\\\l\l\l\i’i§\3\\‘\\ll\7\“\l\‘\\\”Will 3 12 0591 5361 .. " - -v .. Jemb‘-3‘~ .f .. THE-:91! =‘2 3 E3 l_'.'.‘l "\ " a ‘1 t. f. '3’ :11: u ’ I f" a - 1 - an .4. fs. F J -_. ~ ' ,5 - "b q -‘.> (.5!- b 3'. - at I" “ r E 'A_; .4: a J: a")? Nair: AMI—Stunz'x. m- w This is to certify that the dissertation entitled I. Effects Of Brain Or Carotid Body Hypoxia And Hypercapnia On Pulmonary Hemodynamics II. Effects Of Histamine On Lung Water And Hemodynamics Before And After Adrenergic Presented by Blockade Neil C. Olson has been accepted towards fulfillment of the requirements for PhD degree in Physiologx fl \ Of Major prolessor Date March 11, 1982 MS U is an Affirmative Action/Equal Opportunity Institution 0- 12771 MSU * LIBRARIES RETURNING MATERIALS: Place in book drop to remove this checkout from your record. FINES will be charged if book is returned after the date stamped below. I. EFFECTS OF BRAIN 0R CAROTID BODY HYPOXIA AND HYPERCAPNIA 0N PULMONARY HEMODYNAMICS II. EFFECTS OF HISTAMINE ON LUNG WATER AND HEMODYNAMICS BEFORE AND AFTER ADRENERGIC BLOCKADE By Neil C. Olson A DISSERTATION Submitted to Michigan State University in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY Department of Physiology 1982 ABSTRACT 1. EFFECTS OF BRAIN OR CAROTID BODY HYPOXIA AND HYPERCAPNIA ON PULMONARY HEMODYNAMICS II. EFFECTS OF HISTAMINE ON LUNG WATER AND HEMODYNAMICS BEFORE AND AFTER ADRENERGIC BLOCKADE By Neil C. Olson There are no reports in the literature of the effects of local brain hypoxia and hypercapnia on pulmonary hemodynamics and reports con- cerning the role of the carotid bodies are controversial. Therefore, a better understanding of pulmonary vascular control by the brain and carotid bodies is needed. This is particularly true during pathophysio- logical states such as the adult respiratory distress syndrome. Thus, the main purpose of this investigation was to evaluate the effects of local brain or carotid body hypoxia, hypercapnia, and hypoxia-hypercapnia on pulmonary hemodynamics and attempt to relate these findings to the adult respiratory distress syndrome. Numerous etiologies have been proposed for the adult respiratory distress syndrome including brain hypoxia and release of humoral agents. In the latter, histamine is a likely candidate. Therefore, another purpose of this investigation was to evaluate the effects of histamine on lung water and hemodynamics both before and after adrenergic blockade. All experiments were performed on anesthetized dogs. Following ligation of collateral vessels (supplying blood to the brain), brain Neil C. Olson hypoxia or hypercapnia or both were induced by pumping arterial auto- logous blood through an extracorporeal lung to the external carotid arteries for 5 min or 2 hours. Five min hypoxia increased cardiac output, central blood volume, total peripheral resistance, mean pulmonary artery, left atrial, pulmonary artery pulse, and mean aortic pressures. However, these effects were not maintained during 2 hours hypoxia. Pulmonary vascular resistance and lung extravascular thermal volume were unchanged during 5 min and 2 hours of brain hypoxia. I conclude that 5 min of brain hypoxia and/or hypercapnia does not increase pulmonary vascular resistance but may increase microvascular hydrostatic pressure secondary to increased pulmonary blood volume and in this way could cause pulmonary edema and alveolar hemorrhage similar to that observed in the adult respiratory distress syndrome. Failure of these effects to be maintained for 2 hours may reflect deterioration of the preparation. The carotid bodies were perfused bilaterally with hypoxic and/or hypercapnic blood by pumping arterial autologous blood through an extra- corporeal lung to the common carotid arteries. Carotid body hypoxia decreased cardiac output while pulmonary vascular resistance, mean aortic pressure and total peripheral resistance increased. During hypoxia- hypercapnia mean pulmonary artery and aortic pressures increased as did pulmonary vascular and total peripheral resistances. Following alpha blockade with phentolamine pulmonary vascular resistance failed to increase during carotid body hypoxia or hypoxia-hypercapnia. I conclude that carotid body hypoxia and hypoxia-hypercapnia reflexly increase pulmonary vascular resistance and left ventricular afterload, that these Neil C. Olson effects are mediated by alpha adrenergic receptors, and that carotid body stimulation could potentiate the effects of brain hypoxia on pulmonary hemodynamics and thus increase lung water. The effects of 90 min intravenous histamine with and without alpha (phentolamine) or beta (propranolol) receptor blockade on lung water and hemodynamics were also studied. Histamine with and without alpha blockade decreased central blood volume, pulmonary artery wedge, and mean aortic pressures; pulmonary vascular resistance increased while lung extravascular thermal volume was unchanged. Histamine with propranolol decreased mean aortic pressure, cardiac index, and central blood volume. Pulmonary vascular resistance increased and lung extravascular thermal volume was unchanged. In all experiments postmortem extravascular lung water to extravascular dry weight ratio was unchanged from control values. I con- clude that histamine has little or no effect on lung microvascular perme- ability, does not increase lung water, and that these responses are not modified by reflex beta or alpha receptor stimulation on the microvascular membrane. Thus, histamine cannot be considered a likely mediator of the massive increase in lung water and permeability that occur in the adult respiratory distress syndrome. DEDICATION With love to Peggy, Jennifer, and Stephanie ii ACKNOWLEDGMENTS The author wishes to express his sincere appreciation to his Major Professor, Dr. N. Edward Robinson, for his enthusiastic support, constructive criticism and continual encouragement throughout his surgical residency and doctoral program. The author's gratitude is also extended to the members of his Guidance Committee, Dr. R. A. Bernard, Dr. N. R. Dukelow, Dr. T. E. Emerson, Jr., Dr. G. E. Eyster, and Dr. J. B. Scott. This investigation was performed in the laboratory of Dr. Jerry B. Scott, whose recent and untimely death saddened me greatly. I am extremely grateful to Dr. Scott for the enthusiastic and generous sup- port he gave me in all aspects of my graduate training. I feel very honored to have had the privilege to work with such a great scientist, scholar, teacher, writer, and above all, a marvelous human being. The author also expresses appreciation for technical assistance provided by Mr. Donald Anderson, Ms. Roberta Milar, and Mr. William Stoffs and fbr computer programs provided by Mr. Paul Sorenson. This investigation was supported by National Institutes of Health Grants lF32HL05923, lT32HLO7404, and HLl0879. iii TABLE OF CONTENTS Page LIST OF TABLES ................................................... viii LIST OF FIGURES .................................................. xi INTRODUCTION ..................................................... l LITERATURE REVIEW ................................................ 4 1. Functional Anatomy ........................................ 4 Pulmonary Vessels ....................................... 4 Arteries, Capillaries, and Veins ...................... 4 Supernumerary Vessels ................................. 6 Bronchial Vessels ....................................... 6 Origin, Drainage, and Anastomoses ..................... 6 Innervation and Receptors ............................... 7 Nerve Pathways ........................................ 7 Alpha and Beta Adrenergic Receptors ................... 8 Cholinergic Receptors ................................. 8 Lymphatics and Lymph Nodes .............................. 9 Origin and Drainage ................................... 9 Lung Ultrastructure ..................................... l0 Alveolar Epithelium ................................... l0 Vascular Endothelium .................................. ll Interstitium .......................................... ll Alveolar-Capillary Membrane ........................... l2 II. Pulmonary Hemodynamics .................................... l3 Interrelationship of Pressure, Flow, Volume, Distensi- bility and Resistance ................................. 13 III. Factors Controlling Pulmonary Vascular Resistance and Compliance .............................................. l8 Passive Factors ......................................... l8 Ventilation ........................................... lB Lung Volume ........................................... l9 iv TABLE OF CONTENTS--continued Page Gravity ............................................... l8 Hydrostatic Pressure .................................. 19 Transmural Pressure ................................... l9 Interdependence ....................................... l9 Active Factors .......................................... 20 Local Control ......................................... 20 Remote Control ........................................ 20 Nerve Stimulation ....................................... 20 Peripheral Baroreceptors and Chemoreceptors ............. 23 Brain ................................................... 24 Humoral Agents .......................................... 25 IV. Lung Fluid Balance ........................................ 25 Starling Forces ......................................... 25 Pathways for Fluid Movement ............................. 27 V. Pulmonary Edema ........................................... 27 Types and Etiologies .................................... 27 High Pressure ......................................... 27 Permeability .......................................... 28 Sites of Fluid Leakage and Accumulation ................. 28 Alveolar Vessels ...................................... 28 Extra-Alveolar Vessels ................................ 28 Safety Factors .......................................... 30 Interstitial Compliance ............................... 30 Colloid Osmotic Pressure .............................. 30 Lymph Flow ............................................ 30 Quantitation of Lung Water .............................. 30 Antemortem ............................................ 30 Postmortem ............................................ 31 VI. Brain and Pulmonary Edema ................................. 32 Stimuli ............................................... 32 Adrenergic Blockade ................................... 33 VII. Adult Respiratory Distress Syndrome ....................... 34 Definition and Incidence ................................ 34 Etiologic Factors ....................................... 36 Pathophysiology ......................................... 36 Centrineurogenic Hypothesis.... ......................... 38 Preparation ........................................... 38 Circumstantial Evidence ............................... 39 Thromboembolism and Humoral Agents ...................... 41 VIII. Histamine, Microvascular Permeability and Adrenergic Receptors ............................................... 43 TABLE OF CONTENTS--continued Page STATEMENT OF OBJECTIVES ......................................... 47 METHODS ......................................................... 49 I. General .................................................. 49 II. Brain and Carotid Body Studies ........................... 50 III. Brain Hypoxia and Hypercapnia ............................ 51 Thermal-Conductivity Method ............................ 59 Equations for Cardiac Output and Mean Transit Time... 61 Analysis of Lung Composition ......................... 65 IV. Carotid Body Hypoxia and Hypercapnia ..................... 68 V. Histamine and Adrenergic Receptors ....................... 72 RESULTS ......................................................... 75 1. Brain Hypoxia and Hypercapnia ............................ 75 Vagus and Carotid Sinus Nerves Intact .................. 75 Vagus and Carotid Sinus Nerves Cut ..................... 89 II. Carotid Body Hypoxia and Hypercapnia ..................... 108 Vagus Nerves Intact .................................... 108 Vagus Nerves Cut ....................................... 108 III. Histamine and Adrenergic Receptors ....................... 127 Time Control Group ..................................... 127 Histamine Group ........................................ 127 Propranolol-Histamine Group ............................ 127 Phentolamine-Histamine Group ........................... 128 Pulmonary Extravascular Tissue Weight and Lung Extra- vascular Thermal Volume .............................. 138 Extravascular Lung Water to Extravascular Dry Weight Ratio ...... . ......................................... 143 DISCUSSION ............................................ L ......... 144 1. Brain Hypoxia and Hypercapnia ............................ 144 Experimental Preparation ............................... 144 Intact Nerve Group ..................................... 146 Cut Nerve Group (5 Minutes Brain Hypoxia) .............. 146 Differences Between Intact Nerve and Cut Nerve Groups.. 148 Brain Hypoxia and Pulmonary Edema ................ . ..... 149 Adult Respiratory Distress Syndrome .................... 152 vi TABLE OF CONTENTS--continued Page II. Carotid Body Hypoxia and Hypercapnia ..................... 155 Adult Respiratory Distress Syndrome .................... 160 III. Histamine and Adrenergic Receptors ....................... 161 Hemodynamics ........................................... 164 Histamine ............................................ 164 Propranolol and Histamine ............................ 166 Phentolamine and Histamine ........................... 168 Lung Mechanics... ...................................... 169 Adult Respiratory Distress Syndrome .................... 169 SUMMARY AND CONCLUSIONS ......................................... 171 RECOMMENDATIONS ................................................. 174 APPENDIX ........................................................ 176 LIST OF REFERENCES .............................................. 178 vii LIST OF TABLES TABLE Page 1. Cardiovascular effects of perfusing the brain with severely hypoxemic blood under constant pressure conditions (head perfusion pressure = 111 i 2 mmHg) with the vagus and carotid sinus nerves intact ................................ 77 2. Cardiovascular effects of perfusing the brain with mildly hypoxemic blood under constant flow conditions (head flow = 179 i 22 m1/min) with the vagus and carotid sinus nerves intact ..................................................... 79 3. Cardiovascular effects of perfusing the brain with moderately hypoxemic blood under constant flow conditions (head flow = 194 i 22 ml/min) with the vagus and carotid sinus nerves intact ........................................ 81 4. Cardiovascular effects of perfusing the brain with severely hypoxemic blood under constant flow conditions (head flow = 238 i 25 m1/min) with the vagus and carotid sinus nerves intact ..................................................... 83 5. Cardiovascular effects of perfusing the brain with hyper- capnic blood under constant flow conditions (head flow = 218 i 19 ml/min) with the vagus and carotid sinus nerves intact ..................................................... 85 6. Cardiovascular effects of perfusing the brain with hypoxemic-hypercapnic blood under constant flow conditions (head flow = 271 i 26 m1/min) with the vagus and carotid sinus nerves intact ........................................ 87 7. Cardiovascular effects of perfusing the brain with hyper~ capnic blood under constant flow conditions (head flow = 275 i 19 ml/min) with the vagus and carotid sinus nerves 1 cut .................... . ................................... 02 8. Cardiovascular effects of perfusing the brain with hypoxemic-hypercapnic blood under constant flow conditions (head flow = 289 i 19 m1/min) with the vagus and carotid sinus nerves cut ........................................... 104 viii LIST OF TABLES--continued TABLE 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. . Cardiovascular effects of perfusing the brain for two hours with severely hypoxemic blood under constant flow condi- tions (head flow = 298 i 18 m1/min) with the vagus and carotid sinus nerves cut ................................... Cardiovascular effects of perfusing the carotid bodies with severely hypoxemic blood, before vagotomy, under constant flow and pressure conditions ...................... Cardiovascular effects of perfusing the carotid bodies with mildly hypoxemic blood, following vagotomy, under constant flow and pressure conditions ...................... Cardiovascular effects of perfusing the carotid bodies with moderately hypoxemic blood, following vagotomy, under constant flow and pressure conditions..... ................. Cardiovascular effects of perfusing the carotid bodies with severely hypoxemic blood, following vagotomy, under constant flow and pressure conditions ...................... Cardiovascular effects of perfusing the carotid bodies with hypercapnic blood, following vagotomy, under constant flow and pressure conditions ............................... Cardiovascular effects of perfusing the carotid bodies with hypoxemic-hypercapnic blood, following vagotomy, under constant flow and pressure conditions ................ Cardiovascular effects of perfusing the carotid bodies with severely hypoxemic blood, following vagotomy and alpha blockade, under constant flow and pressure conditions ...... Cardiovascular effects of perfusing the carotid bodies with hypoxemic-hypercapnic blood, following vagotomy and alpha blockade, under constant flow and pressure conditions Time control group in which no vasoactive drugs were administered ............................................... Effects of intravenous histamine on the cardiovascular system.... ................................................. ix Page 106 111 113 115 117 119 121 123 125 130 132 LIST OF TABLES--continued TABLE Page 20. Effects of intravenous propranolol and histamine on the cardiovascular system ...................................... 134 21. Effects of intravenous phentolamine and histamine on the cardiovascular system..... ................................. 136 LIST OF FIGURES FIGURE Page 1. Head perfusion preparation ................................ 52 2. Tracing of the cardiovascular response following cessation of flow through the external carotid arteries ............. 56 3. Tracing of thermal-dilution (top) and conductivity- dilution (bottom) curves .................................. 63 4. Linear regression of lung extravascular thermal volume (ETV ) versus pulmonary extravascular tissue weight (PETN) from dogs in which left atrial pressure was increased to product various degrees of pulmonary edema ................ 66 5. Carotid body perfusion preparation ........................ 70 6. Effect of ventilating the extracorporeal lung with a control gas mixture containing 15% 0 5% C0, 80% N and a hypoxic mixture containing 0% 02 5%C0 5% N perfusate gas tensions and pH and2 head perfusion ressure PHP) ..................................................... 7. Effect of perfusing the brain with severely hypoxemic blood on cardiac output (0), left ventricular contractil- ity (dP/dt), heart rate (HR), and stroke volume (VS) ...... 92 9O 8. Effect of perfusing the brain with severely hypoxemic blood on left atrial pressure (P ), pulmonary artery wedge pressure (P ), mean pulmBBary artery pressure H) pulmonary gagery pulse pressure (PPAP), and mean aoFBic pressure (PMA ) ..................................... 94 9. Effect of perfusing the brain with severely hypoxemic blood on total pulmonary vascular resistance (PVR ), pre- large vein pulmonary vascular resistance (PVRP), large vein pulmonary vascular resistance (PVRv ) and total peripheral resistance (TPR) ............................... 96 xi LIST OF FIGURES FIGURE 10. Effect of perfusing the brain with severely hypoxemic blood on lung extravascular thermal volume (ETV ), central blood volume (CBV), and extravascular lbng water to extravascular dry weight ratio (EVLN/EVDN) ............. 11. Linear regression of lung extravascular thermal volume (ETVL) versus pulmonary extravascular tissue weight (PETN) from dogs in the time control group ....................... 12. Linear regression of lung extravascular thermal volume (ETVL) versus pulmonary extravascular tissue weight (PETN) in dogs for the combined histamine, propranolol-histamine, and phentolamine-histamine groups ......................... Al. Diagram showing relationship of the pulmonary artery wedge position to upstream (e.g., small artery) and downstream (e.g., left atrium) locations ............................. xii Page 98 139 141 176 INTRODUCTION In the last two decades a life-threatening pulmonary complication has emerged. The condition, which usually follows hemorrhagic shock and trauma, has been given a wide variety of names including the adult respiratory distress syndrome, shock lung, and capillary leak syndrome (19). That the etiology remains an enigma is evidenced by the numerous hypotheses proposed. One hypothesis, proposed by Moss and Stein (167), is that severe hemorrhagic shock may cause brain hypoxia thus impairing oxidative metabolism in the hypothalamus leading to increased sympathetic outflow to postcapillary vessels and increasing pulmonary vascular resist- ance. Increased postcapillary tone would increase microvascular hydro- static pressure and cause pulmonary congestion, edema, hemorrhage, and hypoxemia (i.e., adult respiratory distress syndrome). There is considerable circumstantial evidence in the literature which suggests that the brain has the ability to increase pulmonary vascular resistance during systemic hypoxemia and/or hypercapnia and that this response may be regulated by the autonomic nervous system. However, no reports exist describing these specific effects. This research is designed to quantitate the change in pulmonary vascular resistance during periods of brain hypoxia, hypercapnia, and hypoxia-hypercapnia and to attempt to elucidate the mechanisms involved. This investigation is of considerable significance because if hypoxic and/or hypercapnic stimulation of the brain (either directly or reflexly through stimulation of peripheral Chemoreceptors) increases pulmonary vascular resistance then neuropharmacologic agents with actions on the brain (e.g., hypothalamus) may become an important modality in the prevention or treatment of pulmonary edema and alveolar hemorrhage during periods of systemic hypoxia and/or hypercapnia. The primary objectives of this investigation were to determine the effects of local brain or carotid body hypoxia and/or hypercapnia on pulmonary vascular resistance both before and after carotid sinus nerve section and cervical vagotomy. Secondary objectives were to localize the site of resistance response, by use of an alpha adrenergic blocker (phentolamine) determine if autonomic mechanisms cause these changes, and measurement of lung extravascular thermal volume and central blood volume during brain hypoxia. Another objective of this investigation was to evaluate the effects of histamine on lung water and hemodynamics with and without alpha or beta receptor blockade. This study was performed for 2 reasons. First, histamine is one of several humoral agents that may be involved in medi- ating the adult respiratory distress syndrome and second, the effect of histamine on lung water during alpha or beta receptor blockade has not been reported. The first several sections of the literature review contain material relevant to functional anatomy, pulmonary hemodynamics, lung fluid balance and pulmonary edema. Subsequent sections present background information on the adult respiratory distress syndrome with particular emphasis on the brain's ability to alter pulmonary hemodynamics and cause pulmonary edema and alveolar hemorrhage. The centrineurogenic hypothesis proposed by Moss and Stein (167) is considered in detail. In addition, histamine is discussed as it relates to the adult respiratory distress syndrome and vascular permeability in general. LITERATURE REVIEW I. Functional Anatomy Pulmonary Vessels The pulmonary circulation originates at the pulmonary semilunar valves and extends to the junction of the pulmonary veins and left atrium. Included in this system is a short pulmonary trunk, right and left pulmonary arteries and their lobar branches, arterioles, capillaries, venules, small and large pulmonary veins (115). In the adult human being three types of pulmonary arteries are described based upon the histology of the arterial wall (115). The wall of elastic arteries, which are greater than 1000 um external diameter, consist predominantly of elastic fibrils with some muscle fibers and collagen. Examples of this type of artery include the pulmonary trunk, right and left pulmonary arteries and their lobar branches. The wall of muscular arteries, whose external diameter is between 100 and 1000 um, consist of smooth muscle fibers interposed between internal and external elastic laminae. The muscular pulmonary arteries lie close to the bronchioles, respiratory bronchioles and alveolar ducts. The wall of arterioles, which include precapillary vessels less than 100 um external diameter, initially has a thin layer of smooth muscle that subsequently disappears so that the wall consists solely of an endothelial lining and a single elastic lamina. Pulmonary arterioles supply blood to alveolar ducts and alveoli. Precapillary vessels accompany the dichotomous branching of the airways to the level of the terminal bronchioles and from this point divide extensively to form the capillary bed (242). The pulmonary capil- laries are lined by an endothelial layer and surrounded by alveoli lined with epithelium. This capillary network is so dense that it is regarded by some as a sheet of flowing blood interrupted in places by "posts" much like an underground parking garage (242). Pulmonary venules are postcapillary vessels less than 100 um external diameter with a histologic structure identical to arterioles. They are formed in the area of the bronchioles and pass into the connective tissue septa between secondary lobules to enter pulmonary veins. The media of pulmonary veins is somewhat irregular in thickness consisting of oblique, circularly arranged smooth muscle fibers and collagen while the adventitia is thick and fibrous. Near the venoatrial junction cardiac muscle appears in the walls of the pulmonary veins (115). Von Hayek (116) has observed cardiac muscle fibers accompanying pulmonary veins as far back as the hilum. The degree of muscularization of pulmonary vessels as well as their distribution within the lung is quite species dependent (85,158). The pulmonary arteries in man, dog and rabbit contain more smooth muscle than veins of comparable size. However, in other species such as the guinea-pig, cat, pig and calf, both arteries and veins have abundant smooth muscle (127). Only in rabbits and cattle does the pulmo- nary circulation contain thick walled precapillary vessels reminiscent of systemic arterioles (85). McLaughlin and co-workers (158) found large variation in the distribution of smooth muscle within pulmonary vessels of several species of animals. They urged extreme caution when comparing findings from other species with human beings since equally marked physio- logic, pathologic and biochemical differences may also exist. It is interesting that of the species examined (i.e., cattle, sheep, pig, dog, cat, monkey, horse, rat, rabbit and guinea-pig), the pulmonary vascular architecture in the horse most closely resembled that found in human- beings (158). The pulmonary arteries also give rise to branches which do not accompany the dichotomous branching of the airways. These are called supernumerary arteries which actually outnumber the conventional ones, providing 25 to 40% of the total cross-sectional area of the pulmonary arterial bed (170). They serve as a collateral source of blood to the terminal respiratory units (i.e., structures distal to and including respiratory bronchioles), particularly following obstruction of conven- tional pulmonary arteries (192). Bronchial Vessels Whereas pulmonary arteries deliver mixed venous blood to the lung at low vascular pressures, the bronchial arteries deliver systemic arterial blood at high vascular pressures. Bronchial arteries vary in number and origin. Three arteries are usually found, one for the right lung and two for the left lung, originating from the thoracic aorta (48) or one of its major branches (161). Bronchial arteries supply blood to the bronchial tree, nervous, lymphoid, and connective tissues within the lung (32,161). The bronchial arteries also supply the pleura of some species, including human beings (161), cattle, sheep, and swine (158). The existence of direct anastomotic connections between pulmonary arteries and bronchial arteries is controversial and may be species dependent (158). Despite this controversy, however, it is generally accepted that anastomoses between bronchial and pulmonary vessels may occur in the precapillary, capillary and postcapillary pulmonary bed (170). Approximately 30% of the bronchial arterial supply is returned through the bronchial, azygos, or intercostal veins into the right atrium. The remaining 70% is returned to the left atrium through the pulmonary veins and thus constitutes a right to left Shunt (170). Although bronchial flow is only approximately 1% of the cardiac output in the normal lung (32), considerable increases in flow may occur during such maladies as neoplasia, congenital anomalies and inflammation (170). Innervation and Receptors The pulmonary circulation is innervated by the sympathetic and parasympathetic nervous systems(115,121,194). In the parasympathetic system, preganglionic efferent fibers have their origin in the dorsal vagal nuclei within the medulla. These fibers descend in the vagus nerve and synapse with ganglia located around blood vessels. Short post- ganglionic fibers then innervate vascular smooth muscle (115,194). Sympathetic preganglionic efferent fibers traveling to the lung originate in the intermediolateral horns of thoracic spinal cord segments two through six. They synapse with cell bodies of postganglionic fibers in the second to sixth spinal ganglia, middle cervical, or stellate ganglia (115). Postganglionic sympathetic fibers emanating from these ganglia and the vagus nerves contribute to the formation of an extrapulmonary (i.e., anterior and posterior pulmonary plexuses) plexus at the roots of the lungs (170). The extrapulmonary plexuses are continued within the lung parenchyma as two main intrapulmonary plexuses, one along conducting airways and the other along the pulmonary artery and its branches (161). The sympathetic postganglionic vasoconstrictor fibers are adrenergic and the parasympathetic vasodilator fibers are Cholinergic. Murray (170) and Fishman (83) consider these vasodilator fibers to be physiologically unimportant in adult human beings. It should be emphasized that tremendous variability in the innerva- tion of the mammalian lung exists between species (117,194). Extrapolation of physiologic responses or anatomic distribution of nerves from one species to another must therefore be done with considerable caution (194). Despite species variability, some generalizations can be made regarding the pattern of innervation of pulmonary vessels (66,85). 