Cassidy. New York: Dekker, 1989, p. 283-307. 16. Kass, D A , W.L Maughan, Z.M. Guo, A. Kono, K. Sunagawa and K. Sagawa. Comparative influence of load versus inotropic states on indexes of ventricular contractility: experimental and theoretical analysis based on pressure-volume relationships. Circulation 76: 1422-1436, 1987. 17. Marban, E. Myocardial stunning and hibernation: the physiology behind the colloquialisms. Circulation 83: 681-688, 1991. 18. Martin, S. E., D.E. Chenoweth, R.L. Engler, D.M. Roth and J.C. Longhurst. C5a decreases regional coronary blood flow and myocardial function in pigs: implications for a granulocyte mechanism. Circ. Res. 63: 483-491, 1988. 19. Maughan, W.L, K. Sunagawa, D. Burkhoff, W.L. Graves, W.C. Hunter and K. Sagawa. Effect of heart rate on the canine end-systolic pressure-volume relationship. Circulation 72: 654-659, 1985. 20. Natanson, C , M. P. Fink, H.K. Ballantync, T.J. MacVittie, J.J. Conklin and J.E. Parrillo. Gram-negative bacteremia produces both severe systolic and diastolic cardiac dysfunction in a canine model that simulates human septic shock. J. Clin. Invest. 78: 259-270,1986. 21. Noshima, S., H. Noda, D.N. Herndon, L.D. Traber. and D.L. Traber. Left ventricular performance during continuous endotoxin-induced hyperdynamic endotoxemia in sheep. J. Appl. Physiol. 74:1528-1533,1993. 134 22. Parker, M.M., J.H. Shelhamer, S.L. Bacharach, M.V. Green, C. Natanson, T.M. Frederick, B.A. Damske, and J.E. Parrillo. Profound but reversible myocardial depression in patients with septic shock. Ann. Intern. Med. 100: 483-490, 1984. 23. Parrillo, J.E., C. Burch, J.H. Shelhamer, M.M. Parker, C. Natanson and W. Schuette.. A circulating myocardial depressant substance in humans with septic shock. J. Clin. Invest. 76:1539-1553-1985. 24. Parrillo, J.E., MM. Parker, C. Natanson, A.F. Suffredini, R.L. Danner, R.E. Cunnion arid F.P. Ognibene. Septic shock in humans. Ann. Intern. Med. 113: 227-242, 1990. 25. Sandhu, R., R.J. Diaz and G.J. Wilson. Comparison of ischemic preconditioning in blood perfused and buffer perfused isolated heart models. Cardiovasc. Res. 27: 602-607, 1993. 26. Suffredini, A.F., R.E. Fromm, M. M. Parker, M. Brenner, J.A. Kovacs, R.A. Wesley and J.E. Parrillo. The cardiovascular response of normal humans to the administration of endotoxin. N. Engl. J. Med. 321: 280-287, 1989. 27. Sugi, K., J. Newald, L.D. Traber, J.P. Maguire, D.N. Herndon, G. Schlag, and D.L. Traber. Cardiac dysfunction after acute endotoxin administration in conscious sheep. Am. J. Physiol. 260 (Heart Circ. Physiol. 29): H1474-H1481,1991. 28. Sunagawa, K., W.L. Maughan, G. Friesinger, P. Guzman, M.S. Chang and K. Sagawa. Effects of coronary arterial pressure on left ventricular end-systolic pressure-volume relation of isolated canine heart. Circ. Res. 50: 727-734,1982. 29. Teplinski, K., M. OToole, M. Olman, K.R. Walley and L.D.H. Wood. Effect of lactic acidosis on canine hemodynamics and left ventricular function. Am. J. Physiol. 258 (Heart Circ. Physiol. 27): H1193-H1199,1990. 30. Walley, K.R., C.J. Becker, R.A. Hogan, K. Teplinsky and L.D.H. Wood. Progressive hypoxemia limits left ventricular oxygen consumption and contractility. Circ. Res. 63: 849-859,1988. 31. Walley, K.R. and D.J. Cooper. Diastolic stiffness impairs left ventricular function during hypovolemic shock in pigs. Am. J. Physiol. 260 (Heart Circ. Physiol. 29): H702-H712, 1991. 32. Walley, K.R., P.C. Hebert, Y. Wakai, P.G. Wilcox, J.D. Road and D.J. Cooper. Decrease in left ventricular contractility after tumor necrosis factor-a infusion in dogs. J. Appl. Physiol. 