1) Both sympa- thetic and parasympathetic fibers are found in close association with pulmonary vessels. 2) The arteries have a more widespread and complete innervation than the veins. 3) The pulmonary vascular innervation in most species is relatively sparse when compared with bronchial or other systemic arteries. 4) The density of nerve fibers is greatest in the larger elastic vessels, less in muscular arteries, and absent in vessels less than 30 pm. 5) Both alpha and beta adrenergic receptors are present; however the alpha receptors predominate (17,229,232). 6) The small intrapulmonary veins are minimally supplied with nerves in comparison to the small precapillary vessels. The calf appears to be an exception because the small veins are more profusely innervated than the smaller arteries (117). Lymphatics and Lymph Nodes The lungs are provided with an abundant supply of lymphatics surpass- ing the number found in several other organs including the liver, kidney, and spleen (161). Grossly, the lymphatics can be divided into a super- ficial and deep network. The superficial network supplies the visceral pleura and the deep network accompanies the cartilagenous airways as well as branches of the pulmonary arteries and veins (161). The two networks anastomose in the pleura and at the hilum (115). Histologically, the walls of larger lymphatics consist of collagen, elastic fibrils, smooth muscle and an endothelial lining. The smaller lymphatics consist of a fibro-elastic coat and an endothelial lining while lymph capillaries consist only of an endothelium (115). Valves, formed by intimal folds, within the lymphatics, are numerous in the pleura and hilar region but infrequent or absent in the lung parenchyma (161). Lymphatics extend at least to the level of the terminal bronchioles (145) but are not present in the walls of alveolar sacs or alveoli (115,161). Lymph vessels between adjacent terminal respiratory units are called juxta- alveolar lymphatics and these constitute the distal blind-ended tubules of the lymphatic system (145). Pulmonary capillaries within the inter- alveolar septum have been estimated to be between 100 um and 1000 um from the nearest juxta-alveolar lymphatic (170). Lymph is formed by normal lungs (26,237) and proceeds toward the hilum along bronchial, arterial, and venous lymph vessels. Exception to this centripetal pattern occurs in the pleura (145). Lymph from the lung drains into tracheobronchial lymph nodes, thoracic duct, and finally the right atrium. Recently, Parker et al. (179) estimated total 10 lymph flow for the normal dog lung to be approximately 0.07 ml/min/lOO gm wet lung weight. The normal concentration of albumin in lung lymph is approximately 80% and the larger globulins approximately 50% of the plasma albumin and globulin concentration, respectively. The total pro- tein concentration in lung lymph is approximately two-thirds the plasma concentration. This is consistent with the concept that protein molecules are sieved according to their molecular size (26,219). Lung,U1trastructure The alveolar epithelium is continuous with airway epithelium. Three cell types constitute the alveolar epithelium. Type I cells (squamous pneumocytes) consist of a nucleus with long thin cytoplasmic extensions that cover approximately 95% of the total alveolar surface (239). Type II cells (granular pneumocytes) are more numerous than type I cells but because of their cuboidal shape occupy less than 5% of the alveolar surface (170). Unlike type I cells, type II cells undergo mitosis and are rich in mitochondria, endoplasmic reticulum, Golgi apparatus, and osmiophilic lamellated bodies. The latter have been associ- ated with the production and removal of surfactant (238). Type II cells can differentiate into type I cells and appear to be the main cell type involved in the repair of injured alveolar epithelium. Type III cells (alveolar brush cells) occur rarely and may serve as receptor cells, although nerve fibers directly associated with these cells have not been identified (239). Between all alveolar epithelial cells exist zonulae occludentes ("tight junctions"). These epithelial tight junctions are impermeable to macromolecules, including horseradish peroxidase (207). Pinocytotic 11 vesicular transport is the only means by which macromolecules may bypass the intercellular junctions of alveolar epithelium (207). The pulmonary vascular endothelium forms a continuous lining in the arteries, capillaries and veins. Ultrastructurally the pulmonary capillary endothelium is continuous and nonfenestrated resembling that found in skeletal and cardiac muscle (207). Each endothelial cell con- sists of a nucleus and long cytoplasmic extensions. The luminal surface is specifically designed for degradation or transformation of several vasoactive substances that include the adenine nucleotides, bradykinin and angiotensin I. The anatomic site for these metabolic functions is the caveolae intracellulares which are localized invaginations of the luminal surface that communicate with the bloodstream through a single lamellar (i.e., diaphragm) membrane (170). Relative to alveolar epithelium the vascular endothelium is “leaky". Schneeberger (207) has demonstrated that junctional slits between endo- thelial cells are sufficiently wide (4 nm) to allow passage of small pro- tein molecules and horseradish peroxidase from plasma into the inter- stitial space. In all mammals the pulmonary alveolar epithelial and capillary endothelial Ultrastructure is essentially identical. This is not the case however, in the interstitium where species variation is great. The size and number of connective tissue fibers are directly proportional to the size of the animal. For example, the primate and canine lung has considerably more and coarser collagenous and elastic fibers than does the rat or mouse lung (238). 12 Elastic, collagenous, and reticular connective tissue fibers are present in the lung interstitium. These fibrous elements are the support- ing lattice for the entire lung and constitutes the so-called "fibrous continuum of the lung" (138). There are three different constituents of this fibrous continuum, albeit they are all interconnected and act as one functional unit: 1) axial connective tissue, 2) peripheral connec- tive tissue, and 3) parenchymatous connective tissue. The axial connec- tive tissue, originating at the hilum, contains the airways, arteries and deep lymphatics. The peripheral connective tissue is essentially the visceral pleura connective tissue and has the superficial lymphatics embedded within it. The parenchymatous connective tissue lies inside alveolar walls and forms a loose basket around each and every alveolus, thus providing structural support for alveoli as well as capillaries (239). In addition to connective tissue elements occupying the inter- stitium, there are cellular constituents (e.g., mesenchymal and mono- nuclear cells ) and proteoglycan filaments. The latter are 98% hyaluronic acid and 2% protein and fill the spaces between connective tissue fibers and cells. The interstitial fluid normally present is entrapped within these proteoglycan filaments and together they are known as the tissue gel. Free fluid is also present but represents a very small compartment in normal lungs (104). The alveolar-capillary membrane is subdivided into a thin portion and thick portion (170). The thin Side is approximately 0.5 pm and is characterized by absence of a morphologic identifiable interstitial space between the fused alveolar and endothelial basement membranes, even 13 during pulmonary edema (239). This narrow membrane combined with a large surface area (approximately 70 m2 in man) makes it an ideal location for diffusion of gases between blood and alveoli (104). In contrast, the thick side with a variable thickness up to several micrometers, contains a distinct interstitial space between the nonfused alveolar and endo- thelial basement membranes (178). The thick portion provides a support- ing structure as well as the primary site for fluid and solute exchange (86). This portion of the interstitium connects without interruption to the interstitial fluid sumps associated with the connective tissue sheaths surrounding airways and blood vessels (239). II. Pulmonary Hemodynamics Interrelationship of Pressure, Flow, Volume, Distensibility and Resistance The complexity of understanding pulmonary hemodynamics is perhaps best summarized in the fbllowing quote: The investigator seeking to understand the pulmonary circulation is confronted by awesome problems: a system of collapsible tubes, of variable number, parts of which narrow while others expand with lung inflation, suspended in an air-filled organ and perfused in a pulsatile fashion with a fluid whose viscosity changes with flow rate, the whole system being subjected to cyclical changes of pressure and volume during breathing (49). Pressure Relative to pressures in the systemic circulation, pressures in the pulmonary circulation are very low. Pulmonary artery systolic and diastolic pressures are about 25 and 8 mmHg, respectively, with a mean pressure of approximately 15 mmHg. Since the left atrial pressure is 14 about 5 mmHg, the vascular pressure gradient across the lung (i.e., pulmonary perfusion pressure) is approximately 10 mmHg (242). Pulmonary artery wedge pressure represents the pressure at the first postcapillary collateral venous vessel (82) and thus is slightly greater than left atrial pressure and less than capillary pressure. In isolated dog lungs the average isogravimetric capillary pressure is 7 (88) to 9 (1) mmHg. Vascular transmural (or distending) pressure, which represents the differ- ence between the intravascular and extravascular pressure is an important concept in pulmonary hemodynamics because even small changes in trans- mural pressure greatly affect vascular caliber (82). The extravascular pressures are subject to continuous fluctuations in the lung due to changes in intrapleural pressure during ventilation (82). Under static conditions alveolar pressure is zero (i.e., atmospheric) and therefore the capillary transmural pressure is the intravascular pressure. Pressure surrounding the parenchymal pulmonary arteries and veins (i.e., extra- alveolar vessels) is considerably less than alveolar pressure. This occurs because as the lung expands the blood vessels are pulled open by radial traction of the lung parenchyma. Since the vessels are relatively more rigid than the surrounding elastic parenchyma, a substantial negative pressure may develop around these vessels thus increasing transmural pressure (185,242). f19w_ Blood flow through the pulmonary circulation is pulsatile (164) and, at least in man, is unevenly distributed. The latter occurs because the low pressure pulmonary circulation is operating in a gravitational field (170). In such a system a vertical distribution of flow exists because 15 of increasing transmural pressure and decreasing vascular resistance from the top to the bottom of the lung. Using radioactive xenon West (242) demonstrated this phenomenon and subsequently divided the lung into 3 zones according to the magnitude of pressures in the pulmonary artery, alveolus, and pulmonary vein. At the apex of the lung (dorsum in quadrapeds) is zone 1 where alveolar pressure exceeds arterial pressure, thus collapsing alveolar capillaries and precluding flow. This zone is thought not to exist under normal conditions. Furthermore, even if alveolar pressure exceeds pulmonary artery pressure, flow through zone 1 is possible through corner vessels. These vessels are located in the corners of 3 alveoli and are not subject to alveolar pressure. Because of surface tension forces at the alveolus, large retracting forces tend to pull these vessels open allowing blood to flow through zone 1 (239). In zone 2 pulmonary arterial pressure exceeds alveolar pressure and both of these exceed venous pressure. The alveolar vessels are essentially collapsible tubes surrounded by a chamber of higher pressure, analogous to a Starling resistor. Under zone 2 conditions blood flow is determined by the arterial-alveolar pressure gradient, rather than the classical arterial-venous pressure difference, and therefore flow is unaffected by the pressure distal to the collapse point until this exceeds alveolar pressure. This phenomenon is known as the sluice or waterfall effect because flow over a waterfall is unaffected by the height of the drop beyond (12). Vascular pressure increases about 1 cm water/cm vertical distance down the lung. Thus, the pressure gradient (i.e., pulmonary artery minus alveolar pressure) increases, resistance decreases, and flow increases down zone 2. The decreasing resistance is due to recruitment 16 and distension of vessels resulting from the increasing transmural pres- sure (242). In zone 3 venous pressure exceeds alveolar pressure causing distention of alveolar vessels and flow once again is determined by the arterial-venous pressure gradient. Flow increases down this zone, albeit less dramatically than in zone 2. This increased flow cannot be explained by an increased pressure gradient since all vascular pressures increase equally down zone 3. However, transmural pressure (across alveolar vessels) does increase and this distends and recruits vessels thus decreasing vascular resistance and increasing conductance (242). At the extreme base of the lung a zone 4 has been described (125) in which blood flow is reduced relative to zone 3. Zone 4 is small at high lung volumes and large at low lung volumes suggesting that vascular resistance is affected by lung retractive forces surrounding extra- alveolar vessels. Increased interstitial pressure has also been suggested as a cause for increased resistance and decreased flow in this zone (170). We. The pulmonary vasculature is a distensible reservoir for blood with the veins holding about 50%, arteries 30%, and capillaries 20% of the pulmonary blood volume (251). The volume of blood within the lungs is determined passively by the balance between inflow from the right heart and outflow from the left heart. Normally pulmonary blood volume is approximately 10% of total blood volume while blood in the heart and lungs (i.e., central blood volume) is 20 to 25% of total blood volume (82). 17 Distensibility (Compliance) Pulmonary vascular distensibility, expressed as the change in vascular volume per unit change in transmural pressure is determined by the elastic properties of the vasculature, smooth muscle tone, peri- vascular air and tissue pressures, alveolar surface tension forces, and mechanical distortions associated with surrounding elastic tissue that is rapidly expanding (82). Despite these complex considerations, several generalizations can be made regarding pulmonary vascular distensibility (12,82,127,l40). l) The pressure-volume characteristics of the entire pulmonary vascular tree resemble those of a large systemic vein (i.e., highly distensible at low transmural pressure but relatively indistensible at high transmural pressure). 2) The pulmonary veins are essentially rigid tubes when left atrial pressure exceeds 15 mmHg. 3) The vascular pressure at a given volume is higher when pulmonary vessels are being filled than when they are emptying. 4) The pulmonary venous-left atrial segment is less distensible than the systemic venous-right atrial segment. 5) Approximately 60% of the distensibility of the entire pulmonary vascu- lar bed lies in the veins, 25% in arteries and 15% in capillaries. 6) In contrast to systemic arterioles (which are relatively rigid), pulmonary arterioles are thin-walled and easily distensible and therefore contribute to the pressure-volume characteristics of the pulmonary circu- lation. Resistance Poiseuille's Law states that resistance to fluid flowing through cylindrical tubes is inversely proportional to the fourth power of the tubes radius and directly proportional to fluid viscosity and tube length. 18 It assumes laminar flow of Newtonian fluid (i.e., constant viscosity) through rigid tubes. To varying degrees these assumptions are not met in the pulmonary circulation and therefore the relationship between pressure and flow is not a linear one (82,127). For example, blood flow through the pulmonary circulation is not laminar nor does blood maintain a constant viscosity as it passes through the microcirculation. With increasing flow or transmural pressure, pulmonary vascular resistance initially falls rapidly and then more slowly. This occurs because increas- ing transmural pressure easily distends all segments of the pulmonary vascular bed as well as recruiting previously collapsed segments. When all the vessels become recruited and maximally distended, pressure then increases linearly with flow and resistance remains constant (127). In contrast to the systemic circulation where resistance to flow is pre- dominantly in the arterioles, resistance through the pulmonary circulation is more longitudinally distributed. Brody et al. (29) found that pul- monary arteries contributed 46%, capillaries 34% and veins 20% of the total lobar vascular resistance in the isolated dog lung. III. Factors Controlling Pulmonary Vascular Resistance and Compliance Passive Factors Factors controlling pulmonary vascular resistance can be passive or active. Passive changes in resistance are secondary to changing mechanical conditions within the lung or systemic circulation. Control of resistance in the normal pulmonary circulation is dominated by mechani- cal factors, especially the effects of ventilation and gravity (12,85). 19 Active changes in pulmonary vascular resistance imply changes in vessel caliber due to contraction or relaxation of vascular smooth muscle in response to neural or humoral stimuli that are under local or remote control. Pulmonary vascular resistance is least at functional residual capacity, vascular resistance increasing when lung volume increases or decreases. At very high lung volumes alveolar vessels are stretched longitudinally and at low lung volumes extra-alveolar vessels become kinked resulting in greater resistance to flow (170). Anything which increases transmural pressure in pulmonary vessels decreases pulmonary vascular resistance. Elevation of left atrial and/or pulmonary artery pressures decreases pulmonary vascular resistance, an effect that is quantitatively greater when initial vascular pressures are low (24). Another consideration is interstitial pressure. At least 3 factors contribute to negative interstitial pressure: 1) pleural pres- sure, 2) tendency for alveoli to collapse thus pulling on corner vessels, and 3) retractive forces pulling open pulmonary vessels as the surrounding lung parenchyma expands (i.e., interdependence). Similarly, an increase in intravascular blood volume will distend other vessels and lung parenchyma causing a decrease in pulmonary vascular resistance. Decreased blood viscosity may also decrease pulmonary vascular resistance. Recog- nizing these passive components is important because interpretation of active vasoconstriction or vasodilation is difficult in the presence of passive changes in vessel caliber. The only exception is when the vasomotor responses are opposite in direction and predominate over con- comitant passive changes (24,170). 20 Active Factors Much of the active pulmonary vascular response to alveolar hypoxia or hypercapnia is mediated locally within the lungs (83). Local alveolar hypoxia (14,123,135,136,l43), and to a lesser extent hypercapnia (14,23, 85), increase pulmonary vascular resistance by constricting primarily the small arteries (83). Pulmonary veins have also been reported to constrict in response to alveolar hypoxia (163,190). That these responses are mediated locally is supported by the fact that they occur in isolated lungs devoid of autonomic innervation, in sympathectomized animals, and in animals following administration of adrenergic blocking agents (82). Although the precise mechanism for this local response is unclear, it is thought that alveolar hypoxia or hypercapnia may directly interfere with excitation-contraction coupling of vascular smooth muscle or cause its effect indirectly through release of intrinsic chemical mediators (83). The most likely candidates for the latter are catecholamines, histamine, prostaglandins and angiotensin (83). Nerve Stimulation Although the pulmonary vasculature has an abundant sympathetic innervation (66,98,117,132), the functional role of the sympathetic nervous system in regulating the pulmonary vascular bed remains contro- versial (66,133). Evidence that pulmonary lobar arteries and veins (external diameter 0.8 to 1.4 mm) have the ability to contract is provided by Greenberg et a1. (98). In this investigation, large contractile responses were elicited during vascular transmural nerve stimulation and this effect was attenuated in a dose-dependent manner by several 21 adrenergic blocking agents including phentolamine, tolazoline, quanethi- dine and bretylium. Cocaine, which blocks reuptake of norepinephrine, enhanced the contractile activity (98). Using intact preparations, Daly et a1. (54) demonstrated that electrical stimulation of the sympathetic chain, stellate or middle cervical ganglions increased pulmonary artery pressure 10 to 15% during constant flow experiments. Furthermore, when pulmonary artery pressure was held constant, stimulations of similar magnitude produced a 30% decrease in flow. Since all passive factors affecting pulmonary vascular resistance were excluded, these investigators (54) concluded that active vasoconstriction caused pulmonary vascular resistance to increase. In more recent studies, electrical stimulation of the stellate ganglion at increasing frequencies caused a frequency- dependent increase in pulmonary vascular resistance in the open- (110,132) and closed- (133) chest dog and this effect was blocked by phentolamine. Using constant flow experiments, Kadowitz et al. (134) demonstrated that the postcapillary vessels contributed as much as 50% to the total increase in pulmonary vascular resistance with the remaining constriction occurring primarily in small arteries. In contrast, other investigators have been unable to show increased pulmonary vascular resistance in response to stellate ganglion stimulation in the perfused canine lung, but found instead that nerve stimulation increased pulse wave velocity and elastic modulus and decreased distensi- bility of large pulmonary arteries. These effects were blocked by the alpha blocker, phenoxybenzamine (129,229). Ingram et al. (129) believe their results are more consistent with the functional anatomy of the pulmonary circulation since the larger pulmonary arteries (> 2 mm diameter) 22 have more abundant elastic tissue, a greater supply of adrenergic nerves, and contain more norepinephrine in their walls than do the small arteries (66,85). The functional significance of increased stiffness of large pulmonary arteries would be to help maintain equal stroke volume between the two ventricles which would be particularly useful during sudden increases in right ventricular cardiac output, as might occur at the start of exercise. Thus, in contrast to the effects of local alveolar hypoxia which acts on small resistance vessels to redistribute blood to better ventilated parts of the lung, reflex sympathetic stimulation acts on larger arteries to synchronize right and left cardiac output by adjust- ing vascular distensibility (229). The absence of any change in pulmonary vascular resistance during stellate ganglion stimulation led Ingram et al. (129) and Szidon and Fishman (229) to conclude that the increased pulmonary vascular resistance demonstrated in earlier studies (54), was probably not due to vasocon- striction but rather due to viscosity changes. They argue that, because of the nonlinear behavior of blood at low shear rates (associated with low blood flow), such studies (54) would be expected to exaggerate the effects of viscosity on the pulmonary vascular bed (129,229). This explana- tion, however, is no longer tenable since more recent investigations demonstrating increased pulmonary vascular resistance during stellate ganglion stimulation have utilized normal pulsatile flows and cannulation techniques that are not likely to alter vascular innervation (110,132,133). Thus, it appears that sympathetic stimulation of pulmonary vessels is capable of both increasing vascular resistance and decreasing vascular compliance (66). Indeed, Daly and Daly (52) and Pace (177) electrically 23 stimulated the stellate ganglion in dogs and found that pulmonary vascular resistance and input impedance both increased. Input impedance, which is the ratio of peak pressure to peak flow, increases as vascular compliance decreases. The latter effect increases the fraction of total hydraulic power associated with pressure oscillations (117). The numerous studies (52,54,110,129,132-134,l77) involving nerve stimulation have firmly established the presence of efferent sympathetic pathways that may participate in neural regulation of pulmonary vessels. However, these studies do not establish the existence of afferent and/or central integrative pathways that are a prerequisite for reflex regulation of the pulmonary circulation (66). Reflex regulation of the pulmonary vasculature could be mediated through the peripheral baroreceptors, peri- pheral or central Chemoreceptors, or through release of humoral agents. Peripheral Baroreceptors and Chemoreceptors Changes in systemic blood pressure may cause alterations in pulmonary vascular resistance that are reflexly mediated through the carotid and aortic sinuses (53). The hemodynamic effects of carotid or aortic body stimulation on the systemic circulation have been well documented (44,118, 180) but the role of the peripheral Chemoreceptors in reflex control of the pulmonary circulation remains unsettled (66). Aviado et a1. (10) and Wilcox et al. (246) reported that systemic hypoxemia stimulated the peripheral Chemoreceptors to reflexly increase pulmonary vascular resist- ance. In contrast, while ventilating the left lower lobe with normoxic gases during systemic hypoxemia, Szidon and Flint (230) demonstrated a decrease in left lower lobe pulmonary vascular compliance but no change ir1 24 pulmonary vascular resistance as determined from pressure-volume and pressure-flow curves, respectively. None of these preparations, however, could distinguish between carotid and aortic body stimulation. In a complicated preparation Daly and Daly (53) reported increased pulmonary vascular resistance during hypoxemic perfusion of the carotid bodies but this effect was consistently found only in the absence of bronchial arterial blood flow. In another study, pharmacologic stimulation of the carotid bodies with nicotine failed to increase pulmonary vascular resist- ance but an increase in pulmonary vascular resistance was elicited during aortic body stimulation (225) and was later shown to be both pre- and post-capillary in origin (224). Whether the aortic bodies reflexly alter pulmonary vascular resistance or compliance in response to local physio- logic stimulation (i.e., altered gas tensions and pH) is unknown (66). An interesting but deleterious effect of peripheral chemoreceptor stimula- tion was reported by Levitzky et al. (142). They reported that hypoxemic stimulation (PaOZ < 40 mmHg) of the peripheral chemoreceptors attenuated the local alveolar hypoxia response causing increased right to left shunting of blood, thus worsening systemic hypoxemia. Denervation of the peripheral chemoreceptors blocked this positive feedback mechanism. m That the brain not only integrates afferent impulses but may also initiate efferent impulses to the pulmonary vasculature is evidenced by decreased pulmonary vascular compliance during brain ischemia (96) or electrical stimulation of the hypothalamus (229). In other studies electrical stimulation of the hypothalamus (4) or increased intracranial pressure (120,151) increased pulmonary vascular resistance. 