76: 1060-1067-1994. 135 33. Wolff, M.R., P.P. deTombe, Y. Harasawa, D. Burkhoff, S. Bier, W.C. Hunter, G. Gerstenblith and K.A. Kass. Alterations in left ventricular mechanics, energetics, and contractile reserve in experimental heart failure. Circ. Res. 70: 516-529, 1992. 34. Yaku, H., B.K. Stinker, E.S.P. Myhre, M.W. Watkins and M.M. Lewinter. Stability of myocardial 02 consumption pressure-volume area relation in red cell-perfused rabbit heart. Am. J. Physiol. 261 (Heart Circ. Physiol. 30): H1630-H1635.1991. 35. Youker, K, C.W. Smith, D.C. Anderson, D. Miller, L.H. Michael, R.D. Rossen and M.L. Entman. Neutrophil adherence to isolated adult cardiac myocytes. J. Clin. Invest. 89: 602-609,1992. CHAPTER IV 1. Barroso-Aranda, J., G.W. Schmid-Schoenbein, B.W. Zweifach' and J.C. Mathison. Polymorphonuclear neutrophil contribution to induced tolerance to bacterial lipopolysaccharide. Circ. Res. 69: 1196-1206, 1991. 2. Bellary, S.S., K.W. Anderson, W.A. Arden and D.A. Butterfield. Effect of lipopolysaccharide on the physical conformation of the erythrocyte cytoskeletal proteins. Life Sciences. 56(2): 91 -98,1995. 3. Bengtsson, A. Cascade system activation in shock. Acta. Anaes. Scand. 98:7-10, 1993. 4. Bengtsson, A., H. Redl, E. Paul, G. Schlag, T.E. Mollnes and J. Davies. Complement and leukocyte activation in septic baboons. Circ. Shock. 39: 83-88, 1993. 5. Dinarello, C.A. Reduction of inflammation by decreasing production of interleukin-1 or by specific receptor antagonism. Int. J. Tiss. Reac. 14(2): 65-75, 1992. 6. Finn, A., S. Strobel, M. Levin and N. Klein. Endotoxin-induced neutrophil adherence to endothelium: relationship to CD11b/CD18 and L-selectin expression and matrix disruption. Annals New York Acad. Sci. 725: 173-182, 1994. 7. Goddard, CM., M.F. Allard, J.C. Hogg, M.J. Herbertson and K.R. Walley. Prolonged leukocyte transit time in coronary microcirculation of 136 endotoxemic pigs. Am. J. Physiol. 269 (Heart Circ. Physiol. 38): H1389-H1397, 1995. 8. Goddard, CM. , M.F. Allard, J.C. Hogg and K.R. Walley. Myocardial morphometric changes related to decreased contractility after endotoxin. Am. J. Physiol. 270 (Heart Circ. Physiol. 39): H1446-H1452, 1996. 9. Granton, J.T., CM. Goddard, M.F. Allard, J.C. Hogg, and K.R. Walley. Removal of circulating leukocytes attenuates myocardial dysfunction in an isolated-supported rabbit heart during endotoxemia. Abstract Am. J. Respir. Crit. Care Med. 151: A316,1995. 10. Haziot, A., B.Z. Tsuberi and S.M: Goyert. Neutrophil CD14: biochemical properties and role in the secretion of tumor necrosis factor-alpha in response to lipopolysaccharide. J. Immu. 150(12): 5556-5565, 1993. 11. Hinshaw, L.B., LT. Archer and B.K. Beller-Todd. Hematologic disturbances during sepsis: platelets and leukocytes. Adv. Shock Res. 7: 1-6,1982. 12. Hogg, J.C. Neutrophil kinetics and lung injury. Physiol. Rev. 67: 1249-1295, 1987. 13. Humer, M.F., P.T. Phang, B.P. Friesen, M.F. Allard, C M . Goddard and K.R. Walley. Heterogeneity of gut capillary transit times and impaired gut oxygen extraction in endotoxemic pigs. J. Appl. Physiol. 81(2): 895-904, 1996. 14. Hung, J. and W.Y.W. Lew. Cellular mechanisms of endotoxin-induced myocardial depression in rabbits. Circ. Res. 73:125-134, 1993. 15. Kelm, M. and J. Schrader. Control pf coronary vascular tone by nitric oxide. Circ. Res. 66:1561-1575,1990. 16. Kreimeier, U., F. Hammerson, M. Ruiz-Morales, Z. Yang and K. Messmer. Redistribution of intraorgan blood flow in acute, hyperdynamic porcine endotoxemia. Eur. Surg. Res. 23: 85-99, 1991. 