25 HumoralgAgents Humoral agents may be released from the adrenal glands (epinephrine and norepinephrine), from within the lungs themselves (histamine, serotonin and prostaglandins), or released elsewhere in the body and activated within the lungs (angiotensin 2). These agents have potent effects on the pul- monary circulation. In the isolated intact lung preparation, with con- stant blood flow and outflow pressures, these agents are generally vaso- constrictors; however, in the intact animal they cause variable responses because they also alter cardiac output, mean arterial and left atrial pressures which secondarily produce passive changes in the lung vasculature (L7). While a large variety of factors may cause release of catecholamines from the adrenal glands or antidiuretic hormone from the posterior pitui- tary (104) it is particularly interesting that carotid body hypoxia also reflexly causes release of these humoral agents (7,209). Thus, it seems clear that sympathetic nerves and receptors supplying lung vessels are part of an extensive control system which can be stimulated centrally and reflexly, thereby modulating pulmonary hemodynamics. IV. Lung Fluid Balance Starling Forces Transvascular movement of water in the lung is believed to be governed by hydrostatic and osmotic forces as described in the Starling equation (88,106,141,221): Qf = Kf [(va'Ppmv) - O ("mv-"pqu ’ where 0f is the rate of net transvascular fluid flow, K1: is the 26 microvascular filtration coefficient, va is the microvascular hydrostatic pressure, P is the perimicrovascular hydrostatic (interstitial) pres- pmv sure, <1 is the weighted plasma protein reflection coefficient, "mv is microvascular colloid osmotic pressure, and n is the perimicrovascular mv colloid osmotic pressure. When discussing chse forces average values for the entire lung are usually used; however, large differences in the distribution of Starling forces exist at different levels in the lung due to the action of gravitational forces on blood flow and ventilation. Until relatively recently it was believed that net reabsorption of fluid into the microcirculation occurred in the lung (105). However, it is now generally agreed that since lymph is continuously formed in normal lungs, the balance of the Starling forces must favor net filtration (26). While the magnitude of intravascular Starling forces (i.e., va approxi- mately 7 to 9 mmHg and "mv approximately 25 to 28 mmHg) are generally well established, the magnitude of extravascular forces is intensely debated. Staub (221) believes P is atmospheric and n is very high pmv pmv (approximately 20 mmHg) whereas Guyton et al. (106) believe P mv is sub- p stantially subatmospheric (approximately - 8 mmHg) and n is only about mv 13 mmHg. Guyton et al. (106) maintain that pulmonary lyfiph protein may become concentrated by lymph nodes and therefore lung lymph overestimates the actual perimicrovascular protein concentration. Regardless of which theory is correct, both agree that the relative magnitude of these extra- vascular Starling forces are such that the lung normally has net filtra- tion occurring from the microcirculation. 27 Pathways for Fluid Movement Several pathways for fluid and protein transport are thought to exist across the pulmonary microvascular endothelium (20,218). These include the small pores (radii 34 angstroms) which allow complete protein sieving, intermediate pores (radii 34 to 300 angstroms) which allow partial sieving of protein, and infrequently occurring large pores (approximately 1000 angstroms) which allow no protein sieving (i.e., all proteins pass through). These pores are all in the intercellular junction and are hydraulically conductive as well as a diffusion pathway for lipid insoluble macromolecules. Their sizes arbitrarily signify their relative reflection coefficients for plasma proteins (i.e., o - 1 for small pores, 0 fbr large pores, and variable for intermediate pores). In addition to pores in the intercellular junctions, extremely small intracellular pores (5 angstroms) also exist. These intracellular pores are thought to con— duct water but are very restrictive to almost all solute molecules (218). Vesicular transport may also account for some protein conductance although this process is considered to be too slow to be physiologically signifi- cant (20,231). V. Pulmonary Edema Iypes and Etiologies Pulmonary edema is a pathologic state in which there is abnormal extravascular water accumulation in the lung (234). According to the Starling equation pulmonary edema can be mediated by two possible mechanisms: 1) increase in pulmonary microvascular hydrostatic pressure or 2) increase in microvascular permeability. The former could be due 28 to an alteration in pulmonary hemodynamics (i.e., increased pulmonary blood volume, pulmonary vascular resistance or decreased pulmonary vascular compliance) and the latter mediated by mechanical, humoral or neural factors (86). Increased microvascular hydrostatic pressure causes increased fluid filtration from the microvasculature and this is mani- fested by a low protein concentration in pulmonary lymph. However, increased endothelial permeability results in an increase in both fluid and protein conductances causing pulmonary lymph to be high in protein (218). Although the latter may be reversible with time, it may be much more serious than conditions which only elevate capillary hydrostatic pressure (e.g., congestive heart failure) because of the marked reduction in the transmural colloid osmotic pressure gradient (222). Sites of Fluid Leakage and Accumulation Leakage of fluid from the pulmonary circulation is not limited to capillaries and venules (2,3,128,l62). Using isolated dog lungs, Iliff (128) increased pulmonary artery or venous pressures while maintaining zone 1 conditions (thus compressing alveolar vessels). In either case lung weight increased substantially, with leakage primarily occurring from extra-alveolar veins. Under zone 3 conditions approximately 60% of transvascular fluid movement was through the extra-alveolar vessels. In similar studies Mitzner and Robotham (162) concluded that 50% of transvascular fluid movement was through alveolar vessels, 23% through extra-alveolar arteries and 27% through extra-alveolar veins. The princi- pal Starling force contributing to leakage of fluid from extra-alveolar vessels is presumably the perivascular hydrostatic pressure, which may 29 become increasingly subatmospheric (< - 20 mmHg) as the surrounding lung parenchyma "pulls" on the relatively indistensible perivascular connective tissue. The extravascular—extracellular space of the lung is composed of an interstitial and an alveolar gas space. The interstitial space has two components, the alveolar wall interstitium (thick septum) and the loose connective tissue space surrounding airways and blood vessels (222). The former site is where excess fluid is first detected by electron microscopy (45) whereas the latter is the first site detected by light microscopy (223). The alveolar-capillary barrier (i.e., thin septum) shows no evidence of fluid accumulation even during severe pulmonary edema resulting from high microvascular hydrostatic pressure (45). The alveolar wall interstitium is continuous with the loose connec- tive tissue space surrounding vessels and airways, and although this space is not large in the normal lung, it has a very large potential for expansion to a volume that is approximately equal to pulmonary blood volume (222,251). The alveolar gas space has an enormous potential for fluid accumulation that is an order of magnitude greater than the poten- tial interstitial space; however, alveoli do not begin accumulating a significant fluid volume until interstitial edema is extensive (222). Fluid filtered into the alveolar interstitium apparently moves by bulk flow to the perivascular and peribronchial regions. Since the interstitial pressure surrounding these structures is thought to be lower: than it is around alveolar capillaries, fluid moves down a hydrostatic pressure gradient toward junctional tissues where arterioles, venules and lymphatic vessels lie (84,106). 30 Safety Factors Safety factors that operate to prevent pulmonary edema when capil- lary filtration pressure is increased include (178): l) a very low compliance of the interstitial space; 2) decreased interstitial colloid osmotic pressure because of protein washout; and 3) increased lymph flow. Because the interstitial space has an initial low compliance, interstitial pressure can increase from -7 to 0 mmHg without development of gross edema. However, when interstitial hydrostatic pressure increases above atmospheric pressure, compliance of the interstitial space dramatically increases resulting in rapid accumulation of water in the interstitial space and alveoli. Under chronic conditions of increased fluid filtration lymphatics may hypertrophy, thus facilitating rapid removal of lymph. During chronic pulmonary venous hypertension in dogs, lymph flow has been found to increase as much as 28-fold (115). Quantitation of Lung Water Quantitation of pulmonary edema is a difficult task, particularly in the intact animal (234). Lung auscultation, lung radiographs or measurements of lung compliance, and blood gases are all nonspecific and often show little evidence fOr increased lung water accumulation until pulmonary edema becomes severe (105,219). Attempts to quantitate pul- monary edema have primarily centered on three techniques: 1) measurement of transthoracic electrical impedance, 2) determination of radiation emission or absorption, and 3) double indicator-dilution studies (219). In the latter technique diffusible (e.g., heat, tritiated water) and non- diffusible (e.g., radiolabeled red blood cells, dye-tagged albumin, 31 sodium) substances are simultaneously injected into the right heart with measurement of these substances in the left heart or aorta. The diffusi- ble indicator measures total tissue volume. For heat (which is extremely diffusible) this includes blood volume, lung water and solids. Tritiated water also measures blood volume but its extravascular volume of distribu- tion is limited to approximately 70% of the lung water space. Thus, heat overestimates the lung water Space while tritiated water underestimates this space. Since the nondiffusible indicator is confined to the intra- vascular space it only measures blood volume. Chinard et al. (41) has Shown that even small ions such as sodium are confined to the normal pulmonary vascular space during their initial transit. In fact, sodium dilution curves are virtually superimposed onto dye-tagged albumin curves. By subtracting intravascular volume from total tissue volume one obtains extravascular volume and this is used as an index for lung water. Although double-indicator dilution techniques provide a quantitative assessment of pulmonary edema it is unlikely that changes less than 15% are detectable (219). Gravimetric analysis provides the most accurate determination of lung water and is accomplished by weighing excised lungs, drying to con- stant weight and correcting for intravascular blood volume (183). The extravascular lung water can then be expressed as a function of the animals dry lung weight (219). Although this provides a reliable index to which in vivo techniques can be compared, it obviously cannot be used to make antemortem assessments of lung water. 32 IV. Brain and Pulmonary Edema The brain's ability to alter pulmonary hemodynamics and cause pulmonary edema and alveolar hemorrhage is demonstrated in numerous experimental and clinical reports. Intracisternal injection of fibrin (34,204) or veratrin (130), blunt head trauma (15,36-38,l47,159), electrical (91,149) or chemical (248) stimulation of the hypothalamus, saline induced hypertension of cerebral vessels (146), chemically induced convulsions (l6), electrolytic destruction of the nucleus tractus soli- tarius (64), and increased intracranial pressure (35,74,151) all cause pulmonary edema and alveolar hemorrhage. Clinical disorders of the brain leading to pulmonary edema and alveolar hemorrhage include seizures (21,124), increased cerebrospinal fluid pressure (73), head trauma (79, 212), cerebral hemorrhage (240), apoplexy and tumors (33). In all of these experimental and clinical reports the following observations appear consistent: l) the edema and hemorrhage are sudden in onset frequently leading to death of the animal, 2) the edema fluid is very high in protein content with numerous red blood cells suggesting large increases in vascular permeability, 3) massive central sympathetic outflow to the cardiovascular system occurs as evidenced by large increases in pulmonary and systemic arterial pressures, 4) pulmonary blood volume increases, and 5) vasodilator agents prevent pulmonary edema and alveolar hemorrhage by preventing large increases in left ventricular afterload, thus pre- venting left ventricular failure and subsequent pooling of blood in the pulmonary circulation. It seems plausible that the brain might regulate pulmonary hemo- dynamics through the autonomic nervous system since the latter extensivelyr 33 innervates the pulmonary vascular bed (117,245) with alpha, and to a lesser extent beta adrenergic receptors (17). The ability of the brain, through the autonomic nervous system, to control the pulmonary vasculature is suggested by several investigations. Pulmonary denervation protects the lungs from edema and alveolar hemorrhage during isolated cerebral hypoxemia (167) or hemorrhagic shock (226,227). By preventing large increases in pulmonary vascular pressures, phenoxybenzamine (an alpha blocker) prevents pulmonary edema associated with chemically induced con- vulsions (16), chemical hypothalamic stimulation (248), blunt head trauma (15,36) and increased intracranial pressure (151). Although a portion of this protective effect is due to less pulmonary vasoconstriction during various brain stimuli, a major portion is due to the fact that the normally large increase in left ventricular afterload (i.e., mean aortic pressure) is prevented. Since afterload is an important determinant of stroke volume (i.e., stroke volume transiently decreases as afterload increases), it follows that left ventricular stroke volume would remain closely matched with right ventricular stroke volume thereby preventing increases in pulmonary blood volume and vascular pressures. In other studies (129,229) sympathetic nerve stimulation, hypothalamic stimulation, or systemic hypoxia decreased pulmonary artery compliance, an effect that was blocked by phenoxybenzamine. Phentolamine (another alpha blocker) blocks the increased pulmonary vascular resistance that occurs during electrical stimulation of the stellate ganglion (132-134), alveolar hypoxia (123) or systemic hypoxia (13). Phentolamine (36,64,248) or cervical cord transection (38,149) prevents pulmonary edema associated with hypothalamic stimulation, blunt head trauma, or electrolytic 34 destruction of the nucleus tractus solitarius. Propranolol or practolol (both beta blockers) potentiate the increase in pulmonary vascular resistance during alveolar hypoxia (123). Propranolol also potentiates the increase in pulmonary vascular resistance during systemic hypoxia (13,233). In other studies propranolol attenuated the increase in pul- monary vascular resistance associated with increased intracranial pressure (151). This seems paradoxical since the effect of removing beta receptor activity should be vasoconstriction and therefore potentiation of any increase in pulmonary vascular resistance. However, propranolol also induced left ventricular failure causing pulmonary vascular pressures to rise. The effect of increasing vascular transmural pressure is to increase vessel diameter and thus passively decrease pulmonary vascular resistance. In contrast to considerable sympathetic control of the pul- monary circulation (18), parasympathetic control is virtually nonexistent (170). Bilateral cervical vagotomy has no effect on elevated pulmonary vascular resistance due to alveolar hypoxia (136); however, acetylcholine antagonizes this effect during alveolar hypoxia (123) or systemic hypoxia (65). Thus, there is considerable circumstantial evidence which suggests that the brain has the ability to alter pulmonary hemodynamics during systemic hypoxia and/or hypercapnia and that this response may be regu- lated by the sympathetic component of the autonomic nervous system. VII. Adult Respiratory Distress Syndrome Definition and Incidence The adult respiratory distress syndrome is a severe pulmonary compli- cation usually occurring within 72 hours of a wide variety of 35 extrapulmonary insults, the most common being hemorrhagic shock and trauma (19). In its most dramatic form, it is characterized by the acute and unexpected onset of severe, life-threatening respiratory distress in patients whose lungs were previously normal. The syndrome is synonymous with shock lung, congestive atelectasis, capillary leak syndrome, DaNang lung, wet lung and numerous other terms (19,113,122). Despite the diverse nature of the adult respiratory distress syn- drome, diffuse damage to the alveolar-capillary membrane is considered a common denominator and once this occurs the ensuing events are similar in most patients (122). It should be emphasized, however, that this may simply represent the limited manner in which the lung responds to injury (19,122). Indeed, this is the strongest argument against combining the many diverse but specific entities into a single diagnostic category of adult respiratory distress syndrome since this may detract from important distinctions that should be made concerning pathophysiology, treatment and prognosis (81,122). The recent development of this syndrome is thought to reflect improved emergency treatmentcfl’severely traumatized patients who would have otherwise succumbed to the initial insult (81,167). Although the incidence of human adult respiratory distress syndrome is increasing, mortality has been slowly decreasing and is now approximately 50% (122, 241). In veterinary medicine the adult respiratory distress syndrome is not well documented (191). This may be due to lack of awareness, financial considerations, and/or death of the animal from the initial insult because of a relative lack of major trauma centers. 36 Etiologic Factors The exact etiology of the adult respiratory distress syndrome is unknown, although a number of factors are known to be involved. The most frequently cited factors are (19,113,167): 1) direct injury to the lung secondary to hemorrhagic shock, 2) oxygen toxicity, 3) ventilator induced injury, 4) fluid overload, 5) central nervous system trauma, 6) thrombo- embolism, 7) sepsis, 8) release of humoral agents (e.g., histamine, serotonin, kinins, prostaglandins, complement activation, catecholamines), 9) lymphatic insufficiency, and 10) brain hypoxia. When considering etiology of the adult respiratory distress syndrome it becomes very diffi- cult to distinguish between initiating and complicating events. For example, a patient in hemorrhagic shock may have some or all the above- mentioned etiologic factors as secondary complications. Therefore, I believe that adult respiratory distress syndrome should be thought of as having a multifactorial etiology rather than a distinct single etiology. That is, the more factors that are present in a clinical situation, the more likely that the adult respiratory distress syndrome will develop. Pathophysiology The gross pathologic changes associated with the adult respiratory distress syndrome are nonspecific. Within a few hours following the clinical insult scattered petechial hemorrhages are evident over the sur- face of the lungs. By approximately 24-48 hours the petechial hemorrhages have become confluent, especially in zone 3, causing the lungs to appear congested and hemorrhagic. However, inflation of the lungs during this phase results in clearing of much of the hemorrhagic appearance, indicatiru; that the primary changes are those of congestion and atelectasis. 37 By 48 to 72 hours the lungs grossly resemble the liver and are hemorrhagic, as evidenced by persistance of a dark red color even when the lungs are inflated. Secondary infection is common causing formation of abscesses and purulent bronchial secretions which may ultimately lead to septicemia and death. When the thorax is opened, little or no collapse of the lungs is observed and the lungs may be several times their normal weight (19,81). Early histologic features in the adult respiratory distress syndrome consist of interstitial and alveolar edema with hemorrhage and focal atelectasis. Pulmonary vessels develop thrombi while cellular debris and proteinaceous fluid accumulate in alveoli. Hyaline membranes, thought to be composed of high molecular weight proteins, are found lining alveoli and alveolar ducts. The damaged alveolar epithelium becomes lined with type 2 alveolar cells. Lesions in alveolar epithelium usually appear more severe than those of the vascular endothelium, however, this is thought to be because of the much more rapid repair capacity of the latter rather than a unique susceptibility of the alveolar epithelium (11). Eventually interstitial fibrosis may develop, although complete healing and repair may occur if the underlying factors causing the adult respira- tory distress syndrome are brought under control (241). Pulmonary function gradually deteriorates as evidenced by increased arterio-venous shunting and respiratory dead space and decreased func- tional residual capacity. By increasing surface tension forces, atelecta- sis and alveolar edema decrease lung compliance (242). Diffusion impairment and ventilation perfusion inequalities exist and shunt fraction may exceed 50%. These effects lead to severe hypoxemia and hypocapnia. The latter occurs because ventilation is increased secondary to hypoxemic 38 stimulation of peripheral chemoreceptors and stimulation of vagal affer- ents located in the lung interstitium. When alveolar flooding occurs carbon dioxide is retained leading to severe respiratory acidosis (241). Centrineurogenic Hypothesis Recently Moss and Stein (167) and Brown (30) proposed a centrineuro- genic etiology for the adult respiratory distress syndrome. These investigators (30,167) believe that brain hypoxia impairs oxidative metabolism in the hypothalamus which increases sympathetic outflow to postcapillary vessels leading to increased pulmonary vascular resistance and microvascular hydrostatic pressure, congestion, surfactant inactiva- tion, atelectasis, edema, and hemorrhage. In their preparation venous blood (P02 35 torr) is pumped from the right atrium to one common carotid artery for 2 hours at a perfusion pressure 20 mmHg greater than mean aortic pressure. They believe the brain is being perfused only with venous blood since all other arterial channels supplying the Circle of Willis should experience retrograde flow because perfusion pressure exceeds mean aortic pressure. Moss and Stein (167) and Brown (30) reported that pulmonary lesions observed at postmortem were similar to those observed in the adult respiratory distress syndrome. They believe the pulmonary lesions secondary to brain hypoxia are species independent since similar responses were obtained in all species studied including dog, calf, pig, goat, rabbit, sheep and monkey (168). These investigators (167) then repeated their experiments following chronic unilateral pul- monary denervation. The denervation procedure involved transection of all ipsilateral hilar lung tissue fbllowed by reanastomoses of bronchi and vessels. TWO months later they induced brain hypoxia in these dogs for 39 two hours. Based upon gross and histologic examination they found that the innervated lung developed lesions consistent with the adult respira- tory distress syndrome whereas the denervated contralateral lung was normal. Moss and Stein (167) believe their experiments are prima-facie evidence that brain hypoxia results in lesions consistent with the adult respiratory distress syndrome and that these events are neurogenically mediated. The centrineurogenic hypothesis is heavily dependent on circumstantial evidence since there are no reports in the literature describing the specific effects of local brain hypoxia and/or hypercapnia on pulmonary hemodynamics. A critical element to the centrineurogenic hypothesis is that pulmonary postcapillary vessels have the ability to constrict in response to sympathetic stimulation. That postcapillary vessels are capable of increasing pulmonary vascular resistance is evidenced by constriction of pulmonary veins during systemic hypoxia (163,190,195,216) electrical stimulation of the stellate ganglion (134), aortic body stimula- tion (224), brisket disease (139) or hemorrhagic shock (137,169,226). Increased postcapillary resistance increases microvascular hydrostatic pressure which may lead to pulmonary congestion, edema, and alveolar hemorrhage (167,169). It should be pointed out, however, that precapillary constriction could also produce similar pathology. Increased pulmonary vascular resistance due to precapillary constriction occurs during alveolar and systemic hypoxia and causes pulmonary edema (126,213,243). Leakage of fluid under these conditions could occur from extra-alveolar vessels (128,162). It has also been suggested that damage to arterial walls proximal to constricted arterioles results in transarterial leakage 40 of fluid (243). An alternative hypothesis is that precapillary constric- tion may be non-uniform in nature so that the increased pulmonary artery pressure is transmitted to pulmonary capillaries not served by con- stricted arterioles (235). Additional circumstantial support for the centrineurogenic hypothe- sis is that electrical stimulation of the hypothalamus (4) increases pulmonary vascular resistance (albeit during nonsteady state measurements of pressure and flow) as does increased intracranial pressure (120,151). In other studies, however, pulmonary vascular resistance did not change during increased intracranial pressure (144,155) or brain ischemia (96), conditions which presumably cause brain hypoxia. During hemorrhagic shock a larger proportion of the cardiac output is delivered to the brain because of excellent autoregulatory ability of the cerebral vasculature. However, this sparing effect is relatively short-lived during severe hemorrhagic shock (167). Considerable evidence suggests that brain hypoxia is an important manifestation of late hemor- rhagic shock. A large reduction in cerebral blood flow (202,215), decreased cerebral cortical P02 (236), increased brain lactate (249) and increased cerebrospinal fluid potassium and pseudocholinesterase (210) all offer circumstantial evidence that brain hypoxia occurs during hemor- rhagic shock in the dog. Since hemorrhagic shock may produce pulmonary edema and alveolar hemorrhage (76,119,137,l69,201,208,226-228) similar to that observed following brain hypoxia (167), Moss and Stein (167) believe brain hypoxia during shock is what initiates the pathophysio- logical events that culminate into the adult respiratory distress syndrome. Consistent with this hypothesis is the fact that during hemorrhagic shock, 41 anemia (and therefore decreased oxygen carrying capacity) predisposes to development of the syndrome (166) while diphenylhydantoin, a neuro- pharmacologic agent that reduces brain excitability, is protective (165). In summary, the centrineurogenic hypothesis for the adult respira- tory distress syndrome is based on a large body of circumstantial evidence that has not been critically tested. For example, there are no descrip- tions in the literature of the specific effects (i.e., changes in pul- monary vascular resistance, central blood volume, and lung water) of local brain hypoxia and/or hypercapnia on pulmonary hemodynamics. However, there are numerous experimental and clinical reports suggesting that a wide variety of brain insults can exert profound effects on pulmonary hemodynamics and ultimately cause pulmonary edema and alveolar hemorrhage similar to that observed in the adult respiratory distress syndrome. The adult respiratory distress syndrome is currently a major clinical problem for which no clear etiology has emerged and research in this area has not been particularly rewarding primarily because a suit- able animal model has not been developed (63a). Clearly, a better understanding of pulmonary vascular control by the brain is needed, particularly during pathophysiological states such as the adult respira- tory distress syndrome. Thromboembolism and Humoral Agents Although there appear to be numerous factors responsible for the adult respiratory distress syndrome, Blaisdell and Lewis (19) believe that following hemorrhagic shock and trauma thromboembolism becomes the primary initiating factor and that other postulated causes are thrombo- plastic in nature. This leads to activation of the coagulation cascade 42 with formation of microemboli consisting of platelets, red and white blood cells, fibrin, cellular debris and neutral fat. These microemboli cause mechanical obstruction of vessels and cause release of histamine, serotonin, and prostaglandins as well as activation of complement and release of proteolytic enzymes from polymorphonucleocytes. These vaso- active agents not only cause vasoconstriction but also increase bronchial and/or pulmonary microvascular permeability (19,84). In this regard, histamine is a potent constrictor of pulmonary veins (92,93) and causes a large increase in bronchial vascular permeability (188). Wilson (247) found a progressive increase in number of mast cells in the lung following hemorrhagic shock and cardiopulmonary by-pass. Since the mast cells were frequently degranulated (suggesting massive release of histamine) and the lungs developed pathologic changes similar to the adult respiratory dis- tress syndrome, he concluded that histamine was a likely mediator of these events. Other investigators also believe that histamine may play an important role in the pathogenesis of the adult respiratory distress syndrome (19,197). Release of histamine in this syndrome also provides an explanation for the characteristic bronchoconstriction that is known to occur (19). Complement activation, especially 03a and 05a fraction, and degranu- lation of polymorphonucleocytes also cause large increases in vascular permeability (63a). The combination of high pulmonary vascular pressures (mediated primarily by vasoconstriction and thromboemboli) and increased vascular permeability all contribute to development of severe pulmonary edema and alveolar hemorrhage. According to Blaisdell and Lewis (19), iatrogenic factors such as oxygen toxicity, ventilator induced injury, 43 and fluid overload simply enhance the development and severity of the adult respiratory distress syndrome. VIII. Histamine, Microvascular Permeability and Adrenergic Receptors Local intra-arterial administration of histamine to the canine forelimb (102,108,152) and equine digit (198,199) causes massive edema through increases in capillary hydrostatic pressure, microvascular surface area, and microvascular permeability (107a). The increase in the latter parameter is clearly the dominant factor in the edema produced by hist- amine. In this regard histamine appears to affect both the large and small pore systems (108a), however, its action on the latter is small (2 to 3 fold increases in capillary filtration coefficient) and transient lasting only 10-20 minutes (137a). Thus the majority of peripheral edema produced by histamine results from increased microvascular permeability to macromolecules, i.e., plasma proteins (108a). Many investigators believe this effect occurs primarily in venules and is due to contraction of actomyosin fibrils within the endothelial cytoplasm leading to increased gap formation between endothelial cells (150,188). The edemagenic effects of local intra-arterial histamine on the forelimb can be blocked by simultaneous intra-arterial infusion of the beta agonist, isoproterenol (152). In contrast, intravenous (i.e., systemic) administration of histamine fails to increase forelimb weight even when calculated blood concentrations of histamine equal or exceed the intra-arterial dose (153). However, in the presence of beta blockade (propranolol) intravenous histamine causes massive forelimb edema 44 comparable to that obtained during intra-arterial histamine alone (100). Thus, it appears that increased blood levels of beta agonists during intravenous histamine infusion antagonize the increased water efflux in the canine forelimb (100). With regard to action of histamine on microvascular permeability in the lung, Goetzman and Visscher (94) found no increase in canine alveolocapillary membrane permeability to radiolabeled albumin when histamine was added to the lung perfusate while Grega et al. (103) reported no increase in lung weight when histamine was infused (2.1-10 ug base/min) into the pulmonary artery of isolated fetal, neo-natal and adult canine lungs. In anesthetized dogs, Pietra et al. (188) found that intravenous histamine (7 ug base/min) caused no leak of colloidal carbon from the pulmonary microcirculation but did find transient leakage from the bronchial vessels that resulted in small increases in lung water. A similar distribution of colloidal carbon following intravenous hist- amine was observed in the rat lung (89). Pietra et al. (189) found no increase in lung weight when histamine was infused into the isolated rabbit lung. More recently, Drake and Gabel (70) found that intravenous histamine failed to increase pulmonary lymph flow and concentration, capillary filtration coefficient, and the calculated permeability-surface area product in the dog lung. 0n the other hand, Brigham and Owen (27) reported that histamine infused into sheep caused a two to six fold increase in pulmonary lymph flow (collected from the efferent duct of the caudal mediastinal lymph node) without a change in the lymph to plasma protein ratio. They also found that histamine increased permeability-surface area product for 45 eight protein fractions and concluded that histamine increased micro- vascular permeability and that its principal site of action was the pulmonary microvascular bed. More recently, Nakahara and co-workers (171) infused histamine into the bronchial or pulmonary circulation of sheep. Under both conditions lymph protein clearance increased in a dose- dependent manner and they concluded that histamine causes a mild increase in pulmonary microvascular permeability. In both investigations (27,171) pulmonary edema was minimal. Critical assumptions in these preparations (27,171) are that the collected lymph is void of systemic contamination, is not modified by lymph nodes and is representative of perimicrovascular interstitial fluid. Drake et al. (72a) recently reported that lymph collected from the efferent duct of the caudal mediastinal node contains significant nonpulmonary contamination, thus raising doubts on conclusions drawn from these studies (27,171). However, even if the assumptions are valid, the effects of histamine on microvascular permeability and lung water in sheep can only be considered mild at best. It should be mentioned that sheep pulmonary vessels have more sympathetic innervation than most mammals (117) as well as more mast cells in their lung. The mast cells have a particularly rich supply of dopamine (78), the latter being a precursor for norepinephrine synthesis. These factors could contribute to a pulmonary vascular bed more reactive to histamine and sympathetic stimulation in sheep. Differences among preparations could also contribute to variable responses to histamine. Recent studies suggest that beta adrenergic receptors may regulate transvascular fluid and protein movement in lungs. In sheep, Hakim et al. (111) reported increased lymph protein clearance following beta blockade 46 with propranolol. Although they believe this effect is partly due to increased microvascular permeability, their data can be explained entirely by an increase in vascular surface area. Persson et al. (186) reported that the beta2 agonist, terbutaline, prevented histamine induced (aerosol) pulmonary edema in the guinea-pig. In most (and possibly all) previous studies concerned with the effects of histamine on lung perme- ability the possibility exists that abnormally high plasma catecholamine levels partially or completely antagonized the permeability effects of histamine and thus prevented pulmonary edema. Increased levels of beta agonists during intravenous histamine administration would be expected to occur because histamine not only causes direct release of catechol- amines from the adrenal glands (217) but also indirectly through reflex sympathoadrenal stimulation secondary to systemic hypotension (100,153). In summary, local intra-arterial administration of histamine increases microvascular permeability and causes edema in the canine fore- limb. During intravenous infusion of histamine these effects are not seen unless there is concomitant beta blockade suggesting that increased blood levels of beta agonists during intravenous histamine antagonize the increased water efflux. Whether a similar phenomenon occurs in the pulmonary microvasculature has not been specifically tested. I believe this is an important question to be answered because of the potential endogenous release of histamine during pathophysiological states and because of increased clinical use of beta blockade therapy. STATEMENT OF OBJECTIVES The first objective of this research was to determine the effects of local brain hypoxia or hypercapnia or both on pulmonary hemodynamics in systemically normoxemic and normocapnic dogs. I determined pulmonary vascular resistance, lung extravascular thermal volume, and central blood volume during brain hypoxia. Lung water was determined gravimetrically following brain hypoxia. In the experimental preparation used by Moss and Stein (167), venous blood is pumped from the right atriUm to one common carotid artery. This means that the carotid bodies and brain are perfused with hypoxemic blood. Therefore, any change in pulmonary hemodynamics could be caused by either brain hypoxia or by carotid body hypoxia. Thus the second objective of this research was to evaluate the effects of carotid body hypoxia and/or hypercapnia on pulmonary hemodynamics. Physiologic stimulation of the carotid bodies was achieved by altering gas tensions of blood perfusing the carotid bodies in systemically (including the brain) normoxemic and normocapnic dogs. A third objective of this research was to evaluate the effects of one humoral agent (i.e., histamine) thought to be involved in the patho- genesis of the adult respiratory distress syndrome. Of particular interest was the effect of histamine on lung water before and after adrenergic blockade. Specifically, I tested the possibility that increased levels of beta agonists may account for the fact that intravenous 47 48 histamine causes little or no increase in microvascular permeability and water content of dog lungs. For completeness, I also examined the possibility that stimulation of alpha adrenergic receptors might alter the response to histamine. I believe these investigations are of considerable significance for both the physiologist and clinician because the specific effects of local brain hypoxia on pulmonary hemodynamics have not been reported, the effects of carotid body hypoxia on pulmonary hemodynamics are contro- versial, and the effects of histamine on lung water following beta or alpha blockade have not been reported. Furthermore, these studies have important therapeutic implications because if it can be demonstrated that the brain (or carotid bodies) increases pulmonary vascular resistance or pulmonary blood volume during local brain (or carotid body) hypoxia and/or hypercapnia, then pharmacologic agents with actions on the brain (or carotid bodies) and vasculature may become an important modality in the prevention or treatment of pulmonary edema and alveolar hemorrhage during periods of systemic hypoxia and/or hypercapnia. In addition, since propranolol is used in certain clinical situations, it is important to determine whether beta blockade alters the effect of histamine on the pulmonary circulation. In this dissertation the experiments involving "Brain Hypoxia and/or Hypercapnia", "Carotid Body Hypoxia and/or Hypercapnia", and "Histamine and Adrenergic Receptors” are each presented sequentially in the "METHODS", "RESULTS", and "DISCUSSION" sections. In the "SUMMARY and CONCLUSIONS" section I have attempted to bring together all aspects of this research. The final section of the text includes "RECOMMENDATIONS" for future research on the brain's role in causing pulmonary edema. METHODS I. General One-hundred nineteen mongrel dogs of either sex weighing 22: 0.6 kg were used in these investigations. All dogs were anesthetized with sodium thiamylal (17.6 mg/kg, Bio-tal, Bio-Ceutic Laboratories, Inc., St. Joseph, MO) and anesthesia was maintained with a mixture of chlora- lose (75 mg/kg) and urethane (500 mg/kg, Sigma Chemical Company, St. Louis, MO). The dogs were intubated with a cuffed endotracheal tube and placed on positive pressure ventilation (Model 613, Harvard Apparatus Company, Millis, MS) with 5 cm water positive end-expiratory pressure. They were placed in a supine position and ventilated with room air supplemented with one to two liters 02 per minute. Tidal volume and respiratory rate were 22 ml/kg and 14 breaths/min, respectively. Systemic pH, PaOz, and PaCO2 were maintained at 7.38 i 0.01, 122 i 4 torr, and 36 i l torr, respectively. Sodium bicarbonate and lactated ringers solution (Cutter Laboratories, Berkeley, CA) were administered intravenously to correct for metabolic acidosis and fluid loss. A blood acid-base alignment nomogram was utilized to determine the base deficit and the estimated lost fluid volume (in open-chest preparations) was 4 m1/kg/hour. Pancuronium bromide (0.08 mg/kg, Pavulon, Organon, Inc., W. Orange, NJ) and heparin sodium (10 mg/kg, Lipo-Hepin, Riker Laboratories, Inc., Northridge, CA) were given intravenously to paralyze the muscles of respiration and for anticoagulation, respectively. 49 50 II. Brain and Carotid Body Studies Polyethylene catheters were placed into the aorta via the femoral artery and following a median sternotomy, into the left atrial appendage. A 7F thermistor-tip Swan-Ganz catheter (Swan-Ganz Directed Monitoring Catheter, Edwards Laboratories, Inc., Santa Ana, CA) was floated into the pulmonary artery. Left atrial (as measured in its appendage), pulmonary artery, pulmonary artery wedge and aortic pressures were measured using Statham P23Gb pressure transducers (Gould Medical Products, Statham Industries, Oxnard, CA) referred to the level of the left atrium. In 30 dogs (brain perfusion experiments), a 20 gauge needle was inserted directly into the left ventricle and the needle hub sutured to the ventricular wall. A polyethylene catheter connected this needle to a pressure transducer. Left ventricular dP/dt was obtained by electrical differentiation of the ventricular pressure signal. All recordings were made continuously on a polygraph recorder (Model 7796A, Hewlett-Packard, Farmington Hills, MI). Cardiac output was measured by the thermodilution-technique. Four ml normal saline (l-2°C) was injected into the right atrium via a side port (located 20 cm from the catheter tip) in the Swan-Ganz catheter. A thermistor positioned near the tip of the catheter detected changes in blood temperature and a computer (Cardiotherm 500, Columbus Instruments, Columbus, Ohio) calculated cardiac output. The mean of 3 to 5 determina- ations was computed as right ventricular output. A blood gas analyzer (Radiometer-blood micro system, Copenhagen, Denmark), which was calibrated daily, was used to determine pH, P02 and PCO2 of systemic arterial blood and of blood from the extracorporeal lung (i.e., perfusate). 51 III. Brain Hypoxia and Hypercapnia Brain hypoxia, hypercapnia, and hypoxia-hypercapnia were caused by pumping arterial autologous blood through an extracorporeal lung (Figure l). Outflow from the extracorporeal lung was pumped to the external carotid arteries. The head was perfused via the external caro— tid arteries with flow to the Circle of Willis via the extensive anastomotic connections that exist in the dog (97). In order to mini- mize systemic blood flow to the brain the following vessels were initially bilaterally ligated: internal carotid, external carotid, occipital, lingual, ascending pharyngeal, cranial laryngeal and cranial thyroid arteries. It is unlikely that any of these ligations resulted in reduced brain blood flow Since Green and Rapela (97) reported that, following bilateral common carotid artery ligation, vertebral artery perfusion pressure is approximately 90% of mean arterial pressure indicating an extensive collateral blood supply to the canine brain. The internal carotid and occipital arteries were ligated cephalad to the carotid sinuses and bodies, respectively (Figure 1). Thus the carotid sinuses were exposed to systemic arterial pressure and the carotid bodies received systemic blood in which gas tensions and pH were normal. Meticulous dissection preserved the carotid sinus nerves and their functional integrity was assumed if arterial pressure increased during temporary bilateral common carotid artery occlusion. The vertebral, costocervical, and omocervical arteries were initially bilaterally tagged with silk suture at their origin from the subclavian artery (Figure l) but were not ligated until after head perfusion had 52 .xemucm uwpocmu coEEou .muo “Aeov accuse Pmu 1w>smuoeo use .Aeov xsmpgm FMUw>Lmoopmou .Am>v Xemagm Fmsampcm> .Anuv xuon uwuosmu an» op umfimgamu Amov sumpcm quwawuuo .Amuv macwm uwuosmu any op umpmnamu Amory agwuLm uwpoemu chcmpcw mgu ucaogm emUMFQ czonm mew macauemwF mmsammmca uwucom come M. Amxmv mcammmga cowmzmcmq com; :wmucwms op umpmanum ermPpwcw mm: zope coan "Amumv mmwgougm uwuosmu Pmccmuxm Amcmumgumu ommumcmpzcuwaoav cmmeaccmu we» op Amv sung swam: comm m zmaogcp mummzmgma mg“ nmem>wpme Amav gang acoumm < .Amv Loumswammg csm>smz a on uwuumccoo mm: masons» one .mIEE m we A>aav mezmmmsa mzo:m> asecoepaa come mum cwmwcwms op “cmmuwmmam mums m we Abuuv asap Pmmcoagoumsuxw esp we zydeco xsmcoepaa ewes mg» ope? Amwv zgwucm Pesoeme one See» AP; n F geaav Bonsaa we: voopm .cowpmceamca cowmaesma end: .P wcamwd 53 CEO ..\¢%\ Do. >QQ 54 begun thus ensuring adequate brain perfusion at all times. It was neces- sary to ligate the costocervical and omocervical arteries because they frequently reanastamose with the vertebral artery as this vessel passes cephalad leading to a reconstitution of flow to the Circle of Willis (244). The ventral spinal artery is primarily supplied through branches from the vertebral arteries (160) and because these anastomoses occur cephalad to the ligatures, the ventral spinal artery was not ligated. A lung was obtained from a donor dog weighing 9-12 kg. Approximately one hour prior to anesthesia the dog was given 300 mg aspirin intravenously to block prostaglandin synthesis that may occur in hypoxic lungs (203). The dog was anesthetized with sodium thiamylal (17.6 mg/kg) and given heparin sodium (10 mg/kg) for anticoagulation. Sufficient blood was removed from the donor dog to prime the extracorporeal system. The heart and lungs were removed and suspended from a large bore latex cannula inserted into the left atrium. Total lung ischemia time was less than 20 minutes. Blood was pumped from the right femoral artery of the experimental animal into the pulmonary artery of the extracorporeal lung at a rate sufficient to maintain its mean pulmonary venous pressure at 5 mmHg. A second pump delivered this perfusate from the extracorporeal lung, through a 38°C water bath, and then to the cannulated external carotid arteries (Figure 1). Blood flow was initially adjusted to maintain head perfusion pressure 3_mean aortic pressure during both constant pressure and constant flow experiments. Since the head perfusion preparation used in this investigation has not been previously described, it is necessary to provide evidence that 55 blood delivered to the external carotid arteries reaches the brain, that other collateral sources of blood flow are virtually excluded, and that the preparation is viable. First, perfusing the head with hypercapnic blood caused slowing of heart rate and movement of respiratory muscles out of phase with mechanical ventilation (before neuromuscular blockade), suggest- ing that the medullary chemoreceptors were being perfused with this blood. Second, hypercapnic blood (PCO2 75 to 85 torr) caused head perfusion pressure to decrease approximately 30%. This means that vascular resist- ance fell an equal amount because blood flow was constant. Since C02 is a potent dilator of cerebral vessels (97) and a weak dilator of vessels supplying skin and skeletal muscle (58), it is likely that a major portion of this fall in vascular resistance was due to dilation of cerebral vessels. Third, when flow into the external carotid arteries was stopped, head perfusion pressure fell to less than 20 mmHg and was nonpulsatile suggesting a virtual absence of collateral flow (Figure 2). Fourth, a central nervous system ischemic response was elicited when flow through the external carotid arteries was stopped (Figure 2). This response is characterized by massive sympathetic outflow to the cardiovascular system causing large increases in mean aortic pressure, thus attempting to relieve brain ischemia (104). Fifth, infusion of 5 m1 of methylene blue dye into the external carotid arteries resulted in diffuse homogenous staining of the brain. When this dye was injected into the dog's left atrium little or no dye was observed in the brain providing head perfusion pressure was approximately equal to mean aortic pressure; however, dye was detectable in the cerebellum and pontomedullary area of some dogs when flow through the external carotid artery was zero. The latter conditions 56 Figure 2. Tracing of the cardiovascular response following cessation of flow through the external carotid arteries. Notice the large oscillations in vascular pressures that are characteristic of the central nervous system ischemic response. PLV’ left ventricular pressure; LV dP/dt, left ventricular contractility; PPA’ mean pulmonary artery pressure; PLA’ left atrial pressure; PHP’ head perfusion ressure, P , mean aortic ressure. MA 57 PLv MM Hg LV dP/dT MMHg/SEC MMHg PLA MMHg PHP MMHg PMA MM Hg l—5 MlN——I Figure 2 58 did not exist during brain hypoxia and/or hypercapnia since flow through the external carotid artery either increased (during constant pressure experiments) or was held constant. In order to change the gas tensions of blood perfusing the brain, the extracorporeal lung was ventilated at a tidal volume of 500 ml and rate of 12 strokes/minute with the following: 1) a control gas mixture containing 15% 02, 5% C02, 80% N2; 2) a mildly hypoxic mixture contain- ing 5% 02, 5% C02, 90% N2; 3) a moderately hypoxic mixture containing 2.5% 02, 5% C0 0% 0 2, 92.5% N2; 4) a severely hypoxic mixture containing 5% C02, 95% N2; 5) a hypercapnic mixture containing 15% 02, 29 15% C02, 70% N2; and 6) a hypoxic-hypercapnic mixture containing 0% 02, 15% C0 85% N2. The extracorporeal lung was ventilated with one of the 29 above gas mixtures for 5 minutes; each test gas was preceded by a ventila- tion period with the control gas mixture for at least 5 minutes. Eighty- seven paired and randomized treatments were made on 31 dogs. No fluids or drugs were administered during data collection periods. Under conditions of constant head perfusion pressure, the extra- corporeal lung was ventilated with a severely hypoxic gas mixture (N = 10) with the dogs' vagus and carotid sinus nerves intact (intact nerve group). During constant flow studies, the extracorporeal lung was ventilated with a mildly hypoxic (N = 11), a moderately hypoxic (N = 12), a severely hypoxic (N = 10), a hypercapnic (N = 10), or hypoxic-hypercapnic (N = 10) gas mixture with the vagus and carotid sinus nerves intact. Following cervical vagotomy and carotid sinus nerve section (cut nerve group) the extracorporeal lung was ventilated with a severely hypoxic (N = 10), a hypercapnic (N = 7) or a hypoxic-hypercapnic (N = 7) gas mixture. 59 Total pulmonary (PVRT), pre-large venous (PVRP) and large venous resistances (PVRV) were calculated by dividing the appropriate pressure gradient by cardiac output as follows: (PPA" PLA) / 0 = PVRT, (P53 - PPAW) / O = PVRP and (PPAw - PLA) / O = PVRV; where P55 equals mean pulmonary artery pressure, PLA equals left atrial pressure, PPAW equals pulmonary artery wedge pressure, 0 equals cardiac output. The pulmonary artery wedge pressure represents downstream pressure at the point of the first post-capillary collateral venous vessel (82). Therefore, PVRP represented resistance proximal to the point of the first collateral venous vessel and PVRv represented resistance distal to the point of the first collateral venous vessel (Appendix A). Total peripheral resistance was calculated by dividing PMA/ 0, where PMA equals mean aortic pressure. Statistical analyses were performed using 2-way analysis of vari- ance (ANOVA) with comparison between means based on Student—Newman-Keuls' test. Data are expressed as mean i standard error (SE). Treatment means were considered significantly different at Ps:0.05. Thermal-Conductivity_Method Using the thermal-conductivity method, additional experiments were perfbrmed in order to determine lung extravascular thermal volume and central blood volume during brain hypoxia. A 5F polyethylene catheter (Edslab Thermodilution Catheter, Edwards Laboratories, Inc., Santa Ana, CA) was introduced into the femoral artery and advanced into the aorta such that its tip was within 5 cm of the aortic valve (determined by palpation). This catheter had a thermistor positioned 0.5 cm from the tip to detect changes in blood temperature. Two platinum ring electrodes 60 positioned 1.0 and 1.5 cm from the tip detected changes in blood conductiv- ity. A 7F Swan-Ganz catheter was inserted into the jugular vein and advanced into the pulmonary artery. When the balloon near the catheter tip was inflated the pulmonary artery wedge pressure was obtained. A second Swan-Ganz angiography catheter (Swan-Ganz Flow Directed Pulmonary Angiography Catheter, Edwards Laboratories, Inc., Santa Ana, CA) was inserted through the same jugular vein and advanced into the right ven- tricle. This catheter had eight holes near its tip, thus facilitating rapid injection and mixing of indicator solution. Swan-Ganz catheter positions were verified by pressure recordings. Cardiac output was measured by the thermodilution technique. Ten ml 3% saline (23-25°C) was rapidly injected into the right ventricle through the angiography catheter. A second cardiac output determination was made after blood temperature had returned to a stable baseline. The thermistor and platinum electrodes positioned near the tip of the aortic catheter detected changes in blood temperature and conductivity, respec- tively. The dilution curves were recorded on a polygraph recorder (Model 7784A, Hewlett-Packard, Farmington Hills, MI) (Figure 3). Using a data digitizer (graf/pen, Science Accessories Corporation, Southport, CT) these curves were sampled every 200 msec to the point where recirculation occurred, which was determined from a log-concentration time curve. These points were fit by a computer (Model POP 11/34, Digital Equipment Corporation, Maynard, MS) to a single exponential decay to time infinity. Area under each curve was integrated and the following equation was used to calculate cardiac output (0): 61 [v h1 [AT] Ed. x s..] 6 = inj - vcat inj S1nj Ct dt d 1 500°() [bxsb where vinj equals volume of sal1ne 1njected. Vcath equals volume of saline in catheter at time of injection, AT equals the difference between blood and injectate temperature in degrees centigrade, dinj equals dens- ity of injectate,s . equals specific heat of injectate, db equals sinJ equals specific heat of blood, .5 C(t) dt equals density of blood, 5b the integral of blood temperature change in degrees centigrade multi- plied by time (t). The product of [dinjx Sian/[db x s b] and catheter heat loss was assumed to equal 1.0. Mean transit time (t) for each curve was calculated as the first moment of the curve from initiation of injection using the fbllowing equation: ‘5‘“) o thLtlIdt CK) lg C(t) dt The product of cardiac output and heat mean transit time was taken as E'= total tissue volume. The product of cardiac output and conductivity (Na+) mean transit time was taken as intravascular volume (i.e., central blood volume). Lung extravascular thermal volume was obtained by subtracting intravascular volume from total lung volume. After normalization for body weight, central blood volume and lung extravascular thermal volume were expressed in ml/kg. In using this technique I made two assumptions. First, I assumed the sodium in the injectate was confined to the vascular compartment dur- ing its initial passage through the lung and that all sodium injected was 62 recovered. In support of this assumption, Chinard et al. (41) reported 22Na+ dilution curves for the mean transit time and shape of T-1824 and pulmonary vascular bed to be almost identical and concluded that both substances are essentially confined to the intravascular compartment dur- ing their initial passage. Furthermore, Pearce (181) reported virtually 24Na+ in normal lungs and only 4.5% loss in high complete recovery of pressure pulmonary edema while Noble et al. (174) found no detectable loss of sodium during pulmonary edema. The second assumption I make is that there is no irreversible loss of heat into the lung. This has previously been shown to be the case even at low cardiac output (8). Furthermore, Pearce and Beazell (182) reported that lung extravascular thermal volume was unaffected by the air in the lungs presumably because of the great disparity in specific heat of alveolar air and lung tissue. Although the thermal-conductivity method for measurement of lung water has been established as a valid technique (173,174), I felt it was necessary to verify the method in our laboratory prior to utilizing this technique during brain hypoxia experiments. In 11 dogs a 5F balloon-tip Foley catheter was inserted directly into the left atrium of open-chested dogs. The balloon was inflated sufficiently to raise left atrial pres- sure to approximately 30 mmHg for l to 2 hours. In most dogs this pro- cedure caused various degrees of pulmonary edema and therefore increased lung extravascular thermal volume. At the completion of each experiment all lung tissue was rapidly excised at the hila. Analysis of lung compo- sition was according to the method of Pearce et al. (183) as modified by Noble and Severinghaus (173). Briefly, the excised lungs were passively drained of blood, weighed and homogenized with 200 ml distilled water in 63 .FE mPP u .cwe\4 em.m u paauao uewvgmu mow mega :H s>ea azocataea .aam we.“ 1 a» .aa. oa.op 1 IS .xeumu meucmcoaxm mpmcwm on» Co cowumpoamcuxm cm “commend; muoo .me:Fo> Passes“ LmFsumm>mgpxm acaF .4>bm “weep uwmcmsu cams A+mzv U xpw>wuuaccou . u mme_u pwmcesu some and; .:p ”mucmgmmmmg uwsgo .c "wuegmwgpcmu momemmc .ua .xpw>_uu:n:ou use mcaumcmaemu coopn cw mmmcecu cmpumpmu mason pmswxoca as» aw Loumcumu »u_>wpu:ucou use Passage < .mpuwcucm> “saws mg» can? umuomncw mum: Fonz gm FE cob .mm>czu Aeouuonv comu:__u-aum>Ppu=ecou use Agony cowpapwuupmsgmgu mo mcwumsk .m assume 64 m mszmwd 65 a blender. Aliquots of homogenate were centrifuged at 15,000 rpms for 30 min producing a clear red supernatant. Hemoglobin determinations were made on systemic arterial blood (HbB) and supernatant (Hbs) by the cyanmethemoglobin method using a spectrophotometer (Model 24, Beckman Instruments Inc., Fullerton, CA ). Aliquots of systemic arterial blood and homogenate (Homog) were dried to constant weight (Wt) in an 85°C oven. The following formulae were used to calculate lung composition: lung blood wt = (Homog wt) (%H20 in Homog/lOO) (HbS/HbB); pulmonary extravascular tissue weight (PETW) = (lung wt) - (lung blood wt); extravascular lung water (EVLW) = (Homog wt) (%H20 in Homog/lOO) - (lung blood wt) (%H20 in blood/100) - (200 ml H20); % H20 in extravascular lung = (EVLW/PETW) (100); and extravascular dry weight (EVDW) = PETW - EVLW. The amount of extravascular lung water and extravascular dry weight in each lung was expressed as a ratio EVLW/EVDW, after correction for residual blood volume. Pulmonary extravascular tissue weight was correlated with lung extravascular thermal volume determinations made immediately before the lungs were excised. Figure 4 shows that a good correlation existed between pulmonary extravascular tissue weight and lung extravascular thermal volume (r = 0.88; P<:0.0l) with a regression equation of: Lung Extravascular Thermal Volume = 1.07 (Pulmonary Extravascular Tissue Weight) - 0.43. Since the thermal-conductivity method proved to be a reliable technique in our laboratory, I utilized this technique in two additional groups of brain hypoxia experiments. In one group of dogs lung 66 .FF u z .mmcezam umemp co eczema we» so comma .uvm amen mo mcw— any we mcw— uwpom .zuwucmue Co mcwp we“ we weep :mxogm .memum agmcospan we mmmgmmc maowsm> muavogn op uwmmwgucp we: mcammmca Fewsue veep sows: cw mace soc; Azhmav agape: damn?» smFaomm>msuxm xsmcoepan mamgm> A4>va we:~o> passes“ cepzume>eguxm m::— we cowmmwcmwc Leone; .