17. Lamas, S., T. Michel, B.M. Brenner and P.A. Marsden. Nitric oxide synthesis in endothelial cells: evidence for a pathway inducible by TNF-a. Am, J . Physiol. 261 (Cell Physiol. 30): C634-C641, 1991. 18. Lassen, N.A. and W. Perl. Tracer kinetics in medical physiology. Raven Press, New York. 1979. 137 19. Martin, S.E., D.E. Chenoweth, R.L Engler, D.M. Roth and J.C. Longhurst. C5a decreases regional coronary blood flow and myocardial function in pigs: implications for a granulocyte mechanism. Circ. Res. 63: 483-491, 1988. 20. Massey, K.D., R.M. Strieter, S.L Kunkel, J.M. Danforth and T.J. Standiford. Cardiac myocytes release leukocyte-stimulating factors. Am. J. Physiol. 269 (Heart. Circ. Physiol. 38): H980-H987, 1995. 21. Natanson, C , W.D. Hoffman, A.F. Suffredini, P.O. Eichacker and R.L. Danner. Selected treatment strategies for septic shock based on proposed mechanisms of pathogenesis. Ann. Intern. Med. 120: 771-783, 1994. 22. Nathan, C , S. Srimal, C. Farber, E. Sanchez, L. Kabbask, A. Asch, J Gailit and S.D. Wright. Cytokine-induced respiratory burst of human neutrophils-dependence on extracellular matrix proteins and CD11/CD18 integrins. J. Cell. Biol. 109: 1341-1349,1989. 23. Nawroth, P.P., I. Bank, D. Handley, J. Cassimeris, L. Chess and D. Stern. Tumor necrosis factor/cachectin interacts with endothelial cell receptors to induce release of interleukin-1. J. Exp. Med. 163: 1363-1375, 1986. 24. Ohgami, M., CM. Doerschuk, D. English, P.M. Dodek and J.C. Hogg. Kinetics of radiolabeled neutrophils in swine. J. Appl. Physiol. 66(4): 1881-1885, 1989. 25. Pugin, J., R.J. Ulevitch and P.S. Tobias. A critical role for monocytes and CD14 in endotoxin-induced endothelial cell activation. J. Exp. Med. 178: 2193-2200, 1993. 26. Saigado, A., J.L. Boveda, J. Monasterio, R.M. Segura, M. Mourelle, J. Gomezjiminez and R. Peracaula. Inflammatory mediators and their influence on haemostasis. Haemostasis 24: 132-138, 1994. 27. Salyer, J.L, J.F. Bohnsack, W.A. Knape, A.O. Shigeoka, E.R. Ashwood, and H.R. Hill. Mechanisms of tumor necrosis factor- alteration of PMN adhesion and migration. Am. J. Path. 136(4): 831-841,1990. 28. Schwartz, M.A, M.D. Schaller and M.H. Ginsberg. Integrins: emerging paradigms of signal transduction. Annli. Rev. Cell Dev. Biol. 11: 549-599, 1995. 29. Steinhoff, G., M. Behrend, B. Schrader, AM. Duijvestijn and K. Wonlgeit. Expression patterns of leukocyte adhesion ligand molecules on human liver endothelia. Am. J. Pathol. 142: 481-488,1993. 138 30. Takahashi, K., K. Ando, A. Ono, T. Shimosawa, E. Ogata and T. Fujita. Tumor necrosis factor-a induces vascular hyporesponsiveness in Sprague-Dawley rats. Life Sciences. 50: 1437-1444, 1992. 31. Vercellotti, G.M., N.W.R. Wickham, K.S. Gustafson, H.Q. Yin, M. Hebert and H.S. Jacob. Thrombin-treated endothelium primes neutrophil functions: inhibition by platelet-activating factor receptor antagonists. J. Leuk. Biol. 45: 483-490, 1989. 32. Walley K.R. Heterogeneity of oxygen delivery impairs oxygen extraction by peripheral tissues: theory. J. Appl. Physiol. 81(2): 885-894, 1996. 33. Witthaut, R., A. Farhood, C.W. Smith and H. Jaeschke. Complement and tumor necrosis factor-alpha contribute to Mac-1 (CD11b/CD18) up-regulation and systemic neutrophil activation during endotoxemia in vivo. J. Leukocyte Biol. 55: 105-111,1994. 34. Xie, J., Y. Wang, J. Kolis, T. Malinski, S. Nelson, W. Summer and S.S. Greenberg. Tumor necrosis factor-a inhibits contractions to sympathetic nerve stimulation by a nitric oxide-dependent mechanism. Proc. Soc. Exp. Biol. Med. 203: 447-453, 1993. CHAPTER V 1. Barroso-Aranda, J., G.W. Schmid-Schonbein, B.W. Zweifach and J.C. Mathison. Polymorphonuclear neutrophil contribution to induced tolerance to bacterial lipopolysaccharide. Circ. Res. 69:1196-1206,1991. 2. Bock, M., C. Krause, M. Wagner, S. Braun, U. Pachmann and W. Memphel. Transfusion of leukocyte-depleted red cell concentrates and whole blood with anew bedside filter system. Plasma Ther. Transfus. Technol. 8: 302-306, 1987. 3. Carli, A., M.C. Auclair, C. Benassayag and E. Nunez. Evidence for an early lipid soluble cardiodepressant factor in rat serum after a sublethal dose of endotoxin. Circ. Shock 8: 301-312, 1981. 4. Deckin, U.K.M., C.W. Flosche, A. Golecke and J. Schrader. Endotoxin-induced contractile dysfunction in guinea pig hearts is not mediated by nitric oxide. Am. J. Physiol. 268: H2460-H2465, 1995. 5. Dhainaut, J.F., M.F. Huyghebaert, J.F. Monsallier, G. Lefevre, J. Dall'Ava-Santucci, F. Brunet, D. Villemant, A. Carli and D. Raichvarg. Coronary hemodynamics and myocardial metabolism of lactate, free fatty acids, 139 glucose, and ketones in patients with septic shock. Circulation 75: 533-541, 1987. 6. Eichacker, P.Q., A. Faresc, W.D. Hoffman, S.M. Banks, T. Mouginis, S. Richmond, G.C. Kuo, T.J. MacVittie and C. Natanson. Leukocyte CD11b/18 antigen-directed monoclonal antibody improves early survival and decreases hypoxemia in dogs challenged with tumor necrosis factor. Am. Rev. Respir. Dis. 145: 1023-1029, 1992. 7. Eichacker, P.Q., Y. Waisman, C. Natanson, a. Farese, W.D. Hoffman, S.M. Banks and T.J. Macvittie. Cardiopulmonary effects of granulocyte colony-stimulating factor in a canine model of bacterial sepsis. J. Appl. Physiol. 77: 2366-2373, 1994. 8. Finkel, M.S., CV. Oddis, T.D. Jacob, S.C Watkins, B.G. Hatter and R.L. Simmons. Negative inotropic effects of cytokines on the heart mediated by nitric oxide. Science 257: 387-389, 1992. 9. Goddard, CM., M.F. Allard, J.C. Hogg, M.J. Herbertson and K.R. Walley. Prolonged leukocyte transit time in coronary microcirculation of endotoxemic pigs. Am. J. Physiol. 269: H1389-H1397, 1995. 10. Goddard, CM., M.F. Allard, J.C. Hogg and K.R. Walley. Myocardial morphometric changes related to decreased contractility after endotoxin. Am. J. Physiol. 270: H1146-H1452, 1996.. 11. Hallstrom, S., C. Vogl, H. Redl and G. Schlag. Net inotropic plasma activity in canine hypovolemic traumatic shock: low molecular weight plasma fraction after prolonged hypotension depresses cardiac muscle function in vitro. Circ. Shock 30: 129-144, 1990. 12. Harris, R.I., P.C.W. Stone and J. Stuart. An improved chromogenic substrate endotoxin assay for clinical use. J. Clin. Pathol. 36: 1145-1149, 1983. 13. Herbertson, M.J., H.A. Werner, C.Mr Goddard, J.A. Russell, A. Wheeler, R. Coxon and K.R. Walley. Anti-tumor necrosis factor-a prevents decreased ventricular contractility in endotoxemic pigs. Am. J. Respir. Crit. Care Med. 152:480-488,1995. 14. Herbertson, M.J., H.A. Werner and K.R. Walley. Nitric oxide synthase inhibition partially prevents decreased LV contractility during endotoxemia. Am. J. Physiol. 270: H1979-H1984, 1996. 140 15. Herbertson, M.J., H.A. Werner and K.R. Walley. Platelet-activating factor antagonism improves ventricular contractility in endotoxemia. Crit. Care Med. 25: 221-227, 1997. 16. Hogg, J.C. and C M . Doerschuk. Leukocyte traffic in the lung. Annu. Rev. Physiol. 57: 97-114, 1995. 17. Jadschke, H., A. Farhood and C.W. Smith. Neutrophil-induced liver cell injury in endotoxin shock is CD11b/CD18- mechanism. Am. J. Physiol. 