¢ mcamwd 67 _§:e 35.. m. o. .2 m: o. m m at AV h L b b P b b D Q \ o \\ \ \\\ A \ 1¢ .m em .0. .N. \ 6.9a 3.0». .3 9.0 -35: 8.. u ._>E o um- \\ x o 'm. 3:5 4>em 68 extravascular thermal volume and central blood volume was determined at 0 and 5 minutes of severe brain hypoxia (N = 6). In a second group these parameters were measured at 0, 30, 60, 90, and 120 minutes of severe brain hypoxia (N = 6). In both groups the vagus and carotid sinus nerves were cut and flow through the external carotid arteries was constant. Following 5 minutes or 2 hours of brain hypoxia the lungs were excised at the hilum and lung composition was determined as described earlier. IV. Carotid Body Hypoxia and Hypercapnia Hypoxic, hypercapnic, and hypoxic-hypercapnic stimulation of the carotid bodies was achieved by pumping arterial autologous blood through the extracorporeal lung ventilated with appropriate gas mixtures (Figure 5). Outflow from the extracorporeal lung was pumped to the common carotid arteries at constant flow and pressure. In order to isolate and perfuse the carotid bodies, the following vessels were bilaterally ligated: internal carotid, occipital, lingual, ascending pharyngeal, and cranial laryngeal arteries. The internal carotid and occipital arteries were ligated cephalad to the carotid sinuses and bodies, respectively (Figure 5). Meticulous dissection preserved the carotid sinus nerves and their func- tional integrity was assumed if arterial pressure increased during temporary bilateral common carotid artery occlusion and if movement of respiratory muscles occurred (before neuromuscular blockade) out of phase with mechanical ventilation during hypoxemic perfusion of the carotid bodies. The preparation was assumed to be isolated when cessation of flow through the common carotid arteries caused perfusion pressure to decrease to < 20 mmHg without pulsations. 69 The extracorporeal lung preparation was identical to that described for the brain studies with the following exceptions. Perfusate from the extracorporeal lung was delivered to the cannulated common carotid arteries. Blood flow was maintained constant (100 to 125 ml/min) as was perfusion pressure. The latter was set equal to mean aortic pressure by adjusting outflow resistance with a screw clamp. The effluent perfusate was collected from the external carotid arteries and returned to the dog via the right femoral vein (Figure 5). Ventilatory parameters for the extracorporeal lung and test gas mixtures were identical to those used in the brain hypoxia and/or hyper- capnia experiments. Ninety paired treatments were made on 30 dogs. With the vagi intact the extracorporeal lung was ventilated with a severely hypoxic gas mixture (N = 12). Following vagotomy the extracorporeal lung was ventilated with a mildly hypoxic (N = 14), a moderately hypoxic (N = 14), a severely hypoxic (N = 10), a hypercapnic (N = 13) or a hypoxic-hypercapnic (N = 10) gas mixture. The purpose of vagotomy was to eliminate buffering effects of the aortic baroreceptors and cardiac stretch receptors on the cardiovascular system (180). In a separate set of experiments, after vagotomy, phentolamine (1 mg/kg) was infused as a slow intravenous bolus to block alpha adrenergic receptors during severe carotid body hypoxia (N = 7) and hypoxia-hyper- capnia (N = 10). An alpha agonist, phenylephrine (100 pg) or norepine- phrine (10 pg), was given intravenously as a bolus both before and after phentolamine. Adequate blockade was assumed when the response (i.e., increased mean aortic and pulmonary artery pressures) to the agonist given before alpha blockade was abolished after alpha blockade. In all 70 Figure 5. Carotid body perfusion preparation. Blood was pumped (Pump 1 = P1) from the femoral artery (fa) into the main pulmonary artery of the extracorporeal lung (ECL) at a rate sufficient to main- tain its mean pulmonary venous pressure (va) at 5 mmHg. The trachea was connected to a Harvard respirator (R). A second pump (P2) delivered the perfusate through a 38°C water bath (8) to the cannu- lated common carotid arteries (cca). Blood flow and perfusion presure (Pp) were maintained constant. Perfusate was collected from the cannulated external carotid arteries (eca) and returned to the dog via the femoral vein (fv). A screw clamp adjusted outflow resistance. Ligatures are shown placed around the eca, internal carotid artery (ica) cephalad to the carotid sinus (cs), and the occipital artery (oa) cephalad to the carotid body (cb). 71 fo Figure 5 72 experiments cardtid sinus perfusion and pulse pressures were constant between control and test gas period. Statistical analysis and vascular resistance calculations were identical to those used for the brain perfusion studies. V. Histamine and Adrenergic Receptors In closed-chest dogs the thermal-conductivity method was used to determine cardiac output, lung extravascular thermal volume, and central blood volume. The theory, calculations, and methods applicable to this procedure were discussed earlier. Aortic, mean pulmonary artery, and pulmonary artery wedge pressures were measured using Statham P23Gb pressure transducers referred to the level of the left atrium. All recordings were made continuously on a polygraph recorder (Model 7796A, Hewlett-Packard, Farmington Hills, MI). A 20 gauge needle was inserted into the oral end of an endotracheal tube. A polyethylene catheter connected this needle to a Statham PM 131 pressure transducer (Gould Medical Products, Statham Industries, Oxnard, CA) for measurement of airway opening pressure. Pre-large vein pulmonary vascular resistance was calculated as (Pfifi-- PPAW)/CI’ where Pfifi equals mean pulmonary artery pressure, PPAW equals pulmonary artery wedge pressure and CI equals cardiac index expressed in ml/sec/kg. Total peripheral resistance was calculated as PMA/CI where PMA equals mean aortic pressure. Data collection was begun approximately 1 hour after the onset of anesthesia when all vascular pressures were in a steady state for at 73 least 20 min. Thermal and conductivity curves were obtained at -20, 0, 30, 60 and 90 minutes. In the time control group (N = 10) no vasoactive drugs were administered during the 110 min protocol. In the histamine group (N = 8), histamine base (10 ug/kg/min) was infused intravenously between 0 and 90 minutes. In the propranolol-histamine group (N = 11), propranolol (2 mg/kg) was infused intravenously between -20 and -15 min and then at 20 ug/kg/min for the remainder of the experiment (i.e., 105 min). A beta agonist, isoproterenol (10 pg), was given intravenously as a bolus both before and after infusion of 2 mg/kg propranolol. Adequate blockade was assumed when the response to the agonist (i.e., increased heart rate) given before beta blockade was abolished after beta blockade. In the phentolamine-histamine group (N = 6), phentolamine (1 mg/kg) was infused intravenously between -20 and -15 min and then at 10 pg/kg/min for the remainder of the experiment. An alpha agonist, phenylephrine (100 pg), was given intravenously as a bolus both before and after infu- sion of 1 mg/kg phentolamine. Adequate blockade was assumed when the response to the agonist (i.e., increased mean aortic and pulmonary artery pressures) given before alpha blockade was abolished after alpha blockade. In the propranolol—histamine and phentolamine-histamine groups, hist- amine base (10 pg/kg/min) was infused intravenously between 0 and 90 minutes. At the completion of each experiment the dogs were euthanized by intravenous injection of 20 ml saturated KCl solution (6 dogs in time control and 6 in propranolol-histamine groups) or sodium pentobartital (all other dogs). The chest was immediately opened and all lung tissues 74 were rapidly excised at the hilum. Analysis of lung composition was as described earlier. Statistical analysis over the 110 minute protocol was performed using 2-way analysis of variance (ANOVA) with comparison of means based on Student-Newman-Keuls' test. An unpaired Students t-test was used to compare the means between groups of dogs. Means were considered significantly different at P‘:0.05. RESULTS I. Brain Hypoxia and Hypercapnia Vagus and Carotid Sinus Nerves Intact Table 1 shows the cardiovascular effects of perfusing the brain with severely hypoxemic blood under constant pressure conditions (perfusion pressure - 111 i 2 mmHg). Perfusate P02 decreased from 96 to 27 torr, PCO2 decreased from 31 to 29 torr and pH increased from 7.37 to 7.41. Flow through the external carotid arteries increased 66% from 104 to 173 ml/min. None of the other measured or calculated variables changed. Tables 2-4 show the cardiovascular effects of perfusing the brain with mildly, moderately, and severely hypoxemic blood in which head flow was constant at 179 i 22, 194 i 22, and 238 i 25 m1/min, respectively. During mild hypoxia perfusate P02 decreased from 112 to 54 torr, during moderate hypoxia from 122 to 40 torr, and during severe hypoxia from 125 to 29 torr. Perfusate pH increased from 7.37 to 7.39 during moderate hypoxia while systemic pH and PaO2 decreased from 7.39 to 7.36 and 106 torr to 97 torr, respectively. During severe brain hypoxia systemic pH and PaO2 decreased from 7.38 to 7.35 and 98 torr to 87 torr, respectively, while PaCO2 increased from 35 to 39 torr. Head perfusion pressure decreased 6% (from 122 to 115 mmHg), 23% (from 127 to 98 mmHg), and 31% (from 143 to 99 mmHg) during perfusion of the brain with mildly, moderate- ly, and severely hypoxemic blood, respectively. None of the other 75 76 measured or calculated variables changed during mild, moderate, or severe brain hypoxia. The cardiovascular effects of perfusing the brain with hypercapnic blood under constant flow conditions (head flow = 218 i 19 m1/min) are shown in Table 5. During this treatment perfusate PCO2 increased from 33 to 79 torr and pH decreased from 7.39 to 7.15. Head perfusion pressure decreased 30% from 139 to 97 mmHg. Stroke volume increased 12% from 11.0 to 12.3 ml, heart rate decreased 6% from 156 to 147 beats/min, and mean aortic pressure increased 10% from 80 to 88 mmHg. Systemic pH decreased from 7.38 to 7.34. None of the other measured or calculated variables changed. Table 6 shows the cardiovascular effects of hypoxic-hypercapnic blood perfusing the brain. In this series blood flow was constant (head flow = 271 i 26 ml/min) and perfusate P02 and pH decreased from 121 torr to 33 torr and 7.43 to 7.18, respectively, while PCO2 increased from 31 to 73 torr. Head perfusion pressure decreased 38% from 138 to 85 mmHg. Stroke volume increased 20% from 10.7 to 12.8 ml while heart rate decreased 8% from 162 to 149 beats/min. Systemic pH and Pa02 decreased from 7.39 to 7.33 and 99 torr to 87 torr, respectively, while PaCO2 increased from 34 to 40 torr. All other measured or calculated variables did not change. 77 Table 1. Cardiovascular effects of perfusing the brain with severely hypoxemic blood under constant pressure conditions (head perfusion pressure = 111 i 2 mmHg) with the vagus and carotid sinus nerves intact. The extracorporeal lung was ventilated with a control gas mixture con- taining 15% 02, 5% C02, 80% N2 and a hypoxic mixture containing 0% 02, 5% C02, 95% N2. Values are means 1 standard error (SE) based on error mean square from ANOVA, N = 10. *, significantly different from control at P <0.05; Pfifi, mean pulmonary artery pressure; PPAW’ pulmonary artery wedge pressure; PLA’ left atrial pressure; PPAP’ pulmonary artery pulse pressure; PVRT, total pulmonary vascular resistance; PVRP, pre-large vein pulmonary vascular resistance; PVRV, large vein pulmonary vascular resistance; 0, cardiac output; HR, heart rate; V , stroke volume; PMA’ S mean aortic pressure; TPR, total peripheral resistance. 78 Table 1 Control Hypoxia SE Perfusate P02 (torr) 96 27 3.50* Perfusate PCO2 (torr) 31 29 0.40* Perfusate pH 7.37 7.41 0.01* Head Blood Flow (ml/min) 104 173 4.40* FEE-(mmHg) 14.4 14.3 0.10 PPAW (mmHg) 6.3 6.4 0.10 PLA (mmHg) 3.7 3.7 0.10 PPAP (mmHg) 8.4 8.1 0.10 PVRT (mmHg/L/min) 4.40 4.41 0.10 PVRP (mmHg/L/min) 3.28 3.24 0.10 PVRv (mmHg/L/min) 1.12 1.17 0.04 6 (L/min) 2.71 2.72 0.04 HR (beats/min) 157 156 1.80 VS (ml) 17.8 18.0 0.20 PMA (mmHg) 106 105 0.40 TPR (mmHg/L/min) 42.3 41.6 0.40 PaO2 (torr) 102 95 2.20 PaCO2 (torr) 34 35 0.80 pH (systemic) 7.36 7.36 0.01 79 Table 2. Cardiovascular effects of perfusing the brain with mildly hypoxemic blood under constant flow conditions (head flow = 179 i 22 m1/min) with the vagus and carotid sinus nerves intact. The extra- corporeal lung was ventilated with a control gas mixture containing 15% 02, 5% C02, 80% N2 and a hypoxic mixture containing 5% 02, 5% 002, 90% N2. Values are means i standard error (SE) based on error mean square from ANOVA. N = 11, unless otherwise indicated. *, signifi- cantly different from control at P<:0.05; PHP’ head perfusion pressure; PPA’ mean pulmonary artery pressure; PPAW’ pulmonary artery wedge pres— sure; PLA’ left atrial pressure; PPAP’ pulmonary artery pulse pressure; PVRT, total pulmonary vascular resistance; PVRP, pre-large vein pulmonary vascular resistance; PVRV, large vein pulmonary vascular resistance; 0, cardiac output; HR, heart rate; VS’ stroke volume; dP/dt, left ventricular contractility (N = 6); PMA’ mean aortic pressure; TPR, total peripheral resistance. 80 Table 2 Control Hypoxia SE Perfusate P02 (torr) 112 54 4.60* Perfusate PCO2 (torr) 35 35 1.00 Perfusate pH 7.37 7.38 0.01 PHP (mmHg) 122 115 1.10* Pfifi-(mmHg) 11.7 11.6 0.10 PPAW (mmHg) 5.4 5.4 0.20 PLA (mmHg) 2.7 2.5 0.10 PPAP (mmHg) 11.1 10.8 0.20 PVRT (mmHg/L/min) 5.63 5.81 0.10 PVRP (mmHg/L/min) 3.89 4.02 0.10 PVRv (mmHg/L/min) 1.74 1.79 0.10 6 (L/min) 1.72 1.69 0.03 HR (beats/min) 165 162 1.20 VS (m1) 10.5 10.4 0.10 dP/dt (mmHg/sec) 2000 1833 62 PMA (mmHg) 86 83 0.90 TPR (mmHg/L/min) 51.7 51.3 0.90 Pa02 (torr) 110 104 2.60 PaCO2 (torr) 33 33 1.10 pH (systemic) 7.39 7.38 0.01 81 Table 3. Cardiovascular effects of perfusing the brain with moderately' hypoxemic blood under constant flow conditions (head flow = 194 t 22 ml/min) with the vagus and carotid sinus nerves intact. The extra- corporeal lung was ventilated with a control gas mixture containing 15% 02, 5% 002, 80% N2 and a hypoxic mixture containing 2.5% 02, 5% C02, 92.5% N2. Values are means i standard error (SE) based on error mean square from ANOVA. N = 12, unless otherwise indicated. *, signifi- cantly different from control at P<:0.05; PHP’ head perfusion pressure; P53, mean pulmonary artery pressure; PPAW’ pulmonary artery wedge pressure; PLA’ left atrial pressure; PPAP’ pulmonary artery pulse pressure; PVRT, total pulmonary vascular resistance; PVRP, pre-large vein pulmonary vascular resistance; PVRV, large vein pulmonary vascular resistance; 0, cardiac output; HR, heart rate; VS, stroke volume; dP/dt, left ventricular contractility (N = 6); PMA’ mean aortic pres- sure; TPR, total peripheral resistance. 82 Table 3 Control Hypoxia SE Perfusate PO2 (torr) 122 40 3.60* Perfusate PCO2 (torr) 35 35 0.70 Perfusate pH 7.37 7.39 0.01* PHP (mmHg) 127 98 5.30* PER-(mmHg) 12.2 12.0 0.10 PPAW (mmHg) 5.7 5.4 0.10 PLA (mmHg) 2.8 2.6 0.10 PPAP (mmHg) 10.7 10.3 0.20 PVRT (mmHg/L/min) 5.79 5.80 0.10 PVRP (mmHg/L/min) 4.01 4.08 0.10 PVRv (mmHg/L/min) 1.78 1.72 0.10 0 (L/min) 1.72 1.70 0.03 HR (beats/min) 161 157 1.10 VS (m1) 10.7 10.8 0.20 dP/dt (mmHg/sec) 1583 1521 21 PMA (mmHg) 77 75 0.70 TPR (mmHg/L/min) 46.7 46.5 0.70 P302 (torr) 106 97 1.50* PaCO2 (torr) 34 34 0.70 pH (systemic) 7.39 7.36 0.01* 83 Table 4. Cardiovascular effects of perfusing the brain with severely hypoxemic blood under constant flow conditions (head flow = 238 i 25 ml/min) with the vagus and carotid sinus nerves intact. The extra- corporeal lung was ventilated with a control gas mixture containing 15% 02, 5% C02, 80% N2 and a hypoxic mixture containing 0% 02, 5% 002, 95% N2. Values are means i standard error (SE) based on error mean square from ANOVA. N = 10, unless otherwise indicated. *, signifi- cantly different from control at P<:0.05; PHP’ head perfusion pressure; PPA’ mean pulmonary artery pressure; PPAW’ pulmonary artery wedge pressure; PLA’ left atrial pressure; PPAP’ pulmonary artery pulse pressure; PVRT, total pulmonary vascular resistance; PVRP, pre-large vein pulmonary vascular resistance; PVRV, large vein pulmonary vascular resistance; 0, cardiac output; HR, heart rate; VS, stroke volume; dP/dt, left ventricular contractility (N = 8); PMA’ mean aortic pres- sure; TPR, total peripheral resistance. 84 Table 4 Control Hypoxia SE Perfusate P02 (torr) 125 29 4.70* Perfusate PCO2 (torr) 34 34 0.80 Perfusate pH 7.41 7.43 0.01 PHP (mmHg) 143 99 4.30* PPA (mmHg) 12.0 12.2 0.20 PPAW (mmHg) 5.8 5.8 0.10 PLA (mmHg) 1.7 1.5 0.10 PPAP (mmHg) 10.7 11.0 0.20 PVRT (mmHg/L/min) 5.74 5.88 0.20 PVRP (mmHg/L/min) 3.49 3.60 0.20 PVRv (mmHg/L/min) 2.25 2.28 0.10 0 (L/min) 1.86 1.88 0.04 HR (beats/min) 176 180 3.20 VS (m1) 10.7 10.6 0.20 dP/dt (mmHg/sec) 1813 1947 84 PM“ (mmHg) 79 84 2.30 TPR (mmHg/L/min) 43.8 45.8 2.00 PaO2 (torr) 98 87 1.40* PaCO2 (torr) 35 39 0.70* pH (systemic) 7.38 7.35 0.01* 85 Table 5. Cardiovascular effects of perfusing the brain with hypercapnic blood under constant flow conditions (head flow = 218 i 19 ml/min) with the vagus and carotid sinus nerves intact. The extracorporeal lung was ventilated with a control gas mixture containing 15% 02, 5% C02, 80% N2 and a hypercapnic mixture containing 15% 02, 15% C02, 70% N2. Values are means i standard error (SE) based on error mean square from ANOVA. N = 10, unless otherwise indicated. *, significantly different from control at Pl<0.05; PHP’ head perfusion pressure; PPA’ mean pulmonary artery pressure; PPAW’ pulmonary artery wedge pressure; PLA’ left atrial pressure; PPAP’ pulmonary artery pulse pressure, PVRT, total pulmonary vascular resistance; PVRP, pre-large vein pulmonary vascular resistance; PVRV, large vein pulmonary vascular resistance; 0, cardiac output; HR, heart rate; VS, stroke volume; dP/dt, left ventricular contractility (N = 8); PMA’ mean aortic pressure; TPR, total peripheral resistance. 86 Table 5 ___ Control Hypercapnia SE Perfusate P02 (torr) 110 97 5.10 Perfusate PCO2 (torr) 33 79 2.50* Perfusate pH 7.39 7.15 0.01* PHP (mmHg) 139 97 3.50* Pix-(mmHg) 11.2 11.6 0.20 PM (mmHg) 4.4 4.5 0.10 PLA (mmHg) 1.3 1.3 0.04 PPAP (mmHg) 12.1 13.0 0.30 PVRT (mmHg/L/min) 5.85 5.77 0.20 PVRP (mmHg/L/min) 4.01 3.98 0.10 PVRV (mmHg/L/min) 1.84 1.79 0.10 0 (L/min) 1.71 1.80 0.05 HR (beats/min) 156 147 2.20* VS (ml) 11.0 12.3 0.40* dP/dt (mmHg/sec) 1866 1975 81 pm (mmHg) 80 88 2.20* TPR (mmHg/L/min) 47.9 50.7 1.30 PaO2 (torr) 94 95 1.80 PaCO2 (torr) 35 38 1.10 pH (systemic) 7.38 7.34 0.01* k 87 Table 6. Cardiovascular effects of perfusing the brain with hypoxemic— hypercapnic blood under constant flow conditions (head flow = 271 i 26 m1/min) with the vagus and carotid sinus nerves intact. The extra- corporeal lung was ventilated with a control gas mixture containing 15% 0 5% CO 80% N and a hypoxic-hypercapnic mixture containing 2’ 2 15% C02, 85% N2. Values are means i standard error (SE) based 2, 0% 02, on error mean square from ANOVA. N = 10, unless otherwise indicated, *, significantly different from control at P<:0.05; PHP’ head perfu- sion pressure; Pfifi, mean pulmonary artery pressure; PPAW’ pulmonary artery wedge pressure; PLA’ left atrial pressure; PPAP’ pulmonary artery pulse pressure; PVR , total pulmonary vascular resistance; PVRP, pre- T large vein pulmonary vascular resistance; PVRV, large vein pulmonary vascular resistance; Q. cardiac output; HR, heart rate; VS, stroke volume; dP/dt, left ventricular contractility (N = 8); PMA’ mean aortic pressure; TPR, total peripheral resistance. 88 Table 6 Control Hypoxia-Hypercapnia SE Perfusate P02 (torr) 121 33 5.80* Perfusate PCO2 (torr) 31 73 1.90* Perfusate pH 7.43 7.18 0.02* PHP (mmHg) 138 85 4.00* FEE-(mmHg) 10.8 11.2 0.20 PPAW (mmHg) 5.1 5.1 0.20 PLA (mmHg) 2.0 2.0 0.10 PPAP (mmHg) 10.6 11.0 0.20 PVRT (mmHg/L/min) 5.42 5.34 0.20 PVRP (mmHg/L/min) 3.45 3.54 0.10 PVRv (mmHg/L/min) 1.97 1.80 0.10 0 (L/min) 1.70 1.87 0.07 HR (beats/min) 162 149 3.20* Vs (m1) 10.7 12.8 0.50* dP/dt (mmHg/sec) 1566 1750 87 PMA (mmHg) 68 77 3.80 TPR (mmHg/L/min) 41.6 44.4 1.70 PaO2 (torr) 99 87 2.50* PaCO2 (torr) 34 40 1.00* pH (systemic) 7.39 7.33 0.01* 89 Vagus and Carotid Sinus Nerves Cut Figures 6-9 show the cardiovascular effects of perfusing the brain with severely hypoxemic blood under constant flow conditions (head flow = 259 i 16 ml/min) in which perfusate P02 was reduced from 101 to 22 torr (SE =5.70) and pH increased from 7.39 to 7.42 (SE = 0.01). Head perfusion pressure decreased 19% from 106 to 86 mmHg (SE = 3.40). Mean pulmonary artery pressure increased 14% from 11.1 to 12.6 mmHg (SE = 0.20), left atrial pressure 65% from 1.7 to 2.8 mmHg (SE = 0.30), and pulmonary artery pulse pressure 17% from 10.1 to 11.8 mmHg (SE = 0.50). Pulmonary vascular resistance values (i.e., PVRT, PVRP, PVRv) were not significantly different from control. Cardiac output increased 18% from 1.36 to 1.61 L/min (SE = 0.06), stroke volume 21% from 6.8 to 8.2 ml (SE = 0.30), dP/dt 41% from 1850 to 2600 mmHg/sec (SE = 126); mean aortic pressure increased 35% from 60 to 81 mmHg (SE = 3.30) and total peripheral resist- ance 13% from 46.2 to 52.1 mmHg/L/min (SE = 1.80). Systemic pH and PaO2 decreased from 7.38 to 7.34 (SE = 0.01) and 114 torr to 97 torr (SE = 4.00), respectively. None of the other measured or calculated variables changed. In a separate set of experiments, 5 min brain hypoxia caused no significant change in lung extravascular thermal volume (i.e., control = 8.27 ml/kg, hypoxia = 8.23 m1/kg, SE = 0.20) but central blood volume increased 18% from 12.5 to 14.7 ml/kg (SE = 0.30) (Figure 10). In this group the hemodynamic changes were similar to that reported in Figures 6-9. Postmortem extravascular lung water to extravascular dry weight ratio was 3.56 i 0.07, which is not significantly different from control values (3.64 i 0.14; N = 10) in our laboratory. 90 .mo.ouva um Fospcou soc» newcomewu xFuceuewwcmwm .4 mop n z .<>oz< Ease menacm came posed no women Lasso vsmccmum A memos mew mwapm> .acmumcou umcwmucmms we: zerm new: .920 mew: mm>gmc macwm kuogmo vce mamm> one .Aazav wsammmga commamsma new; use In new mcopmcmu mom mpmmamsma co Nz Nmm N .mou xm .mo go mcwcwmpcou wgauxwe u_xoaz; a ten 2 Row .mou Rm .No «mp mcwcwmucoo mcszws mam Focpcou e sue: acaF megonsoumsuxm on» mappePwucm> mo uummmu .m weaned 0N. 30E m wcammd 0.93 9Q” "' o ’9 .. V ‘b o 0:. 0'. O O. O‘O'O' ocwr coo» d’d’ 96%1 V 3‘ SE)... rcon 03.2 '6' co 4% V O O O 9.. O O v ‘W% o q¢ *1— nth. 5x8»: 2.48% Bo. 92 .mo.ouva pm Poeucou soc» pcmgmmwpu xpuceuwmwcmwm .« ”or u z .<>oz< seem menacm cams Loggm co comma Loscw usmwcmpm A mamas use .Nou em .No Nae manpa> .Nz Ram .Nou am .No so esteeapeou massxas uexoa»; a use Nz New mewcweucou weapxwe mam Focused m suwz emamepcm> we: wear Pmmcoagoumeuxm one .gau mew: mm>gmc m::_m uwuoemu vcm mamm> as“ new Acws\Fe mF A mmmv acmpmcou me: new; mg» op 30pm .Am>v m23Fo> mxogum use .Amzv wan; peed; .Auuxauv auwrwpumcucoo sepzowspcm> Hemp .on unease umwucmu co noopn owemxoa»; apmsm>mm new: :Pesn mg» mcwmawgma mo uumwum .n msamwd 93 .Om 5:582. .00. r.00. 100m ,Sx0a>zz§§m_ n meamwd 000. 0001 033165 mummmnw JOEFZOU-I... 0007 000w. comma x . I‘ \ V1 \. ~; K K ' X s>»%<, l . l I ‘J/ K\! ,' 'l ‘A YX‘ / \ v a? X X /\ / A‘f / >- '7 r X ‘1 XX 2‘ 7.25.” ‘ . >37.” I} f l l . I n in" \ / fll‘ ( 7" r- W /‘\)<. I . ,/ . / .oo._ 8.. .55) low; um»; 94 .mo.o vs we Posesou scam psmsmwwwu a~pseuwmwsmwm .« .m u z mews: 3oz< soc» mseucm sews gossm so ummen Locum ugeuseum H msews use waspe> .N2 gas .Nou Re .No as usesseesou essaxes u_xos»s e ese Nz Noe .Nou we .mo an mswsmepsou msaaxwe mem Focused e saw: umuepwusm> me: asap Pemsoacouesuxm use .926 mew; mm>sms maswm uwuoseu use mame> esp use AswE\FE up “.mmmv useumsou me: uem; ms» op 30pm .A<2sv msammmsq uwpgoe sews use .Aamm saw: swesn ms» mswmzmsmn Co pumemm .w acumen 95 0_.L :1 DIES «use»... 2.55 g 6......on I 0'9' '0 0.0.0:: v v .oyyovopv 3. V 0.9 o ’0 O'O'O' 0.0.. v '0'. O O 9292. o o o ’o v . '9 o 0: O .7 o 9 0:9. 0 " 96 .88; ee 33:8 5.: “seat; boseoetso: .5 .o u 2 was: > a m>s use m>s ooooxe .op u z .<>oz< Eosw oseoom sees Loses so uomeo posse useusepm H .N N mseos ose me=~e> z Nmm . 00 am .No No mswsweusoo esopxwe owxoszs e use Nz Now .Nou Rm .mo amp msmsweusoo osoost mem Fososoo e sow: uooepmoso> me: asap Feosossooespxe use .uso osez mo>sos mosmm uwuoseo use mome> esp use AswE\PE up M mmmv useumsoo me: ueoc esp op zops .Amsev ooseemwmos Fesesswsos Feuou use .A>m>sv moseomwmos sepoome> usesospos swo> omsep .Asm>sv mosepmwmos sepoome> xsesospoo swo> ease—loss .Ahm>sv ooseamwmms sepoume> xsesoE—os Feuop so uoops oesoxosas zposo>em sew: swess esp ms_momsos mo uoommm .m essays m osamws . Wm _ . . mm. 6%». . , 97 5x8»... 54% g .5528 I I -3. 5 5533155 “me -00 98 .mo.ou.s we Posusoo sosm usesomewu apuseowwssmwm .4 .m u z oses; 3s>u\zo>m useoxo .o u z .<>oz< 5osm oseoom sees sosse so uomeo sosse useuseem u mseme ose moope> .Nz Nmm .mou gm .No so mswsweusoo osouxwe owxosxs e use Nz Rom .Nou em .No amp msmsweusoo osoust mew ~osusoo e sow: uouepwpso> me: msoF Feesoosooesuxo use .eoo osoz mo>sos moswm uwuoseo use mome> esp use Ass5\_5 0F H mmmv useumsoo me: ueos use on sops .A3o>m\34>mv oeues psmwoz >.su sepoome> -esuxo op sooez asap sepoome>esexo use .A>muv oEoFo> uoopo Fespseo .A4>huv osoFo> Feesosp sepoome>espxo asap so uoopo oesoonAs aposm>em sue: sweso esp mswmomsoo so poemem .oP ossmws 99 536%... z.0.50). 139 .meseoom pmeop mo uospos esp so uemeo .ppp pmos so ospp esp mp espp uppom .zpppsoup so ospp esp mp ospp soxosm .soosm pospsoo oepp esp sp mmou 5osp Azhmsv psmpoz oommpp sepoome>espxo asesospoo momso> Ap>hmv eso—o> pessosp se—oome>espxo asap so sopmmosmes seospp .PP osomps 140 .s essops 935 2.5.. o. o e s o e e o p n h r P p pl)“ \\ o \\\ .1 \\ I? \ \\ \\ .e To a. po re so... so... .0. ~o..+3pm&.e.ou..>eu . 872.88 .258 55.... 9.25 5>pe 141 .mmgmacm ammu— $o nocume mg“ :0 ummma .uww pmmn mo mew? mg“ mp mcwp uwpom .auwucmuw mo mcpp mgp mm mcwp :mxogm .maaogm mewsmumw;-mcw5m~0pcmga can .mcwsmumwg-popocwgaogn .mcwEmumw; vwcwnsou any so» mmou cw fighmav pcmwmz mammwp Lopaumm>mguxm xgmcospzn mamgm> Ag>pmv manpo> FmEmeu Lm_:umm>mgpxm mcap we cowmmmgmmg gmmcwd .NP mgamwm 142 mp mgamwm 2:5 3 .53 o. m m h m 0 ¢ 0 p p p p p P - J\‘\|Ho \\ \\ ;\ \\\ ufiv 5.0» n. \\ vmd u . \ 9N +3kma¢mdfl>kw \\ .m Tm l. 935 4>km .m .m o Amuz. m5£o§I$£EoBE2Q< :0. 2 72. 8_E£m_1 . _o_oc2ao..m . Amuzv $560350 143 Extravascular Lunggwater to Extravascular Dry Weight Ratio No significant difference was found in extravascular lung water to extravascular dry weight ratio between time control and propranolol- histamine groups receiving the same euthanasia solution; however, a significant difference did exist depending on the type of euthanasia solution used. Time control dogs euthanized with KCl had an extra- vascular lung water to extravascular dry weight ratio of 2.87 i 0.17 whereas propranolol-histamine dogs had a ratio of 2.66 i 0.19. Because these ratios are lower than reported values (183), I began euthanizing subsequent dogs with sodium pentobarbital. Dogs in the time control group euthanized with sodium pentobarbital had an extravascular lung water to extravascular dry weight ratio of 3.97 i 0.17 whereas dogs in the propranolol-histamine group had a ratio of 3.88 i 0.12, dogs in the phentolamine-histamine group 3.98 i 0.09 and dogs in the histamine group 3.75 i 0.04. The extravascular lung water to extravascular dry weight ratio for dogs euthanized with KCl (i.e., 6 dogs in time control and 6 in propranolol-histamine groups) was 2.76 i 0.13 and with sodium pentobarbital (i.e., all other dogs) was 3.88 i 0.05 (P<:0.01). Because lung dry weight should not be affected by the type of euthanasia solution used, the most likely cause for this difference was reabsorption of extravascular lung water as the extremely hypertonic bolus of KCl passed through the lung suggesting that saturated KCl should be avoided for euthanasia purposes when performing investigations involving lung water determinations. DISCUSSION I. Brain Hypoxia and Hypercapnia Experimental Preparation In the experimental preparation of this investigation it was imperative that collateral (i.e., systemic) blood flow to the brain be negligible. Bilateral ligation of the common carotid and vertebral arteries often does not eliminate blood flow to the dog brain because of the extensive collateral blood supply (56,97,175,244). Daly et al. (56) reported that even following bilateral ligation of the vertebral and common carotid arteries, backflow pressure (from the Circle of Willis) in the common carotid artery could not be lowered below 60 mmHg suggest- ing maintenance of considerable collateral flow to the brain. Whisnant et al. (244), demonstrated that the costocervical and omocervical arter- ies are the main source of this collateral circulation because these vessels frequently reanastomose with the vertebral artery as the latter vessel courses cephalad. In my preparation these vessels were ligated. With zero flow through the external carotid arteries, head perfusion pressure decreased to < 20 mmHg and was nonpulsatile. This suggests that if any collateral flow remained it was minimal since normally when vertebral arteries and collateral vessels are intact, approximately 70% of the mean arterial pressure is transmitted back from the Circle of Willis to the common carotid artery (175). 