261: G1051-G1056, 1991. 18. Kass, D.A., W.L. Maughan, Z.M. Guo, a. Kono, K. Sunagawa and K. Sagawa. Comparative influence of load versus inotropic states on indexes of ventricular contractility: experimental and theoretical analysis based on pressure-volume relationships. Circulation 76: 1422-1436, 1987. 19. Lefer, A. and T. Inge. Differentiation of a myocardial depressant factor present in shock plasma from known plasma peptides and salts. Proc. Soc. Exp. Biol. 142: 422-433, 1973. 20. Ley, K., P. Gaehtgens, C. Fennie, M.S. Singer, L.A. Laskey and S.D. Rosen. Lectin-like cell adhesion molecule 1 mediates leukocyte rolling in mesenteric venules in vivo. Blood 77: 2553-2555, 1991. 21. Lund-Johansen, F. and L.W.M.M. Terstappen. Differential surface expression of cell adhesion molecules during granulocyte maturation. J. Leukocyte Biol. 54: 47-55, 1993. 22. Ma, A.L., P.S. Tsao and A. Lefer. Antibody to CD-18 exerts endothelial and cardiac protective effects in myocardial ischemia and reperfusion. J. Clin. Invest. 88: 1237-1243, 1991. 23. Ma., W.I., A.S. Weyrich, D.J. Lefer, M. Buerke, K.H. Albertine, T.K. Kishimoto and A.M. Lefer. Monoclonal antibody to L-selectin attenuates neutrophil accumulation and protects ischemic reperfused cat myocardium. Circulation 88: 649-658, 1993. 24. Massey, K.D., R.M. Steiter, S.L Kunkel, J.M. Danforth and T.J. Standiford. Cardiac myocytes release leukocyte-stimulating factors. Am. J. Physiol. 269: H980-H987, 1995. 25. Matsuda, T., I. Rubenstein, R.A. Robbins, S. Koyama, W.J. Joyner and S.I. Rennard. Role of neutrophils in endotoxih-mediated microvascular injury in hamsters. J. Appl. Physiol. 71: 307-313,1991. 141 26. Maughan, W.L, K. Sunagawa, D. Burkhoff, W.L Graves, W^C. Hunter and K. Sagawa. Effect of heart rate on the canine end-systolic pressure-volume relationship. Circulation 72: 654-659, 1995. 27. Parillo, J.E., C. Burch, J.H. Shelhamer, M.M. Parker, C. Natanson and W. Schuette. Circulating myocardial depressant substance in humans with septic shock. J. Clin. Invest. 76: 1539-1553,1985. 28. Parker, J. and R. Adams. Development of myocardial dysfunction in endotoxin shock. Am. J. Physiol. 248: H818-H826, 1985. 29. Parker, M.M., J.H. Shelhamer, S.L Bacharach, M.V. Green, C. Natanson, T.M. Frederick, B.A. Damske and J.E. Parillo. Profound but reversible myocardial depression in patients with septic shock. Ann. Intern. Med. 100: 483-490, 1985. 30. Sheridan, F.M., I.M. Dauber, I.F. McMurtry, E.J. Lesnefsky and LD. Horwitz. Role of leukocytes in coronary vascular endothelial injury due to ischemia and reperfusion. Circ. Res. 69: 1566-1574, 1991. 31. Smjth, J.A. Neutrophils, host defense, and inflammation: a double-edged sword. J. Leukocyte Biol. 56: 672-686, 1994. 32. Suffredini, A.F., R.E. Fromm, M.M. Parker, M. Brenner, J.A. Kovacs, R.A Wesley and J.E. Parillo. The cardiovascular response of normal humans to the administration of endotoxin. N. Engl. J. Med. 321; 280-287, 1989. 33. Sunagawa, K., G. Maughan, P. Freisinger, P. Guzman, M.S. Chang and K. Sagawa. Effects of coronary arterial pressure on left ventricular end-systolic pressure-volume relation of isolated canine heart. Circ. Res. 50: 727-734, 1982. 34. Thomas, J.R., J.M. Harian, C L . Rice and R.K. Winn. Role of leukocyte CD11/CD18 complex in endotoxic and septic shock in rabbits. J. Appl. Physiol. 73: 1510-1516, 1992. 35. Tiegs, G., J. Barsig, B. Matiba, S. Uhlig and A. Wendel. Potentiation by granulocyte macrophage colony-stimulating factor of lipopolysaccharide toxicity in mice. J. Clin. Invest. 93: 2616-2622,1994. 36. Wajley, K.R., P.C Hebert, Y. Wakai, P.G. Wilcox, J.D. Road and D.J. Cooper. Decrease in left ventricular contractility after tumor necrosis factor-a infusion in dogs. J. Appl. Physiol. 76:1060-1067, 1994. 37. Walzog, B., R. Seifert, a. Zakrzewicz, P. Gaehtgens and K. Ley. Cross-linking of CD18 in human neutrophils induces an increase of intracellular 142 free C a 2 + , exocytosis of azurophilic granules, quantitative up-regulation of CD18, shedding of L-selectin, and actin polymerization. J. Leukocyte Biol. 56:625-635,1994. Youker, K., C.W. Smith, D. Anderson, D. Miller, LH. Michael, R.D. Rossen and M.L. Entman. Neutrophil adherence to isolated adult cardiac myocytes. J. Clin. Invest. 89: 602-609, 1992. 143 APPENDIX I SYMBOLS AND ABBREVIATIONS USED IN THE TEXT S Y M B O L S a alpha 0 degree(s) < less than > greater than C 0 2 carbon dioxide H 2 0 water micro p.Ci micro-Curie(s) microgram(s) pi micrometer(s) u.m micrometer(s) 0 2 oxygen r 2 regression coefficient 85sr strontium-85 99 T c technetium-99 X magnified (multiplied) by + plus or minus A ANOVA analysis of variance 144 ARDS adult respiratory distress syndrome B BPM C °c C5a CD cm cmH20 C 0 2 CSF D dP dP/dtm a x DREG E EDTA ^max ESPVR beats per minute, beats/min degrees Celsius complement fraction 5a Common designation centimeter(s) centimeter(s) of water content of oxygen colony stimulating factor change in pressure maximal rate of change of intraventricular pressure down regulated antigen (L-selectin) ethylene diamine tetraacetic acid maximal elastance end-systolic pressure-volume relationship 145 EU endotoxin unit(s) F F French unit(s), Fahrenheit . Fig figure FITC fluorescein isothiocyanate G g gram(s) G-CSF granulocyte colony stimulating factor GM-CSF granulocytemacrophage colony stimulating factor H h hour(s) H heart H(-) heart (non-activated) H(+) heart (activated) Hb hemoglobin HPF high powered field(s) Hz Hertz (cycle(s) per second) I ICAM intercellular adhesion molecule 146 ICAM-1 ID ig igG IL IL-1 IL-6 IM IU IV K kD kg L I L L(-) L(+) LAD LAL LV LVEDP LVESP intercellular adhesion molecule one inner diameter immunoglobulin immunoglobulin G interleukin interleukin-one interleukin-six intramuscular international unit(s) intravenous kiloDaltons kilogram(s) liter(s) leukocyte(s) leukocyte(s) (non-activated) leukocyte(s) (activated) left anterior descending (artery) Limulus amebocyte lysate (assay) left ventricle(ular) left ventricular end-diastolic pressure left ventricular end-systolic pressure 147 LYM lymphocyte(s) M MFI mg min ml mmHg mmol N n NO NS NZW O OD P P PaC02 PAF mean fluorescence intensity milligram(s) minute(s) milliliter(s) millimeter(s) of mercury millimdle(s) number nitric oxide non-significant New Zealand White (rabbit) outer diameter probability arterial partial pressure of carbon dioxide platelet activating factor 148 Pa02 arterial partial pressure of oxygen PBS phosphate buffered saline PC02 partial pressure of carbon dioxide pH power of hydrogen PMN polymorphonuclear neutrophil P02 partial pressure of oxygen R RBC red blood cell(s) S s second(s) SD standard deviation SE standard error SEM standard error of the mean S02 saturation of oxygen Sr strontium T Tc technetium TNF-a tumor necrosis factor-alpha TOT total leukocytes 149 u U unit(s) V vol volume vs versus W WBC wt white blood cell(s) weight 

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
https://iiif.library.ubc.ca/presentation/dsp.831.1-0088672/manifest

Comment

Related Items