144 145 In some of the cut nerve experiments, it was not possible under constant flow to maintain head perfusion pressure above mean aortic pres- sure since the former always decreased while the latter always increased during brain hypoxia and/or hypercapnia. This group nevertheless had the greatest systemic responses to brain hypoxia and/or hypercapnia. Such systemic responses would not be expected if collateral circulation was of any consequence. When the extracorporeal lung was ventilated with a hypoxic or hypoxic-hypercapnic gas mixture PaOZ and systemic pH decreased and Paco2 increased. Ventilation of the extracorporeal lung with a hypercapnic gas mixture caused no change in PaO2 while PaCO2 increased and systemic pH decreased. Thus, changes in systemic gas tensions and pH tended to follow changes in perfusate P02, P002 and pH, but were of much lesser magnitude. Previous work (77) using a similar extracorporeal system reported no changes in systemic gas tensions and pH. However, flow through the extracorporeal lung in my experiments was approximately two- fold greater and represented approximately 15% of the cardiac output. In the cut nerve group these small changes in systemic gas tensions and pH were of no consequence because the peripheral chemoreceptors were denervated. In the intact nerve group these changes are unlikely to have affected my results or interpretation since the reflex vascular effects of increased Paco2 on the peripheral chemoreceptors is considerably attenuated without a concomitant large decrease in pH (184) or ineffec- tive when PaOZ is approximately 100 torr (57). Although pH did decrease, the change was of small magnitude and P302 was 94 i 3 torr (Tables 1-6) during brain hypoxia and/or hypercapnia. 146 Intact Nerve Group In all experiments in which the vagus and carotid sinus nerves were intact, head perfusion pressure decreased during constant flow perfusion. Hypoxia-hypercapnia caused a 38%, hypoxia a 31%, and hyper- capnia a 30% decrease in head perfusion pressure. During constant head perfusion pressure experiments hypoxia caused a 66% increase in flow to the head. These responses can be attributed to the local vasodilator effect of hypoxia and hypercapnia (58,97). All systemic and pulmonary hemodynamic parameters measured were unchanged from control during hypoxia. Similar results were obtained during hypercapnia and hypoxia-hypercapnia except that the addition of 002 to the perfusate caused a slowing of heart rate. This cardiac effect of perfusing the brain with hypercapnic blood has been previously reported (56) and was attributed to stimulation of vagal nuclei in the medulla leading to increased vagal tone to the heart. Stroke volume increased during hypercapnia and hypoxia-hypercapnia probably because of a slower heart rate. The latter effect would increase filling time resulting in increased end-diastolic volume. The tendency of both cardiac output and total peripheral resistance to increase contributed to an overall increase in mean aortic pressure during hypercapnia. Cut Nerve Group(5 Minutes Brain Hypoxia) As in the intact nerve group hypoxic, hypercapnic, and hypoxic- hypercapnic blood caused local dilation of vessels supplying the head resulting in decreased head perfusion pressure. The increase in left ventricular dP/dt, cardiac output, total peripheral resistance, and mean aortic pressure observed during brain hypoxia and hypoxia-hypercapnia are 147 consistent with other studies (62,68) that demonstrate the effect of brain hypoxia on the systemic vasculature. Although brain hypoxia has also been reported to increase heart rate (68), I did not observe this probably because control heart rates were high. While brain hypercapnia did not increase cardiac output, it increased dP/dt, total peripheral resistance, and mean aortic pressure, effects previously demonstrated in dogs (68) and more recently, in the monkey (157). Total pulmonary vascular resistance was unaffected by any of the treatments despite increases in mean pulmonary artery and left atrial pressures. The net effect of these changes in pressure is to increase vascular transmural pressure, the passive effects of which would decrease PVRT. Therefore, it is possible that some active vasoconstriction occurred and prevented a significant decrease in PVRT. These data do not support the hypothesis that brain hypoxia increases PVRT (30,167) but are consistent with other investigations involving brain ischemia (96) or increased intracranial pressure (l44, 155); conditions which presumably cause brain hypoxia and also do not elevate PVRT (96,144,l55). Data from the cut nerve groups demonstrate that the major effect of brain hypoxia or hypoxia-hypercapnia on the pulmonary circulation is to increase pulmonary blood volume and secondarily increase microvascular hydrostatic pressure. The increase in central blood volume probably occurred as a result of increased left ventricular afterload as evidenced by an abrupt increase in mean aortic pressure. Under these conditions right ventricular cardiac output may transiently exceed left ventricular cardiac output causing a shift in blood volume from the systemic 148 vasculature to the pulmonary vasculature. Further evidence for this is the rise in mean pulmonary artery and left atrial pressures. It is possible that decreased left ventricular and pulmonary vascular compli- ance also contributed to increased mean pulmonary artery and left atrial pressures. Decreased vascular compliance may have also contributed to the increased pulmonary artery pulse pressure; though this was in part due to increased right ventricular stroke volume. Stellate ganglion stimulation (129) and brain ischemia (96) have been reported to decrease pulmonary vascular compliance resulting in an increased pulmonary artery pulse pressure without a change in PVRT. Differences Between Intact Nerve and Cut Nerve Groups Differences in response to brain hypoxia and/or hypercapnia were obtained between the intact nerve and cut nerve groups. In the intact nerve group no changes were observed in dP/dt, total peripheral resist- ance, mean aortic pressure or pulmonary artery pulse pressure but in the cut nerve group these parameters increased. The vagus and carotid sinus nerves (i.e., intact nerve group) apparently buffered out the primary response observed in the cut nerve group. The primary response to brain hypoxia and/or hypercapnia is an increased sympathetic discharge to the cardiovascular system that results in increased ventricular contractility (67), heart rate and total peripheral resistance (68). The vagus and carotid sinus nerves increase their afferent nerve activity when the baroreceptors are stimulated by increased mean aortic pressure. Stimula- tion of cardiac stretch receptors also increases afferent vagal activity. Because these afferent impulses are inhibitory to medullary sympathetic 149 outflow, they are the likely explanation for the lack of responses in the intact nerve group compared to the cut nerve groups. Similar buffering effects on the cardiovascular system have been reported previously (62,67,184). Perfusion of the brain with hypercapnic or hypoxic-hypercapnic blood decreased heart rate when the vagus nerves were intact. The likely explanation for this is that vagal tone to the heart increased during CO2 stimulation of vagal nuclei in the medulla (56). This is further supported by the observation that after the vagus nerves were cut heart rate did not change during hypercapnia or hypoxia-hypercapnia. The lower head perfusion pressures observed in the cut nerve group, despite higher blood flows to the head, probably occurred because in the dog cervical vagotomy eliminates sympathetic innervation to blood vessels supplying the head. This lack of sympathetic tone was also probably responsible for the smaller change in head perfusion pressure from con- trol values during hypoxia, hypercapnia, and hypoxia-hypercapnia. Brain Hypoxia and Pulmonary Edema The hemodynamic results in the cut nerve group are consistent with those found in neurogenic pulmonary edema (albeit of lesser magnitude) which develops following a variety of brain disorders in the absence of primary cardiac and pulmonary disease (232). In this disorder massive sympathetic outflow occurs; a situation more analogous to the cut nerve than the intact nerve group of this investigation since in the latter sympathetic outflow was inhibited by the peripheral baroreceptors as discussed earlier. 150 The key factor in the development of neurogenic pulmonary edema appears to be a large increase in left ventricular alterload resulting in increased left atrial and pulmonary vascular pressures and volume (36-38,64,74,159,204). The increase in left ventricular afterload is caused by intense peripheral vasoconstriction and is mediated by alpha adrenergic receptors (204,232). Indeed, neurogenic pulmonary edema caused by chemically induced convulsions (16), chemical hypothalamic stimulation (248), blunt head trauma (15,36) and increased intracranial pressure (151) can be prevented by phenoxybenzamine (an alpha blocker). This has important clinical implications because in order to prevent a shift in blood volume from the systemic to the pulmonary circulation (and high microvascular pressure), alpha adrenergic blockade must be directed primarily with systemic arterial pressure in mind. It should also be kept in mind that there are two distinct phases to neurogenic pulmonary edema (232). First, a transient systemic hypertension and second, a return to normal vascular pressures accompanied by increased pulmonary microvascular permeability. The latter probably is induced by the transient capillary hypertension (196). Obviously for alpha blockade therapy to be effective it must be given prior to or during systemic hypertension to prevent a large shift in blood volume to the pulmonary circulation. The increase in left atrial, mean pulmonary artery and aortic pres- sures during 5 min hypoxia (cut nerve group) suggest increased pulmonary blood volume. Indeed, central blood volume increased and although some of this may reflect increased left heart volume, it is likely that pulmonary blood volume increased as well and this certainly would 151 contribute to the increased pulmonary vascular pressures observed. In contrast, these increases in vascular pressures and central blood volume were not detected during 2 hours hypoxia. Although this might sug- gest that the effects of brain hypoxia on pulmonary hemodynamics is a transient phenomenon, I believe a more likely explanation is a progressive deterioration of my preparation as evidenced by falling total peripheral resistance and mean aortic pressure. The lack of any change in lung extravascular thermal volume and postmortem lung water content is not surprising in view of the small hemodynamic changes elicited during brain hypoxia. The increase in plasma filtrate (i.e., lymph) caused by l to 2 mmHg increase in left atrial pressure would easily be removed by the lymphatics and therefore prevent any accumulation of extravascular lung water. In fact, Guyton and Lindsey (105) found that left atrial pressure had to increase to approximately 24 mmHg before lymphatics failed to drain the excess filtered fluid and thus prevent pulmonary edema. Although the results of this investigation could be interpreted to mean that brain hypoxia does not cause neurogenic pulmonary edema, I believe such a conclusion is inappropriate. In a clinical situation acute systemic hypoxia would cause direct and reflex (through stimulation of the peripheral chemo- receptors) stimulation of the brain as well as release of catecholamines from the adrenal glands. It is conceivable that these events, occurring in an unanesthetized and previously healthy individual, could lead to a dramatic increase in mean aortic pressure and secondarily cause large increases in left atrial and pulmonary vascular pressures. The latter being mediated by a massive shift in blood volume to the pulmonary 152 circulation causing capillary hypertension, and hence, neurogenic pulmonary edema. Indeed, a large increase in pulmonary blood volume is the likely explanation for development of pulmonary edema and alveolar hemorrhage during such clinical disorders as seizures (21,124), increased cerebro- spinal fluid pressure (73), head trauma (79,212), cerebral hemorrhage (240), apoplexy and brain tumors (33). With regard to the effects of anesthesia on the development of neurogenic pulmonary edema, Bean and Beckman (15) found that unanesthe- tized rats subjected to mechanical head injury developed severe pulmonary edema and alveolar hemorrhage. In contrast, rats pretreated with anesthetic agents (similar to those used in this investigation) prior to head trauma developed little or no pulmonary pathology presumably because excitability of the central nervous system was depressed. Thus it is not unreasonable to suggest that in the present investigation anesthesia may have attenuated the pulmonary hemodynamic effects of brain hypoxia. Adult Respiratory Distress Syndrome Recently Moss and Stein (167) and Brown (30) proposed a centri- neurogenic etiology for the adult respiratory distress syndrome ("shock lung"). In their preparation venous blood (P02 35 torr) is pumped from the right atrium to a common carotid artery for two hours at a perfusion pressure that is 20 mmHg greater than mean aortic pressure thus ensuring hemodynamic isolation of the Circle of Willis. Moss and Stein (167) reported that the pulmonary lesions observed at postmortem are similar to those observed in the adult respiratory distress syndrome (i.e., conges- tion, edema, atelectasis and hemorrhage). They then repeated these 153 experiments following chronic unilateral pulmonary denervation. Based upon gross and histologic examination they found that the innervated lung developed lesions similar to the adult respiratory distress syndrome whereas the denervated contralateral lung was normal. Based on their work and a large body of circumstantial evidence (95,137,169,202,206,210- 212,215,226-228,236,249) they hypothesized that brain hypoxia impairs oxidative metabolism in the hypothalamus which increases sympathetic out- flow to the pulmonary postcapillary vessels leading to increased pulmonary vascular resistance and microvascular hydrostatic pressure, congestion, surfactant inactivation, atelectasis, edema, and hemorrhage. My findings support an increased microvascular hydrostatic pressure (volume mediated) during 5 min brain hypoxia but are not tenable with the hypothesis that this increase is secondary to increased postcapillary resistance because PVRT, PVRP, and PVRv did not increase in either the intact nerve or cut nerve groups. I propose that brain hypoxia elevates microvascular hydro- static pressure secondary to increased pulmonary blood volume and is therefore a form of neurogenic pulmonary edema. Comparison of data between investigators is admittedly difficult when considerable differences exist between experimental preparations. However, I believe the pulmonary lesions reported by Moss and Stein (167) could also be explained by increased pulmonary blood volume. One of the intriguing aspects of Moss' work is that unilateral pulmonary denervation during brain hypoxia is protective. However, this should not be construed as prima-facie evidence that pulmonary nerves are responsible for the lesions observed in innervated lungs. Since the denervation procedure involved transection of all ipsilateral hilar lung tissue it is possible 154 that post-surgical fibrous tissue reduced the diameter of the reanasto- mosed pulmonary artery and veins. The former would increase upstream vascular resistance and preferentially direct more flow to the "innervated" lung. Furthermore, increased downstream (i.e., venous) resistance in the denervated lung would preferentially cause blood to "dam-up" in the "innervated" lung thus increasing microvascular hydro- static pressure and possibly causing the pulmonary pathology reported by Moss and Stein (167). Although the findings of this investigation during 5 min brain hypoxia are significant and the preparation was obviously viable, the findings during 2 hours hypoxia raise some questions about preparation viability inasmuch as it was difficult to keep the animals alive. There appeared to be a progressive deterioration of the cardiovascular system as evidenced by decreasing total peripheral resistance and mean aortic pressure despite intravenous fluids (to correct for fluid loss). This problem, however, is not unique to my preparation. I have had personal communication with 3 independent physiologists (Drs. M. Maron, B. Petersen, J. Reeves) who have used the experimental preparation described by Moss et a1. (167,168). In all cases they were not able to maintain survival long enough to duplicate the studies of Moss and Stein (167). Crockard et a1. (46) perfused the brain of monkeys using the Moss preparation and all animals died of brain edema approximately 95 minutes after the onset of hypoxemic perfusion. An interesting feature of their study was that the monkeys allowed to breathe spontaneously developed incidental pul- monary congestion and edema but those with positive pressure ventilation did not. This raises the possibility that much of the atelectasis and 155 systemic hypoxemia described by Moss and Stein (167) was exaggerated by the effects of spontaneous breathing. II. Carotid Body Hypoxia and Hypercapnia The second objective of this research was to evaluate the effects of carotid body hypoxia, hypercapnia, and hypoxia-hypercapnia on pul- monary hemodynamics. The data demonstrate that hypoxic and hypoxic- hypercapnic blood perfusing the carotid bodies can reflexly cause a small increase in PVRT in vagotomized dogs. Increased PVRT observed with hypoxia or hypoxia-hypercapnia was pre-large vein in origin [i.e., (PPA" PPAw)/0 = PVRP] and although precapillary vessels are the most likely sites of increased resistance, venules and small pulmonary veins cannot be ruled out because PVRP includes the resistance of all vessels upstream from the first post-capillary collateral venous vessel. The increases in resistance must have resulted from active vasoconstric- tion because they cannot be explained by the passive effects of vascular transmural pressure. During hypoxia aVerage prevenous vascular trans- mural pressure [i.e., (PPA'+ PPAW/z] was unchanged from control (8.3 to 8.4 mmHg) despite decreased cardiac output and during hypoxia-hypercapnia it increased above control (8.5 to 9.2 mmHg) without a change in cardiac output. The passive effects of increased transmural pressure would act to increase vessel diameter and decrease PVRT (82) and this is the likely explanation for a lesser increase in PVRT during hypoxia-hypercapnia than during hypoxia. Increased PVRT during carotid body hypoxia and hypoxia-hypercapnia agree with the findings of other investigators (10,53,246). However, 156 Daly and Daly (53) reported that increased PVRT during carotid body hypoxia in a non-vagotomized preparation only occurred consistently in the absence of bronchial flow. They concluded the primary effect of carotid body hypoxia on the pulmonary circulation was to increase PVRT but that this effect was masked by hemodynamic events associated with the bronchial circulation. Despite an intact bronchial circulation in my preparation PVRT increased during hypoxia and hypoxia-hypercapnia but only after bilateral cervical vagotomy. Thus, it appears that a vagotom- ized preparation (which eliminates vagal afferent inhibition of central sympathetic outflow) might be a more important factor in eliciting increased PVRT during carotid body hypoxia and hypoxia-hypercapnia than is exclusion of the bronchial circulation. I consider the presence of an intact bronchial circulation in my experiments to have negligible effects on the data or its interpretation for several reasons. First, Bruner and Schmidt (32) reported that approxi- mately 1% of the cardiac output drains into the pulmonary veins from the bronchial circulation. This small amount of flow negligibly affects the pulmonary vascular resistance values. 'Second, if substantial transfer of blood from the bronchial arteries to the pulmonary veins occurred I would have expected left atrial and/or pulmonary artery wedge pressures to increase and these were unchanged. Third, results of more recent investigations involving changes in pulmonary vascular resistance (132) or vascular compliance (129) during electrical stimulation of stellate ganglion were not significantly different whether the bronchial circula- tion was intact or not. Thus, I believe an intact bronchial circulation in this study was of little or no consequence. 157 It is well known that changes in lung volume and active movement of respiratory muscles may passively influence pulmonary vascular resistance (53,82). In my preparation these factors were minimized by: l) ventilating at constant volume, rate and end-expiratory pressure; 2) paralyzing the dogs respiratory muscles; 3) opening the thorax. It is unlikely local alveolar hypoxia caused the observed increases in pulmonary vascular resistance since PaO2 was unchanged from control values during carotid body stimulation (both > 100 torr). Stern et al. (225) reported nicotine stimulation of aortic bodies reflexly increased PVRT approximately 31% and with the carotid sinus nerves cut approximately 49%. Stern and Braun (224) reported nicotine stimulation of the carotid bodies caused no change in PVRT either before or after vagotomy and concluded that carotid body stimulation does not reflexly increase PVRT. This conclusion is at variance with those of others (10,53,246) and the results of the present study. While it is difficult to compare the responses to physiologic and pharmacologic stimulation it should be kept in mind that the pulmonary vascular response to severe hypoxic or hypoxic-hypercapnic carotid body stimulation was not of large magnitude. It may be that stimulation of the aortic bodies is more effective in elevating PVRT than stimulation of the carotid bodies. Comroe (44) has emphasized this point with regard to the systemic circulation in the dog. Since sympathetic innervation to the pulmonary vasculature has been clearly demonstrated (117), with alpha receptors predominating (17), it is likely that the increased PVRT and PVRP observed before alpha blockade was due to activation of alpha adrenergic receptors either through stimulation of sympathetic nerves to 158 the pulmonary circulation or release of catecholamines from the adrenal glands (7). This conclusion is supported by the fact that following alpha blockade with phentolamine, carotid body hypoxia and hypoxia- hypercapnia failed to increase PVRT and PVRP. However, it must be recognized that baseline PVRT was greater in the alpha-block compared to the non alpha-block control groups and this could have masked any further vasoconstriction during carotid body hypoxia and hypoxia-hypercapnia. During hypoxic-hypercapnic stimulation of the carotid bodies an increased pulmonary artery pulse pressure was observed (without a signifi- cant change in stroke volume). This effect was also blocked by phentol- amine suggesting that carotid body stimulation is capable of reflexly decreasing compliance of pulmonary vessels, an effect that has been previously reported (230). Despite complete alpha blockade (as determined from blood pressure recordings) to an alpha agonist administered intravenously, an attenuated increase (P<:0.05) in total peripheral resistance still occurred during hypoxia-hypercapnia. This is consistent with the concept that alpha adrenergic antagonists are generally more effective in blocking responses to circulating alpha agonists than those resulting from adrenergic nerve activity (172). The decreased cardiac output during carotid body hypoxia, under conditions of controlled ventilation and intact vagus nerves, has also been reported by other investigators (55,69). This investigation demonstrates a similar decrease following vagotomy and suggests that the decrease in cardiac output during carotid body hypoxia is due to with- drawal of sympathetic tone to the heart rather than increased vagal 159 outflow. Although hypothesizing sympathetic withdrawal from the heart while the vasculature receives increased sympathetic tone is difficult to reconcile teleologically, this has been previously reported. In vagotomized dogs, Downing et al. (69) found a general sympathetic withdrawal from the heart and increased sympathetic tone to the vascula- ture during carotid body stimulation with hypoxic blood. Their evidence for this was increased total peripheral resistance, decreased atrial and ventricular contractility, and decreased heart rate; the latter effect was abolished by the ganglionic blocker hexamethonium. During hypercapnic and hypoxic-hypercapnic perfusion (before alpha blockade) of the carotid bodies, cardiac output was unchanged. Following alpha blockade cardiac output increased (P<:0.05) during hypoxia-hypercapnia and tended to increase during hypoxia (P<:0.10). These findings seem inconsistent with those reported during carotid body hypoxia before alpha blockade (i.e., cardiac output decreased), however, I believe the explanation lies in the fact that control mean aortic pressure during hypoxia was 110 mmHg (vagi intact) and 102 mmHg (vagi cut), during hypercapnia 79 mmHg, and during hypoxia-hypercapnia 74 mmHg. Following alpha blockade control mean aortic pressure was 50 mmHg during hypoxia and 54 mmHg during hypoxia-hypercapnia. It seems likely that baseline sympathetic outflow to the heart was greater in those groups with the lower carotid sinus perfusion pressures (set equal to mean aortic pressure). Downing et al. (69) reported that carotid body hypoxia in the presence of high carotid sinus pressure caused decreased ventricular contractility and increased vascular resistance but in the presence of low carotid sinus pressure both of these parameters increased. 160 Thus, it appears that the sympathetic activity to the heart during carotid body stimulation is somewhat dependent on the degree of afferent activity from the carotid sinus and perhaps aortic baroreceptors. During carotid body stimulation at high carotid sinus pressures, sympathetic withdrawal from the heart might occur but at low carotid sinus pressures ”cardiac sympathetic stimulation may occur. It seems likely that the interaction of these reflexes would occur within the central nervous system. Central interaction of baroreceptor and chemoreceptor reflexes has been previously suggested (118). Adult Respiratory Distress Syndrome In the experimental preparation of Moss and Stein (167), brain hypoxia is induced by pumping venous blood from the right atrium to one common carotid artery. This means that the ipsilateral carotid bodies (and probably the contralateral as well) are also being perfused with hypoxemic blood. The results of this investigation suggest that carotid body stimulation alone is not likely to produce the adult respiratory distress syndrome or neurogenic pulmonary edema because left atrial pres- sure was unaffected and only small increases in PVRP and pulmonary vascular pressures were observed. Furthermore, although I did not deter- mine water content in these lungs, no gross evidence of edema was present. It seems likely, however, that carotid body hypoxia would contribute to the pulmonary lesions and hypoxemia obtained by Moss and Stein (167) for three reasons. First, carotid body hypoxia does reflexly increase PVRP and this would increase fluid filtration from alveolar and/or extra- alveolar vessels. Second, peripheral chemoreceptor stimulation with hypoxemic blood (PaO2 < 40 torr) may attenuate the local alveolar hypoxia 161 response and cause increased right to left shunting of blood, thus worsen- ing systemic hypoxemia (142). Third, a substantial increase in left ventricular afterload does occur during carotid body hypoxia and/or hypercapnia. The latter effect appears to be the precipitating factor that causes neurogenic pUlmonary edema as discussed earlier. Systemic hypoxemia would obviously stimulate the brain both directly and reflexly through primary stimulation of the aortic and carotid bodies. In this regard aortic body stimulation constricts pulmonary arteries and veins (224) and, compared to carotid body stimulation, is more effective in elevating mean aortic pressure (44). Thus the combined effects of acute hypoxic stimulation of the brain and peripheral chemoreceptors could result in pulmonary edema and alveolar hemorrhage similar to that observed in the adult respiratory distress syndrome. III. Histamine and Adrenergic Receptors The third objective of this research was to evaluate the effect of histamine on lung water and hemodynamics before and after adrenergic blockade and attempt to relate these findings to the adult respiratory distress syndrome as well as vascular permeability in general. Ninety minutes of histamine infusion (10 pg/kg/min) failed to increase lung extravascular thermal volume and postmortem extravascular lung water/ extravascular dry weight ratio. These results are consistent with previous reports emphasizing the lack of ability of histamine to increase lung water (70,103). During simultaneous histamine and propranolol or phentolamine infusion, lung extravascular thermal volume was again unchanged as was the extravascular lung water/extravascular dry weight 162 ratio (compared to time control dogs receiving the same euthanasia solu- tions). Thus, it appears that the inability of histamine to markedly increase lung water in the dog is not due to enhanced beta (or alpha) receptor stimulation on the lung microvascular membrane. It should be emphasized that in the canine forelimb intravenous histamine with beta blockade causes massive edema (100). Indeed, in the systemic circulation beta agonists antagonize the action of histamine on both the large (100, 152,153) and small (194a) pore systems and prevent edema formation. Thus, if similar mechanisms were operative in the lung, I believe edema should have been detectable under the conditions of the present investi- gation. Therefore these results suggest a fundamental difference in response of the lung and systemic microvasculature to administration of histamine in the presence of beta blockade. These results do not rule out the possibility that vascular perme- ability increased since I did not measure pulmonary lymph flow and protein concentration; however, if increased microvascular permeability had occurred it would have increased volume of distribution for Na+. This would have increased central blood volume values and decreased lung extravascular thermal volume values (relative to control values), neither of which occurred with histamine alone or in the presence of alpha or beta blockade. Furthermore, lung extravascular thermal volume over- estimatedpulmonary extravascular tissue weight a similar amount (approxi- mately 17%) in dogs receiving and not receiving histamine. Increased volume distribution for Na+ would have Caused a lesser overestimation of pulmonary extravascular tissue weight (or possibly an underestimation) in dogs receiving histamine. In addition, the fact that the slope for the 163 regression line and correlation coefficient between pulmonary extra- vascular tissue weight and lung extravascular thermal volume were not significantly different between time control dogs and dogs receiving histamine further suggests that histamine did not markedly alter the volume of distribution for Na+ and therefore significant increases in microvascular permeability seem unlikely. Investigations (89,188) using colloidal carbon as a tracer molecule suggest that histamine does not affect pulmonary microvascular permeabil- ity but does cause increased permeability in the bronchial microcircula- tion (especially in the venules) that is rapid and brief in duration. This observation is consistent with histamine's effect on forelimb micro- vasculature (102,108,152) and is not surprising since the bronchial circulation is part of the systemic circulation (188). Ultrastructurally Pietra et al. (188) observed that bronchial venules differed from corre- sponding vessels in the pulmonary microcirculation. Bronchial venules had a thicker endothelium, a greater number of pericytes, and a greater supply of cytoplasmic filaments within endothelial cells. Cytoplasmic filaments in endothelial cells of pulmonary veins were either absent or in very small amounts and they (188) concluded that the bronchial venular endothelium has a much greater potential for contraction than does the pulmonary venular endothelium. Therefore, it is possible that there was a transient increase in lung extravascular thermal volume because of peribronchial edema and that this effect disappeared prior to my first measurement (i.e., 30 minutes). However, since flow through the bronchial circulation is. approximately 1% of the pulmonary circulation (32) any increase in total lung water would likely be of small magnitude 164 and in all likelihood undetectable by double-indicator dilution tech- niques. Furthermore, since there was no difference in postmortem lung water content between beta or alpha blocked and unblocked dogs any transient effect would have little physiological significance. Hemodynamics Histamine The large decreases in mean aortic pressure, total peripheral resistance, cardiac index (at 30 min) and stroke volume caused by hist- amine are consistent with previously reported actions of histamine on the systemic circulation (107,193). Decreased mean aortic pressure was caused by both a fall in total peripheral resistance and cardiac index. The fall in total peripheral resistance is primarily due to arteriolar dilatation (107). Because histamine may cause venous constriction (107), pooling of blood occurs peripherally leading to decreased central venous pressure, venous return and cardiac index (193). Stroke volume decreases presumably because of decreased end-diastolic volume secondary to decreased filling pressure and time. The latter occurs because of the baroreceptor-induced tachycardia secondary to systemic hypotension as well as release of catecholamines from the adrenal glands (100,153, 217). Although some of the increase in PVRP caused by histamine can be attributed to passive mechanisms (due to decreased vascular transmural pressure), it is likely that some of the increase was also due to active vasoconstriction. Histamine is a potent postcapillary constrictor in the dog lung (28,92,93). Gilbert et al. (92) reported the predominant site of this heightened resistance to be distal to the first post-capillary 165 collateral venous vessel (i.e., pulmonary artery wedge position) but proximal to the large veins. Increased resistance at this site would tend to increase pulmonary artery wedge pressure; however, in the present investigation pulmonary artery wedge pressure decreased. It is unlikely increased venous compliance caused pulmonary artery wedge pres- sure to decrease since, if anything, compliance would have decreased because of increased sympathetic outflow secondary to systemic hypoten- sion. Thus, in order to explain the fall in pulmonary artery wedge pressure a proportionately greater decrease in venous volume than an increased in venous resistance must have occurred. Indeed, central blood volume (which in this investigation represents the volume of blood between the right ventricle and proximal aorta) decreased and although it is likely that a portion of this reduction was secondary to decreased left heart volume, the change was too large not to involve a substantial reduction in pulmonary blood volume. Furthermore, a decrease in cardiac index (at 30 min) was not associated with an increase in conductivity mean transit time, which suggests a concomitant decrease in vascular cross-sectional area. Previous investigations using excised dog (28) and cat (61) lungs reported that histamine decreased post-capillary blood volume but was without effect on pre-capillary blood volume. Although decreased venous volume seems the most likely site for reduced pulmonary blood volume, the present data do not exclude a change in volume in the capillaries and pre-capillaries. The decrease in central blood volume caused by histamine probably resulted from decreased left ventricular afterload (as evidenced by an abrupt decrease in mean aortic pressure) and decreased venous return 166 secondary to peripheral pooling of blood. Under these conditions left ventricular stroke volume may transiently exceed right ventricular stroke volume causing a shift in blood from the pulmonary vasculature to the systemic vasculature. When both ventricles regain equal stroke volumes central blood volume would remain in a steady state. Indeed, central blood volume did not change between 30 and 90 minutes of histamine. Propranolol and Histamine Infusion of the beta blocker, propranolol, decreased cardiac index and stroke volume, presumably because of decreased ventricular contrac- tility. Mean aortic pressure was maintained by an increase in total peripheral resistance. Following beta blockade, PVRP did not increase as might be expected when a vasodilating effect is removed (13,17,18). However, in the present investigation vascular transmural pressure increased, as evidenced by increased mean pulmonary artery and pulmonary artery wedge pressures. The passive effects of increased transmural pressure would act to increase vessel diameter and decrease PVRP (22,82). Indeed, the fact that PVRP did not decrease in the presence of rather large increases in transmural pressure is evidence that pulmonary vaso- constriction occurred following beta blockade. Since central blood volume increased it is likely that the increase in pulmonary artery wedge pressure was volume mediated, albeit venous constriction and decreased venous compliance cannot be excluded as contributing factors. The negative inotropic effect of propranolol decreased stroke volume. The decrease in left ventricular stroke volume presumably exceeded the decrease in right ventricular stroke volume (at least transientlY). thus shifting blood from the systemic vasculature to 167 the pulmonary vasculature and left heart (i.e., central blood volume). Increased central blood volume probably increased vascular cross- sectional area. This effect, as well as decreased cardiac output, caused a 62% increase in conductivity mean transit time. The hemodynamic effects of histamine following beta blockade were qualitatively similar to those observed for histamine infusion except for heart rate and conductivity mean transit time. In the presence of beta blockade and histamine, heart rate did not change and conductivity mean transit time increased. This probably occurred because beta receptor blockade also eliminates any sympathetically mediated chronotropic response that would result secondary to systemic hypotension. Increased conductivity mean transit time was most likely due to a very low cardiac output. Cardiac index was significantly lower than during histamine infusion resulting also in a significantly lower mean aortic pressure. During histamine infusion cardiac index (60-90 min) returned to control values but during propranolol-histamine infusion this effect was abolished. Presumably catecholamines released from the adrenal gland (217) were responsible for maintaining not only a higher cardiac index but also increasing cardiac index (at 60 and 90 min) towards control values. Histamine with beta blockade significantly potentiated (P<:0.05) the increase in PVRP observed during histamine alone. Bergofsky (18) reported beta blockade with histamine more than doubled the pulmonary vasoconstrictor response to histamine, possibly by unmasking previously antagonized alpha receptors. 168 Phentolamine and Histamine Alpha blockade with phentolamine decreased total peripheral resist- ance which decreased mean aortic pressure. Stroke volume decreased but cardiac index was maintained by an increased heart rate. Following alpha blockade PVRP did not decrease as is frequently reported (13,17,18). However, as with beta blockade, the passive effects of vascular transmural pressure apparently predominated. In contrast to increased vascular transmural pressure with beta blockade, alpha blockade decreased vascular transmural pressure. This should have increased PVRP and since PVRP did not change it suggests that alpha blockade caused pulmonary vasodilation. Pulmonary artery pulse pressure decreased and although decreased stroke volume was a contributing factor, it is possible that blockade of alpha receptors decreased vascular compliance. Following alpha blockade mean aortic pressure, pulmonary artery wedge pressure, and central blood volume decreased in a manner similar to infusion of histamine only. The mechanism discussed for decreased central blood volume with histamine is also the likely explanation for decreased central blood volume with phentolamine. Decreased pulmonary vascular cross-sectional area probably occurred during alpha blockade because conductivity mean transit time decreased without a change in cardiac output. Infusion of histamine following alpha blockade caused no further change in any hemodynamic parameter measured except for increased PVRP, decreased mean aortic pressure and total peripheral resistance, and a gradual return of mean pulmonary artery pressure to control values. These results are consistent with the ability of histamine to cause 169 pulmonary vasoconstriction and peripheral vasodilation independent of alpha adrenergic receptors. However, this does not mean that alpha receptors are not utilized by histamine. Attenuation of histamine- induced increase in PVRP by alpha blockade has been reported (18). Lung Mechanics Histamine,propranolol with histamine, and phentolamine with hist- amine increased airway opening pressure approximately 23%, 46%, and 21%, respectively. Since tidal volume and respiratory rate were held constant, this represents a decrease in dynamic compliance and can best be explained by airway constriction (241). These findings are consistent with previous reports that emphasize the ability of histamine to cause constriction of alveolar ducts and airways (42,63). Histamine with beta blockade significantly potentiated the decrease in dynamic compliance compared to the decrease observed with histamine or alpha blockade with histamine. Histamine induced release of catecholamines from the adrenal gland (217) presumably accounts for the lesser increase in airway open- ing pressure during histamine or histamine with alpha blockade. This is consistent with previous reports (42,63) that catecholamines released from the adrenal gland primarily affect beta adrenergic receptors (in airways) to antagonize the constricting effect of histamine. Adult Respiratory Distress Syndrome Histamine has been implicated as a possible mediator of the adult respiratory distress syndrome (19,247). In the present investigation, large doses of histamine failed to increase lung extravascular thermal volume and postmortem lung water and this response was unaffected by 170 alpha or beta blockade. Thus, it appears unlikely that histamine is an important contributing factor to the large increases in microvascular permeability and lung water that occur in the adult respiratory distress syndrome. However, histamine is a potent postcapillary constrictor (92, 93) in the lung and this would contribute to increased microvascular hydrostatic pressure, which in the presence of increased vascular perme- ability, could contribute substantially to development of pulmonary edema (84). On the other hand, the results of this investigation show that histamine decreases pulmonary artery wedge pressure by decreasing pul- monary blood volume. The effect of this would be to attenuate any increase in microvascular hydrostatic pressure brought about by venous constriction. Thus, histamine has considerable hemodynamic effects on the pulmonary circulation that include changes in vascular resistance, volume, blood flow, and probably vascular compliance. If the net effect of these changes is to increase microvascular hydrostatic pressure then histamine would obviously contribute to edema formation and in this way could play a role in mediating the adult respiratory distress syndrome. In dogs of this investigation, histamine caused impairment of gas exchange as evidenced by decreased P302 and increased PaCO2 (at 90 min). This effect can best be explained by peripheral airway constriction resulting in greater ventilation-perfusion inequalities. Therefore, if endogenous release of histamine occurs in the adult respiratory distress syndrome (247), it would likely worsen the clinical malady by potentiating the hypoxemia and decreasing dynamic compliance. Thus the deleterious effects of histamine on lung mechanics and gas exchange may be a more important factor in mediating the adult respiratory distress syndrome than are histamines' effects on pulmonary hemodynamics. SUMMARY AND CONCLUSIONS Following ligation of collateral vessels (supplying blood to the brain) brain hypoxia, hypercapnia, and hypoxia-hypercapnia were induced by pumping arterial autologous blood through an extracorporeal lung to the external carotid arteries for 5 min or 2 hours. When the vagus and carotid sinus nerves were intact, hypoxia and hypoxia-hypercapnia caused no change in any pulmonary or systemic parameter measured or calculated. Hypercapnia increased mean aortic pressure and stroke volume and decreased heart rate. Following vagotomy and carotid sinus nerve section (which eliminated vagal and carotid sinus nerve afferent inhibition of central sympathetic outflow), 5 min hypoxia and hypoxia- hypercapnia increased cardiac output, dP/dt, total peripheral resistance, mean pulmonary artery, left atrial, pulmonary artery pulse, and mean aortic pressures. However, these effects were not maintained during 2 hours hypoxia. Central blood volume increased during 5 min hypoxia but pulmonary vascular resistance and lung extravascular thermal volume were unchanged during 5 min and 2 hours of brain hypoxia. Postmortem extravascular lung water to extravascular dry weight ratio was not altered following 5 min or 2 hours of brain hypoxia. I conclude that 5 min of brain hypoxia and/or hypercapnia does not increase pulmonary vascular resistance but may increase microvascular pressure secondary to increased pulmonary blood volume and in this way could cause pulmonary edema and alveolar hemorrhage similar to that observed in the adult 171 172 respiratory distress syndrome. Failure of these effects to be maintained for 2 hours may reflect deterioration of the preparation. The carotid bodies were perfused bilaterally with hypoxic and/or hypercapnic blood by pumping arterial autologous blood through an extra- corporeal lung to the common carotid arteries. Before vagotomy, carotid body hypoxia decreased cardiac output and increased total peripheral resistance but did not alter any pulmonary parameter measured or calcu- lated. After vagotomy, carotid body hypoxia decreased cardiac output while pulmonary vascular resistance, mean aortic pressure and total peripheral resistance increased. During hypoxia-hypercapnia mean pul- monary artery, pulmonary artery pulse, and mean aortic pressures increased as did pulmonary vascular and total peripheral resistances. Hypercapnia increased mean aortic pressure and total peripheral resistance but did not change pulmonary vascular resistance. Following alpha blockade with phentolamine pulmonary vascular resistance failed to increase during carotid body hypoxia or hypoxia-hypercapnia. I conclude that carotid body hypoxia and hypoxia-hypercapnia reflexly increase pulmonary vascular resistance and left ventricular afterload, that these effects are mediated by alpha adrenergic receptors, and that carotid body stimulation could potentiate the effects of brain hypoxia on pul- monary hemodynamics and contribute to increased lung water. The effects of 90 min intravenous histamine with and without alpha (phentolamine) or beta (propranolol) receptor blockade on lung water and hemodynamics were also studied. Histamine with and without alpha blockade decreased central blood volume, pulmonary artery wedge, and mean aortic pressures; pulmonary vascular resistance increased while lung 173 extravascular thermal volume was unchanged. Histamine with propranolol decreased mean aortic pressure, cardiac index, and central blood volume. Pulmonary vascular resistance increased and lung extravascular thermal volume was unchanged. In all experiments postmortem extravascular lung water to extravascular dry weight ratio was unchanged from control values. I conclude that histamine has little or no effect on lung micro- vascular permeability, does not increase lung water, and that these responses are not modified by reflex beta or alpha receptor stimulation on the microvascular membrane. Thus, histamine cannot be considered a likely mediator of the massive increase in lung water and permeability that occur in the adult respiratory distress syndrome. RECOMMENDATIONS A major limitation of these studies is that the dog requires extensive surgical intervention to eliminate collateral blood flow to the brain, thus making the preparation a difficult one to study and maintain. A further complicating factor is that these experiments must be done in open-chested animals. In contrast, surgical manipulation required to induce local brain hypoxia in sheep should be minimal because practically the entire brain is supplied by blood through the external carotid arteries. The internal carotid artery is absent in the adult and the vertebral artery supplies only the cervical spinal cord to the level of the obex (11a). In addition, utilizing sheep would make a thoracotomy unnecessary since vessels originating off the subclavian artery do not contribute to cerebral blood flow. Even more attractive is the possibility that local brain hypoxia could be induced in the conscious animal. Another advantage of using sheep is that the efferent duct of the caudal mediastinal node can be easily cannulated (223a) thus allowing for collection of lung lymph during brain hypoxia. Finally, sheep pulmonary vessels are considered to be more responsive (compared to the dog) to nervous and humoral stimuli (111). Therefore, I believe utilizing sheep as a model to study the effects of brain hypoxia on pulmonary hemodynamics and lung water would be more appropriate in future investigations. If a viable and stable brain hypoxia model could 174 175 be developed, it would be extremely useful in further elucidating central control mechanisms on pulmonary hemodynamics as well as evalua- tion of various therapeutic interventions. APPENDIX 176 . mocwum www.» Sm PDUmm> xsmcoe—aa cmm> manor .>m>m “mocmumwmmg Lapaumw> agmcospaa cwm> wmsmpuwgn .am>m “mucmumpmog .23.: 22953 .32 CE .233: :2: E .3 $23: as: is. $053 q .3: . a .233;$212355”; .13 :3: .0 sagas... .1 3:3,; 5 as... .a .Pmmmm> maocm> Pmcmpmppoo mng—wamoumoa umgww any do muwm mgu um ogammmca Emmgumczou m mw mcammmca mono: Acmugm zgmcoepaq mash .mcwm> wmgmfi mg» «comma was was zgmpp_amo mg» cuppa umuowawv mw Emcmmwu mvgu cw sows; Pmmmm> Fmgmumrpoo pmgww use can gm“ gmumgumu ms“ cmmzumn mcauuo cowumcnP—wacm mgzmmwga .gmpmgumo cmmum: a »a mo .umnaopm m? xgmugm zgmcoepza PFmEm m sonoggu 3on con: .mcowumuop Asawcpm ammF ..m.mv Emwgpmczov can Axgmpgm FFmEm ..m.mv Emmcumqa ou :owuwmoa mmumz acwpgm xgmcospaa mg» we awzmcowumpmc mcwzozm Emcmupo .p< ocamwu M, _.. . 177 < xwucmaa< o > o . o . Sm Izfim u m>n_ 34.515 u m>a Saufim u m>a 23.1.2 5.qu j 25> moto=aoo >524 __oEm ALOCO...F3Q Came) $31.5. a) -. -muUCflHmnrwwL LOFDUWaw) —. Esra. .GUCDUMrme LGPJUMQ) ELMCOEPDQ Crmw> WWLMPIULQ 00570 mvaanPMrmecp I‘ll”|.l|l \l. LIST OF REFERENCES 10. 11. LIST OF REFERENCES . Agostoni, E. and J. Piiper. Capillary pressure and distribution of vascular resistance in isolated lung. Am. J. Physiol. 202: 1033-1036, 1962. . Albert, R. K., S. Lakshminarayan, T. W. Huang and J. Butler. Fluid leaks from extra-alveolar vessels in living dog lungs. J. Appl. Physiol. 44: 759-762, 1978. . Albert, R. K., S. Lakshminarayan, W. Kirk and J. Butler. Lung inflation can cause pulmonary edema in zone 1 of in situ dog lungs. J. Appl. Physiol. 49: 815-819, 1980. . Anderson, F. L. and A. M. Brown. Pulmonary vasoconstriction elicited by stimulation of the hypothalamic integrative area for the defense reaction. Circ. Res. 12: 747-756, 1967. . Anderson, W. P., L. J. Dunegan, D. C. Knight, G. F. Fitzpatrick, N. E. O'Connor and A. P. Morgan. Rapid estimation of pulmonary extravascular water with an instream catheter. J. Appl. Physiol. 39: 843-845, 1975. . Andreyev, L. A. Functional changes in the brain of the dog after reduction of cerebral blood supply. Arch. Neurol. Psychiat. 34: 481-507, 1935. . Anichkov, S. V., E. I. Malyghina, A. N. Poskalenko and V. E. Ryzhenkov. Reflexes from carotid bodies upon the adrenals. Arch. Int. Pharmacodyn. 79: 156-165, 1960. . Arfors, K. E., P. Malmberg and K. Pavek. Conservation of thermal indicator in lung circulation. Cardiovasc. Res. 5: 530-534, 1971. . Aviado, D. M. The pharmacology of the pulmonary circulation. Pharmacol. Rev. 12: 159-239, 1960. Aviado, D. M., Jr., J. S. Ling and C. F. Schmidt. Effects of anoxia on pulmonary circulation: reflex pulmonary vasoconstriction. Am. J. Physiol. 189: 253-262, 1957. Bachofen, M., H. Bachofen and E. R. Weibel. Lung edema in the adult respiratory distress syndrome. In: Pulmonary Edema. Edited by A. P. Fishman and E. M. Renkin, Bethesda,TAmerican Physiological Society, 1979, p. 241-252. 178 11a. 12. 13. 14. 15. 16. 17. 18. 19. 20. 21. 22. 23. 179 Baldwin, B. A. and F. R. Bell. The anatomy of the cerebral circula- tion of the sheep and ox. The dynamic distribution of the blood supplied by the carotid and vertebral arteries to cranial regions. J. Anat. (London) 97: 203-215, 1963. Barer, G. R. The physiology of the pulmonary circulation and methods of study. Pharmac. Ther. 2: 247-273, 1976. Barer, G. R. and J. R. McCurrie. Pulmonary vasomotor responses in the cat; the effects and interrelationships of drugs, hypoxia and hypercapnia. Q. J. Exp. Physiol. 54: 156-172, 1969. Barer, G. R., P. Howard and J. W. Shaw. Stimulus response curves for the pulmonary vascular bed to hypoxia and hypercapnia. J. Physiol. (London) 211: 139-155, 1970. Bean, J. W. and D. L. Beckman. Centrogenic pulmonary pathology in mechanical head injury. J. Appl. Physiol. 27: 807-812, 1969. Bean, J. W., D. Zee, and B. Thom. Pulmonary changes with convulsions induced by drugs and oxygen at high pressure. J. Appl. Physiol. 21: 865-872, 1966. Bergofsky, E. H. Humoral control of the pulmonary circulation. In: Annual Review of Physiology. Edited by I. S. Edelman, Vol. 42, Palo Alto, 0a.: Annual Reviews, Inc., 1980, p. 221-233. Bergofsky, E. H. Mechanisms underlying vasomotor regulation of regional pulmonary blood flow in normal and disease states. Am. J. Med. 57: 378-394, 1974. Blaisdell, F. W. and F. R. Lewis, Jr. In: Respiratory Distress Syndrome of Shock and Trauma. Major Problems in Clinical Surgery. Vol. 21, edited by P. A. Ebert. Philadelphia, W. B. Saunders, 1977. Blake, L. H. Mathematical modeling of steady state fluid and protein exchange in lung. In: Lunngater and Solute Exchange--Lung Biology in Health and Disease. Vol. 7, edited by N. C. Staub, New York, Marcel Dekker, Inc., 1978, p. 99-128. Bonbrest, H. C. Pulmonary edema following an epileptic seizure. Am. Rev. Resp. Dis. 91: 97-100, 1965. Borst, H. G., E. Berglund and M. McGregor. The effects of pharma- cologic agents on the pulmonary circulation in the dog. Studies on epinephrine, nor-epinephrine, 5-hydroxytryptamine, acetylcholine, histamine and aminophylline. J. Clin. Invest. 36:669-675, 1957. Borst, H. G., J. L. Whittenberger, E. Berglund and M. McGregor. Effects of unilateral hypoxia and hypercapnia on pulmonary blood flow distribution in the dog. Am. J. Physiol. 191:446-452, 1957. 24. 25. 26. 27. 28. 29. 30. 31. 32. 33. 34. 35. 36. 37. 180 Borst, H. G., M. McGregor, J. L. Whittenberger and E. Berglund. Influence of pulmonary arterial and left atrial pressures on pulmonary vascular resistance. Circ. Res. 4: 393-399, 1956. Bressack, M. A. and R. D. Bland. Intravenous infusion of tolazoline reduces pulmonary vascular resistance and net fluid filtration in the lungs of awake, hypoxic newborn lambs. Am. Rev. Respir. Dis. 123: 217-221, 1981. Brigham, K. L. Lung lymph composition and flow in experimental pulmonary edema. In: Pulmonary Edema. Edited by A. P. Fishman and E. M. Renkin, Bethesda, American Physiological Society, 1979, p. 161-173. Brigham, K. L. and P. J. Owen. Increased sheep lung vascular perme- ability caused by histamine. Circ. Res. 37: 647-657, 1975. Brody, J. S. and E. J. Stemmler. Differential reactivity in the pulmonary circulation. J. Clin. Invest. 47:800-808. 1968. Brody, J. S., E. J. Stemmler and A. B. DuBois. Longitudinal distri- bution of vascular resistance in the pulmonary arteries, capil- laries and veins. J. Clin. Invest. 47: 783-799, 1968. Brown, R. W. Centroneurogenic control of the respiratory distress syndrome. Fed. Proc. 35: 324, 1976. Bruderman, I., K. Somers, W. K. Hamilton, W. H. Tooley and J. Butler. Effect of surface tension on circulation in the excised lungs of dogs. J. Appl. Physiol. 19: 707-712, 1964. Bruner, H. D. and C. F. Schmidt. Blood flow in the bronchial artery of the anesthetized dog. Am. J. Physiol. 148: 648-666, 1947. Cameron, G. R. Pulmonary oedema. Br. Med. J. 1: 965-972, 1948. Cameron, G. R. and S. N. De. Experimental pulmonary oedema of nervous origin. J. Pathol. Bacteriol. 61: 375-387, 1949. Campbell, G. 5., F. J. Haddy, W. L. Adams and M. B. Visscher. Circu- latory changes and pulmonary lesions in dogs following increased intracranial pressure, and the effect of atropine upon such changes. Am. J. Physiol. 158: 96-102, 1949. Chen, H. I. and C. Y. Chai. Pulmonary edema and hemorrhage as a consequence of systemic vasoconstriction. Am. J. Physiol. 227: 144-151, 1974. Chen, H. I., J. D. Lin and J. F. Liao. Participation of regional sympathetic outflows in the centrogenic pulmonary pathology. Am. J. Physiol. 240: H109-H115, 1981. 38. 39. 40. 41. 42. 43. 44. 45. 46. 47. 48. 49. 50. 51. 181 Chen, H. I., S. C. Sun and C. Y. Chai. Pulmonary edema and hemorrhage resulting from cerebral compression. Am. J. Physiol. 224: 223-229, 1973. Chinard, F. P. Estimation of extravascular lung water by indicator- dilution techniques. Circ. Res. 37: 137-145, 1975. Chinard, F. P. Exchanges across the alveolar-capillary barrier. In: The Pulmonary_Circulation and Interstitial Space. Edited by A. P. Fishman and’H. H. Hecht, Chicago, The University of Chicago Press, 1969, p. 79-98. Chinard, F. P., T. Enns and M. F. Nolan. Indicator-dilution studies with ”diffusible" indicators. Circ. Res. 10: 473-490, 1962. Colebatch, H. J. H. and L. A. Engel. Constriction of the lung by histamine before and after adrenalectomy in cats. J. Appl. Physiol. 37:798-805, 1974. Comroe, J. H., Jr. In: Physiolpgy of Respiration. Second edition, Chicago, Year Book Medical Publishers, Inc.,‘l977. Comroe, J. H., Jr. The location and function of the chemoreceptors of the aorta. Am. J. Physiol. 127: 176-191, 1939. Cottrell, T. S., 0. R. Levine, R. M. Senior, J. Weiner, D. Spiro and A. P. Fishman. Electron microscopic alterations at the alveolar level in pulmonary edema. Circ. Res. 21: 783-797, 1967. Crockard, H. A., J. C. Clark, M. L. Warnock, A. M. Churg and K. Hanlon. The significance of pulmonary changes associated with cerebral perfusion with hypoxic blood in monkeys. Surg. 82: 588- 598, 1977. Crone, C. The permeability of capillaries in various organs as determined by use of the ’indicator diffusion' method. Acta Physiol. Scand. 58: 292-305, 1963. Cudkowicz, L. In: The Human Bronchial Circulation In Health And Disease. Baltimore, The Williams & Wilkins Co., 1968. Culver, B. H. and J. Butler. Mechanical influences on the pulmonary microcirculation. In: Annual Review of Physiology. Edited by I. S. Edelman, Vol. 42, Palo Alto, 0a.: AnnualiReviews, Inc., 1980, p. 187-198. Daly, I. de B. Intrinsic mechanisms of the lung. 0. J. Exp. Physiol. 43: 2-26, 1958. Daly, I. de B. and C. Hebb. Pulmonary vasomotor fibers in the cervical vagosympathetic nerve of the dog. 0. J. Exp. Physiol. 37: 19-43, 1952. 52. 53. 54. 55. 56. 57. 58. 59. 60. 61. 62. 63. 182 Daly, I. de B. and M. de B. Daly. Sympathetic nerve control of pulmonary vascular resistance and impedance in isolated perfused lungs of the dog. J. Physiol. (London) 234: 106P-108P, 1973. Daly, I. de 8. and M. de B. Daly. The effects of stimulation of the carotid body chemoreceptors on the pulmonary vascular bed in the dog: the 'vasosensory controlled perfused living animal' prepara- tion. J. Physiol (London) 148: 201-219, 1959. Daly, I de 8., H. Duke, C. 0. Hebb and J. Weatherall. Pulmonary vasomotor fibers in the sympathetic chain and its associated ganglia in the dog. 0. J. Exp. Physiol. 34: 285-313, 1948. Daly, M. de B. and M. J. Scott. The cardiovascular responses to stimulation of the carotid body chemoreceptors in the dog. J. Physiol. (London) 165: 179-197, 1963. Daly, M. de B., C. J. Lambertsen and A. Schweitzer. The effects upon the bronchial musculature of altering the oxygen and carbon dioxide tensions of the blood perfusing the brain. J. Physiol. 119: 292-314, 1953. Daly, M. de 8., J. L. Hazzledine and A. Howe. Reflex respiratory and peripheral vascular responses to stimulation of the isolated perfused aortic arch chemoreceptors of the dog. J. Physiol. 177: 300-322, 1965. Daugherty, R. M., Jr., J. B. Scott, J. M. Dabney and F. J. Haddy. Local effects of 0 and CO on limb, renal, and coronary vascular resistances. Am. 3. Physial. 213: 1102-1110, 1967. Dawson, C. A., D. J. Grimm and J. H. Linehan. Effects of lung inflation on longitudinal distribution of pulmonary vascular resistance. J. Appl. Physiol. 43: 1089-1092, 1977. Dawson, C. A., D. J. Grimm and J. H. Linehan. Influence of hypoxia on the longitudinal distribution of pulmonary vascular resistance. J. Appl. Physiol. 44: 493-498, 1978. Dawson, 0. A., T. E. Forrester, and L. H. Hamilton. Effect of hypoxia and histamine infusion on lung blood volume. J. Appl. Physiol. 38: 811-816, 1975. DeGeest, H., M. N. Levy and H. Zieske. Reflex effects of cephalic hypoxia, hypercapnia, and ischemia upon ventricular contractility. Circ. Res. 17: 349-358, 1965. Diamond, L. Potentiation of bronchomotor responses by beta adrenergic antagonists. J. Pharmac. Exptl. Therap. 181: 434-445, 1972. 183 63a. Divertie, M. B. and T. L. Petty. Adult respiratory distress syndrome. In: Current Concepts. Kalamazoo, MI, The Upjohn Company, 1979. 64. Doba, N. and D. J. Reis. Acute fulminating neurogenic hypertension produced by brainstem lesions in the rat. Circ. Res. 32: 584-593, 973. 65. Dollery, C. T. and J. B. Glazier. Pharmacological effects of drugs on the pulmonary circulation in man. Clin. Pharmacol. Ther. 7: 807-818, 1966. 66. Downing, S. E. and J. C. Lee. Nervous control of the pulmonary circulation. In: Annual Review of Physiology. Edited by I. S. Edelman, Vol. 42, Palo Alto, 0a.: Annual Reviews, Inc., 1980, p. 199-210. 67. Downing, S. E. and J. H. Siegel. Baroreceptor and chemoreceptor influences on sympathetic discharge to the heart. Am. J. Physiol. 204: 471-479, 1963. 68. Downing, S. E., J. H. Mitchell and A. J. Wallace. Cardiovascular responses to ischemia, hypoxia, and hypercapnia of the central nervous system. Am. J. Physiol. 204: 881-887, 1963. 69. Downing, S. E., J. P. Remensnyder and J. H. Mitchell. Cardiovascular responses to hypoxic stimulation of the carotid bodies. Circ. Res. 10: 676-685, 1962. 70. Drake, R. E. and J. C. Gabel. Effect of histamine and alloxan on canine pulmonary vascular permeability. Am. J. Physiol. 239: H96-H100, 1980. 71. Drake, R. E., J. H. Smith and J. C. Gabel. Estimation of the filtra- tion coefficient in intact dog lungs. Am. J. Physiol. 238: H430-H438, 1980. 72. Drake, R., K. A. Gaar and A. E. Taylor. Estimation of the filtration coefficient of pulmonary exchange vessels. Am. J. Physiol. 234: H266-H274, 1978. 72a. Drake, R., T. Adair, D. Traber, and J. Gabel. Contamination of caudal mediastinal node efferent lymph in sheep. Am. J. Physiol. 241: H354-H357, 1981. 73. Ducker, T. B. Increased intracranial pressure and pulmonary edema. Part 1: Clinical study of 11 patients. J. Neurosurg. 28: 112-117, 1968. 74. Ducker, T. B., R. L. Simmons and R. W. Anderson. Increased Intra- cranial pressure and pulmonary edema. Part 3: the effect of increased intracranial pressure on the cardiovascular hemodynamics of chimpanzees. J. Neurosurg. 29: 475-483, 1968. 75. 76. 77. 78. 79. 80. 81. 82. 83. 84. 85. 86. 87. 88. 184 Duke, H. N. Pulmonary vasomotor responses of isolated perfused cat lungs to anoxia and hypercapnia. Q. J. Exp. Physiol. 36: 75-88, 1951. Eaton, R. M. Pulmonary edema: Experimental observations on dogs following acute peripheral blood loss. J. Thoracic. Surg. 16: 668-694, 1947. Ehrhart, I. C., P. E. Parker, W. J. Weidner, J. M. Dabney, J. B. Scott and F. J. Haddy. Coronary vascular and myocardial responses to carotid body stimulation in the dog. Am. J. Physiol. 229: 754-760, 1975. Falck, B., T. Nystedt, E. Rosengren and J. Stenflo. Dopamine and mast cells in ruminants. Acta Pharmacol. et Toxicol. 21: 51-58, 1964. Felman, A. H. Neurogenic pulmonary edema. Observations in 6 patients. Am. J. Roentgenol. Radium Ther. Nucl. Med. 112: 393-396, 1971. Fillenz, M. B. Innervation of pulmonary and bronchial blood vessels of the dog. J. Anat. 106: 449-461, 1970. Fishman, A. P. Adult respiratory distress syndrome. In: Pulmonary Diseases and Disorders. Vol. 2, New York, McGraw-Hill Book Co., 1980, p. 1667-1681. Fishman, A. P. Dynamics of the pulmonary circulation. In: Handbook of Physiology-Circulation, Section 2, Vol. 2, Bethesda, American Physiological Society, 1965, p. 1667-1743. Fishman, A. P. Hypoxia on the pulmonary circulation - How and where it acts. Circ. Res. 38: 221-231, 1976. Fishman, A. P. Pulmonary edema. In: Pulmonary Diseases and Disorders. Vol. 1, New York, McGraw-Hill Book Co., 1980, p. 733-753. Fishman, A. P. Respiratory gases in the regulation of the pulmonary circulation. Physiol. Rev. 41: 214-280, 1961. Fishman, A. P. and G. G. Pietra. Hemodynamic pulmonary edema. In: Pulmonary Edema. Edited by A. P. Fishman and E. M. Renkin, Bethesda, American Physiological Society, 1979, p. 79-96. Gaar, K. A., Jr., A. E. Taylor and A. C. Guyton. Effect of lung edema on pulmonary capillary pressure. Am. J. Physiol. 216: 1370-1373, 1969. Gaar, K. A., Jr., A. E. Taylor, L. J. Owens and A. C. Guyton. Pulmonary capillary pressure and filtration coefficient in the iso- lated perfused lung. Am. J. Physiol. 213: 910-914, 1967. 89. 90. 91. 92. 93. 94. 95. 96. 97. 98. 99. 100. 101. 185 Gabbiani, G., M. C. Badonnel, C. Gervasoni, B. Portmann and G. Majno. Carbon deposition in bronchial and pulmonary vessels in response *0 vasoactive compounds. Proc. Soc. Exp. Biol. Med. 140: 958-962, 972. Gabel, J. C. and R. E. Drake. Pulmonary capillary pressure in intact dog lungs. Am. J. Physiol. 235: H569-H573, 1978. Gamble, J. E. and H. D. Patton. Pulmonary edema and hemorrhage from preoptic lesions in rats. Am. J. Physiol. 172: 623-631, 1953. Gilbert, R. P., L. B. Hinshaw, H. Kuida and M. B. Visscher. Effects of histamine, 5 hydroxytryptamine and epinephrine on pulmonary hemodynamics with particular reference to arterial and venous seg- ment resistances. Am. J. Physiol. 194: 165-170, 1958. Glazier, J. B. and J. F. Murray. Sites of pulmonary vasomotor reactivity in the dog during alveolar hypoxia and serotonin and histamine infusion. J. Clin. Invest. 50: 2550-2558, 1971. Goetzman, B. W. and M. B. Visscher. The effects of alloxan and histamine on the permeability of the pulmonary alveolocapillary barrier to albumin. J. Physiol. (London) 204: 51-61, 1969. Golden, P. F. and J. J. Jane. Survival following profound hypo- volemia: Role of heart, lung, and brain. J. Trauma 9: 784-798, 1969. Grand, M. G. and S. E. Downing. Metabolic and reflex influences on pulmonary vasomotion. Am. J. Physiol. 218: 654-661, 1970. Green, H. D. and C. E. Rapela. Cerebral vascular responses to localized and systemic hypotension induced by hemorrhage and shock. In: Microcirculation as Related to Shock. New York, Academic Press, 1968, p. 93-119. Greenberg, 5., P. J. Kadowitz, A. Hyman and F. A. Curro. Adrenergic mechanisms in canine intralobar pulmonary arteries and veins. Am. J. Physiol. 240: H274-H285, 1981. Grega, G. J., D. E. Dobbins, J. B. Scott and F. J. Haddy. Effects of histamine and increased venous pressure on transmicrovascular protein transport. Microvasc. Res. 18: 95-104, 1979. Grega, G. J., D. L. Marciniak, B. S. Jandhyala and R. M. Raymond. Effects of intravenously infused histamine on canine forelimb transvascular protein efflux following adrenergic receptor blockade. Circ. Res. 47: 584-591, 1980. Grega, G. J., J. J. Maciejko, R. M. Raymond and D. P. Sak. The inter- relationship among histamine, various vasoactive substances, and macromolecular permeability in the canine forelimb. Circ. Res. 46: 264-275, 1980. 102. 103. 104. 105. 106. 107. 108. 108a. 109. 110. 111. 112. 186 Grega, G. J., R. L. Kline, D. E. Dobbins and F. J. Haddy. Mechanisms of edema formation by histamine administered locally into canine forelimbs. Am. J. Physiol. 223: 1165-1171, 1972. Grega, G. J., R. M. Daugherty, Jr., J. B. Scott, D. P. Radawski and F. J. Haddy. Effect of pressure, flow, and vasoactive agents on vascular resistance and capillary filtration in canine fetal, newborn, and adult lung. Microvasc. Res. 3: 297-307, 1971. Guyton, A. C. In: Textbook of Medical Physiolpgy. Sixth edition, Philadelphia, W. B. Saunders Co., 1981. Guyton, A. C. and A. W. Lindsey. Effect of elevated left atrial pressure and decreased plasma protein concentration on the development of pulmonary edema. Circ. Res. 7: 649-657, 1959. Guyton, A. C., J. C. Parker, A. E. Taylor, T. E. Jackson and D. S. Moffatt. Forces governing water movement in the lung. In: Pulmonary Edema. Edited by A. P. Fishman and E. M. Renkin, Bethesda, American Physiological Society, 1979, p. 65-77. Haddy, F. J. Effect of histamine on small and large vessel pressures in the dog foreleg. Am. J. Physiol. 198: 161-168, 1960. Haddy, F. J., J. B. Scott and G. J. Grega. Effects of histamine on lymph protein concentration and flow in the dog forelimb. Am. J. Physiol. 223: 1172-1177, 1972. Haddy, F. J., J. B. Scott and G. J. Grega. Peripheral circulation: Fluid transfer across the microvascular membrane. In: International Review of Physiology-Cardiovascular Physiology II, VOlume 9, edited by A. C. Guyton and A. W. Cowley, Baltimore, University Park Press, 1976, p. 63-109. Haggendal, E. and B. Johansson. Effects of arterial carbon dioxide tension and oxygen saturation on cerebral blood flow autoregula- tion in dogs. Acta Physiol. Scand. 258: 27-53, 1965. Hakim, T. S. and C. A. Dawson. Sympathetic nerve stimulation and vascular resistance in a pump-perfused dog lung lobe. Proc. Soc. Exp. Biol. Med. 160: 38-41, 1979. Hakim, T. S., F. L. Minnear, H. van der Zee, P. S. Barie and A. B. Malik. Adrenoreceptor control of lung fluid and protein exchange. J. Appl. Physiol. 51: 68-72, 1981. Hakim, T. S., H. van der Zee and A. B. Malik. Effects of sympathetic nerve stimulation on lung fluid and protein exchange. J. Appl. Physiol. 47: 1025-1030, 1979. 113. 114. 115. 116. 117. 118. 119. 120. 121. 122. 123. 124. 125. 187 Halmagyi, D. F. J. Role of lymphatics in the genesis of "shock lung": A hypothesis. In: Lung Water and Solute Exchange- Lung Biology in Health and Disease. Vol. 7, edited by N. C. Staub, New York, Marcel Dekker, Inc., 1978, p. 423-435. Hamer, J., S. Hoyer, E. Alberti and F. Weinhardt. Cerebral blood flow and oxidative brain metabolism during and after moderate and profound arterial hypoxaemia. Acta Neurochir. 33: 141-150, 1976. Harris, P. and 0. Heath. In: The Human Pulmonary Circulation. Second edition, London, ChurChill Livingstone,T977. Hayek, H. von. In: The Human Lung. New York, Hafner Publishing Co., Inc., 1960. Hebb, C. Motor innervation of the pulmonary vessels of mammals. In: The Pulmonary Circulation and Interstitial Space. Edited by A. P. Fishman and H. H. Hecht, Chicago, The University of Chicago Press, 1969, p. 195-222. Heistad, D. D., F. M. Abboud, A. L. Mark and P. G. Schmid. Inter- action of baroreceptor and chemoreceptor reflexes. Modulation of the chemoreceptor reflex by changes in baroreceptor activity. J. Clin. Invest. 53: 1226-1236, 1974. Henry, J. N., A. H. McArdle, H. J. Scott and F. N. Gurd. A study of the acute and chronic respiratory pathophysiology of hemorrhagic shock. J. Thoracic Cardiovasc. Surg. 54: 666-681, 1967. Hessler, J. R. and S. Cassin. Effects of increased intracranial pressure on pulmonary vascular resistance of fetal and neonatal goats. Am. J. Physiol. 232: H671-H675, 1977. Hirsch, E. F. and G. C. Kaiser. In: The Innervation of the Lung. Springfield, C. C. Thomas, 1969. Hopewell, P. C. and J. F. Murray. The adult respiratory distress syndrome. Ann. Rev. Med. 27: 343-356, 1976. Howard, P., G. R. Barer, B. Thompson, P. M. Warren, C. J. Abbott and I. P. F. Mungall. Factors causing and reversing vasoconstric- tion in unventilated lung. Respir. Physiol. 24: 325-345, 1975. Huff, R. W. and H. L. Fred. Postictal pulmonary edema. Arch. Intern. Med. 117: 824-828, 1966. Hughes, J. M. B., J. B. Glazier, J. E. Maloney and J. B. West. Effect of extra-alveolar vessels on distribution of blood flow in the dog lung. J. Appl. Physiol. 25: 701-712, 1968. 126. 127. 128. 129. 130. 131. 132. 133. 134. 135. 136. 137. 137a. 138. 188 Hultgren,+L N., C. E. Lopez, E. Lundberg and H. Miller. Physiologic studies on pulmonary edema at high altitude. Circulation 29: 393-408, 1964. Ann. Rev. Med. 17: 431-446, Hyman, A. L. The pulmonary veins. 1966. Iliff, L. D. excised dog lungs. Extra-alveolar vessels and edema development in Circ. Res. 28: 524-532, 1971. Ingram, R. H., J. P. Szidon, R. Skalak and A. P. Fishman. Effects of sympathetic nerve stimulation on the pulmonary arterial tree of the isolated lobe perfused in situ. Circ. Res. 22: 801-815, 1968. Jarisch, A. H., H. Richter and H. Thoma . Klin. Wchnschr. 18: 1440-1443, 1939. Zentrogenes Lungenddem. Joiner, P. 0., P. J. Kadowitz, J. P. Hughes and A. L. Hyman. NE and ACh responses of intrapulmonary vessels from dog, swine, sheep and man. Am. J. Physiol. 228: 1821-1827, 1975. Effect of sympathetic nerve stimu- Circ. Res. Kadowitz, P. J. and A. L. Hyman. lation on pulmonary vascular resistance in the dog. 32: 221-227, 1973. Kadowitz, P. J., P. D. Joiner and A. L. Hyman. Effect of sympathetic nerve stimulation on pulmonary vascular resistance in the intact spontaneously breathing dog. Proc. Soc. Exp. Biol. Med. 147: 68-71, 1974. Kadowitz, P. J., P. D. Joiner and A. L. Hyman. Influence of sympathetic stimulation and vasoactive substances on the canine pulmonary veins. J. Clin. Invest. 56: 354-365, 1975. Kato, M. and N. C. Staub. unilobar hypoxia and hypercapnia. Response of small pulmonary arteries to Circ. Res. 19: 426-440, 1966. Kazemi, H., P. Bruecke and E. Parsons. Role of the autonomic nervous system in the hypoxic response of the pulmonary vascular bed. Respir. Physiol. 15: 245-254, 1972. Keller, C. A., R. J. Schramel, A. L. Hyman and O. Creech, Jr. The cause of acute congestive lesions of the lung. J. Thoracic and Cardiovasc. Surg. 53: 743-753, 1967. Transient effects of Microvasc. Res. Korthuis, R. J., C. Y. Wang and J. B. Scott. histamine on microvascular fluid movement. (In Press), 1982. Krahl, V. E. Anatomy of the mammalian lung. In: Handbook of Ph siolo -Res iration. Vol. 1, Sect. 3, Baltimore, Williams and Wiikins, 1964, p. 267-284. 139. 140. 141. 142. 143. 144. 145. 146. 147. 148. 149. 150. 151. 189 Kuida, H., T. J. Tsagaris and H. H. Hecht. Evidence for pulmonary venoconstriction in brisket disease. Circ. Res. 12: 182-189, 1963. Lee, G. deJ. Regulation of the pulmonary circulation. Brit. Heart J. 33(supp1ement): 15-26, 1971. Levine, O. R., R. B. Mellins, R. M. Senior and A. P. Fishman. The application of Starling's law of capillary exchange to the lungs. J. Clin. Invest. 46: 934-944, 1967. Levitzky, M. G., J. C. Newell, J. A. Krasney and R. E. Dutton. Chemoreceptor influence on pulmonary blood flow during unilateral hypoxia in dogs. Resp. Physiol. 31: 345-356, 1977. Lloyd, T. 0., Jr. Effect of alveolar hypoxia on pulmonary vascular resistance. J. Appl. Physiol. 19: 1086-1094, 1964. Lloyd, T. C., Jr. Effect of increased intracranial pressure on pulmonary vascular resistance. J. Appl. Physiol. 35: 332-335, 1973. Low, F. N. Lung interstitium-development, morphology, fluid con- tent. In: Lung Water and Solute Exchange-Lung Biology in Health and Disease. Vol. 7, edited by N. C. Staub, New York, Marcel Dekker, Inc., 1978, p. 17-48. Luisada, A. A. and S. J. Sarnoff. Paroxysmal pulmonary edema con- sequent to stimulation of cardiovascular receptors. 1. Effect of intra-arterial and intravenous infusions. Am. Heart. J. 31: 270-281, 1946. MacKay, E. M. Experimental pulmonary edema. IV. Pulmonary edema accompanying trauma to the brain. Soc. Exp. Biol. Med. (Proc.) 74: 695-697, 1950. Macmillan, W. H. and J. R. Vane. The effects of histamine on the plasma potassium levels of cats. J. Pharmacol. Exp. Ther. 118: 182-192, 1956. Maire, F. W. and H. D. Patton. Neural structures involved in the genesis of 'preoptic pulmonary edema', gastric erosions and behavior changes. Am. J. Physiol. 184: 345-350, 1956. Majno, G., S. M. Shea and M. Leventhal. Endothelial contraction induced by histamine-type mediators. J. Cell Biol. 42: 617-672, 1969. Malik, A. B. Pulmonary vascular response to increase in intra- cranial pressure: role of sympathetic mechanisms. J. Appl. Physiol. 42: 335-343, 1977. 152. 153. 154. 155. 156. 157. 158. 159. 160. 161. 162. 163. 164. 190 Marciniak, D. L., D. E. Dobbins, J. J. Maciejko, J. B. Scott, F. J. Haddy and G. J. Grega. Antagonism of histamine edema formation by catecholamines. Am. J. Physiol. 234: H180-H185, 1978. Marciniak, D. L., D. E. Dobbins, J. J. Maciejko, J. B. Scott, F. J. Haddy and G. J. Grega. Effects of systemically infused histamine on transvascular fluid and protein transfer. Am. J. Physiol. 233: H148-H153, 1977. Maron, M. B. Differential effect of histamine on protein perme- ability in the dog lung and forelimb. Physiologist 24: 54, 1981. Maron, M. B., T. S. Hakim and C. A. Dawson. Pulmonary hemodynamic responses to elevated cerebral spinal fluid pressure in the dog. J. Appl. Physiol. 46: 84-88, 1979. Maseri, A., P. Caldini, P. Harward, R. C. Joshi, S. Permutt and K. L. Zierler. Determinants of pulmonary vascular volume. Recruitment versus distensibility. Circ. Res. 31: 218-228, 1972. McGillicuddy, J. E., G. W. Kindt, J. E. Raisis and C. A. Miller. The relation of cerebral ischemia, hypoxia and hypercarbia to the Cushing response. J. Neurosurg. 48: 730-740, 1978. McLaughlin, R. F., Jr., W. S. Tyler and R. O. Canada. The compara- tive anatomy of the pulmonary vascular tree. Med. Thorac. 19: 523-527, 1962. Millen, J. E., F. L. Glauser and M. Zimmerman. Physiological effects of controlled concussive brain trauma. J. Appl. Physiol. 49: 856-862, 1980. Miller, M. E., G. C. Christensen and H. E. Evans. The heart and arteries. In: Apatomy of the Dog. Philadelphia, W. B. Saunders Co., 1964, p. 267-429. Miller, W. S. In: The Lung. Second Edition, Springfield, C. C. Thomas, 1947. Mitzner, W. and J. L. Robotham. Distribution of interstitial compliance and filtration coefficient in canine lung. Lymphology 12: 140—148, 1979. Morgan, B. C., S. C. Church and W. G. Guntheroth. Hypoxic constric- tion of pulmonary artery and vein in intact dogs. J. Appl. Physiol. 25: 356-361, 1968. Morkin, E., J. A. Collins, H. S. Goldman and A. P. Fishman. Pattern of blood flow in the pulmonary veins of the dog. J. Appl. Physiol. 20: 1118-1128, 1965. 165. 166. 167. 168. 169. 170. 171. 172. 173. 174. 175. 176. 177. 191 Moss, G. and A. A. Stein. Cerebral etiology of the acute respira- tory distress syndrome: Diphenylhydantion prophylaxis. J. Trauma 15: 39-41, 1975. Moss, G. and A. A. Stein. Shock lung: Anemia as a predisposing factor. Am. J. Surg. 126: 419-420, 1973. Moss, G. and A. A. Stein. The centrineurogenic etiology of the respiratory distress syndrome--Induction by isolated cerebral hypoxemia and prevention by unilateral pulmonary denervation. Am. J. Surg. 132: 352-357, 1976. Moss, G., C. Staunton and A. A. Stein. The centrineurogenic etiology of the acute respiratory distress syndromes. Universal, species- independent phenomenon. Am. J. Surg. 126: 37-41, 1973. Murakami, T., S. D. Wax and W. R. Webb. Pulmonary microcirculation in hemorrhagic shock. Surg. Forum 21: 25-27, 1970. Murray, J. F. In: The Normal Lung;y_The Basis for Diagnosis and Treatment of Pulmonary Disease. Philadelphia, W. B. Saunders Co., 1976. Nakahara, K., K. Ohkuda and N. C. Staub. Effect of infusing hista- mine into pulmonary or bronchial artery on sheep pulmonary fluid balance. Am. Rev. Resp. Dis. 120: 875-882, 1979. Nickerson, M. and B. Collier. Drugs inhibiting adrenergic nerves and structures innervated by them. In: The Pharmacological Basis of Therapeutics. Edited by L. S. Goodman and A. Gilman, 5th edition. New York, Macmillan Publishing Co., 1975, p. 533-564. Noble, W. H. and J. W. Severinghaus. Thermal and conductivity dilution curves for rapid quantitation of pulmonary edema. J. Appl. Physiol. 32: 770-775, 1972. Noble, W. H., J. Obdrzalek and J. C. Kay. A new technique for measuring pulmonary edema. J. Appl. Physiol. 34: 508-512, 1973. Opdyke, D. F. Circulatory effects of partial cerebral ischemia. Amer. J. Physiol. 146: 467-477, 1946. Oppenheimer L., V. B. Elings and F. R. Lewis. Thermal-dye lung water measurements: Effects of edema and embolization. J. Surg. Res. 26: 504-512, 1979. Pace, J. B. Sympathetic control of pulmonary vascular impedance in anesthetized dogs. Circ. Res. 29: 555-568, 1971. 178. 179. 180. 181. 182. 183. 184. 185. 186. 187. 188. 189. 190. 192 Parker, J. 0., A. C. Guyton and A. E. Taylor. Pulmonary transcapil- lary exchange and pulmonary edema. In: International Review of Physiology-Cardiovascular Physiology III. Vol. 18, edited by A. C. Guyton and D. B. Young. Baltimore, University Park Press, 1979, p. 261-315. Parker, J. C., R. Allison and A. E. Taylor. Total lymph flow in normally hydrated dogs. Physiologist 24: 17, 1981. Parker, P. E., J. M. Dabney, J. B. Scott and F. J. Haddy. Reflex vascular responses in kidney, ileum, and forelimb to carotid body stimulation. Am. J. Physiol. 228: 46-51, 1975. Pearce, M. L. Sodium recovery from normal and edematous lungs studied by indicator dilution curves. Circ. Res. 24: 815-820, 1969. Pearce, M. L. and J. W. Beazell. The measurement of pulmonary parenchymal volume by thermal indicator dilution (Abstract). Clin. Res. 14: 182, 1966. Pearce, M. L., J. Yamashita and J. Beazell. Measurement of pulmonary edema. Circ. Res. 16: 482-488, 1965. Pelletier, C. L. Circulatory responses to graded stimulation of the carotid chemoreceptors in the dog. Circ. Res. 31: 431-443, 1972. Permutt, S. Mechanical influences on water accumulation in the lungs. In: Pulmonary Edema. Edited by A. P. Fishman and E. M. Renkin. Bethesda, American Physiological Society, 1979, p. 75-193. Persson, C. G. A., M. Ekman and I. Erjefalt. Vascular antiperme- ability effects of beta receptor agonists and theophylline in the lung. Acta Pharmacol. Toxicol. 44: 216-220, 1979. Peterson, B. T., S. E. Grauer, R. W. Hyde, C. Ortiz, H. Moosavi and M. J. Utell. Response of pulmonary veins to increased intra- cranial pressure and pulmonary air embolization. J. Appl. Physiol. 48: 957-964, 1980. Pietra, G. G., J. P. Szidon, M. M. Leventhal and A. P. Fishman. Histamine and interstitial pulmonary edema in the dog. Circ. Res. 24: 323-337, 1971. Pietra, G. G., M. Magno and L. Johns. Morphological and physio- logical study of the effect of histamine on the isolated perfused rabbit lung. Lymphology 12: 165-176, 1979. Quiroz, A. C., G. Sanchez, T. D. Giles and G. E. Burch. Angiographic demonstration of pulmonary venomotor reactions in the intact dog. Angiology 21: 724-727, 1970. _ __ ‘_ 191. 192. 193. 194. 194a. 195. 196. 197. 198. 199. 200. 201. 202. 193 Rawlings, C. A. The lungs. In: Pathophysiology in Small Animal Sur er . Edited by M. J. Bojrab. Philadelphia, Lea and Febiger, 1981, p. 377-393. Reid, L. Structural and functional reappraisal of the pulmonary artery system. In: Sci. Basis of Med. Ann. Rev. 1968, p. 289-307. Remington, J. W. and C. H. Baker. Blood and plasma specific gravity changes during acute alterations in hemodynamics in splenectomized dogs. Circ. Res. 6: 146-154, 1958. Richardson, J. B. Nerve supply to the lungs. Am. Rev. Resp. Dis. 119: 785-802, 1979. Rippe, B. and G. J. Grega. Effects of isoprenaline and cooling on histamine induced changes in capillary permeability in the rat hindquarter vascular bed. Acta Physiol. Scand. 103: 252-262, 1978. Rivera-EStrada, C., P. W. Saltzman, D. Singer and L. N. Katz. Action of hypoxia on the pulmonary vasculature. Circ. Res. 6: 0-14, 1958. Robin, E. D. Permeability pulmonary edema. In: Pulmonary Edema. Edited by A. P. Fishman and E. M. Renkin, Bethesda, American Physiological Society, 1979, p. 217-228. Robin, E. 0., C. E. Cross and R. Zelis. Pulmonary Edema. N. Eng. J. Med. 288: 239-246, 292-304, 1973. Robinson, N. E., G. A. Jones, J. B. Scott and J. M. Dabney. Effects of histamine and acetylcholine on equine digital lymph flow and composition. Proc. Soc. Exp. Biol. Med. 149: 805-807, 975. Robinson, N. E., J. B. Scott, J. M. Dabney and G. A. Jones. Digital vascular responses and permeability in equine alimentary laminitis. Am. J. Vet. Res. 37: 1171-1176, 1976. Robinson, R. L. and K. E. Jochim. The effect of histamine and anaphylactic shock on the secretion of adrenal medullary hormones. Fed. Proc. 18: 129, 1959. Rounthwaite, H. L., H. G. Scott and F. N. Gurd. Changes in the pulmonary circulation during hemorrhagic shock and resuscitation. Surg. Forum 3: 454-460, 1952. Rutherford, R. B., S. Kaihara, E. P. Schwentker and H. N. Wagner. Regional blood flow in hemorrhagic shock by the distribution of labeled microspheres. Surg. Forum 19: 14-15, 1968. 203. 204. 205. 206. 207. 208. 209. 210. 211. 212. 213. 214. 194 Said, S. I., T. Yoshida, S. Kitamura and C. Vreim. Pulmonary alveolar hypoxia: release of prostaglandins and other humoral mediators. Science 185: 1181-1183, 1974. Sarnoff, S. J. and L. C. Sarnoff. Neurohemodynamics of pulmonary edema. I. Autonomic influence on pulmonary vascular pressures €3d the acute pulmonary edema state. Dis. Chest 22: 685-696, 52. Sarnoff, S. J. and L. C. Sarnoff. Neurohemodynamics of pulmonary edema II. The role of sympathetic pathways in the elevation of pulmonary and systemic vascular pressures following the intra- cisternal injection of fibrin. Circ. 6: 51-62, 1952. Sayweed, M. M., I. H. Chaudry and A. E. Baue. Na+- K+ transport and adenosine nucleotides in the lung in hemorrhagic shock. Surg. 77: 395-402, 1975. Schneeberger, E. E. Barrier function of intercellular junctions in adult and fetal lungs. In: Pulmonary Edema. Edited by A. P. Fishman and E. M. Renkin. Bethesda, American Physiological Society, 1979, p. 21-37. Sealy, W. C., S. Ogino, A. M. Lesage and W. G. Young, Jr. Functional and structural changes in the lung in hemorrhagic shock. Surg. Gynec. Obstet. 122: 754-760, 1966. Share, L. and M. N. Levy. Effect of carotid chemoreceptor stimu- lation on plasma antidiuretic hormone titer. Am. J. Physiol. 210: 157-161, 1966. Simeone, F. A. and M. M. Witoszka. Changes in composition of cerebrospinal fluid in hemorrhagic shock. Surg. Forum 21: 53-55, 1970. Simeone, F. A. and M. M. Witoszka. The central nervous system in experimental hemorrhagic shock. Am. J. Surg. 119: 427-432, 1970. Simmons, R. L., A. M. Martin, Jr., C. A. Heisterkamp, III, and T. B. Ducker. Respiratory insufficiency in combat casualties. II. Pulmonary edema following head injury. Ann. Surg. 170: 39-44, 1969. Singh, 1., P. K. Hanna, M. C. Srivastava, M. Lal, S. B. Roy and C. S. V. Subramanyam. Acute mountain sickness. N. Eng. J. Med. 280: 175-184, 1969. Sjfistrand, T. Blood volume. In: Handbook of PhysiologygyCircula- tion, Section 2, Vol. 1, Washington, D.C., American Physiological Society, 1962, p. 52. 215. 216. 217. 218. 219. 220. 221. 222. 223. 223a. 224. 225. 226. 227. 228. 195 Slater, G., B. C. Vladeck, R. Bassin, R. S. Brown and W. C. Shoemaker. Sequential changes in cerebral blood flow and distri- I bution of flow within the brain during hemorrhagic shock. 1 Ann. Surg. 181: 1-4, 1975. Snow, H. M., C. M. Furnival and R. J. Linden. Pulmonary venous response to hypoxia in dogs.. Brit. Heart J. 32: 557, 1970. Staszewska-Barczak, J. and J. R. Vane. The release of catechol- amines from the adrenal medulla by histamine. Brit. J. Pharmacol. 25: 728-742, 1965. Staub, N. C. Pathways for fluid and solute fluxes in pulmonary edema. In: PulmonaryeEdema. Edited by A. P. Fishman and E. M. Renkin, Bethesda, American Physiological Society, 1979, p. 113-124. Staub, N. C. Pulmonary Edema. Physiol. Rev. 54: 678-811, 1974. Staub, N. C. Pulmonary edema due to increased microvascular permeability to fluid and protein. Circ. Res. 43: 143-151, 1978. Staub, N. C. The forces regulating fluid filtration in the lung. Microvas. Res. 15: 45-55, 1978. Staub, N. C. The pathophysiology of pulmonary edema. Hum. Path. 1: 419-432, 1970. Staub, N. C., H. Nagano and M. L. Pearce. Pulmonary edema in dogs, especially the sequence of fluid accumulation in lungs. J. Appl. Physiol. 22: 227-240, 1967. Staub, N. C., R. D. Bland, K. L. Brigham, R. Demling, A. J. Erdmann, and W. C. Woolverton. Preparation of chronic lung lymph fistulas in sheep. J. Surg. Res. 19: 315-320, 1975. Stern, S. and K. Braun. Effect of chemoreceptor stimulation on the pulmonary veins. Am. J. Physiol. 210: 535-539, 1966. Stern, 5., R. E. Ferguson and E. Rapaport. Reflex pulmonary vaso- constriction due to stimulation of the aortic body by nicotine. Am. J. Physiol. 206: 1189-1195, 1964. Sugg, W. L., W. D. Craver, W. R. Webb and R. R. Ecker. Pressure changes in the dog lung secondary to hemorrhagic shock: Protective effect of pulmonary reimplantation. Ann. Surg. 169: 592-598, 1969. Sugg, W. L., W. R. Webb and R. R. Ecker. Prevention of lesions of the lung secondary to hemorrhagic shock. Surg. Gynec. and Obstet. 127: 1005-1010, 1968. Sugg, W. L., W. R. Webb, S. Nakae, T. Theodorides, D. N. Gupta and W. A. Cook. Congestive atelectasis: An experimental study. Ann. Surg. 168: 234-242, 1968. 229. 230. 231. 232. 233. 234. 235. 236. 237. 238. 239. 240. 241. 196 Szidon, J. P. and A. P. Fishman. Autonomic control of the pulmonary circulation. In: The Pulmonary Circulation and Interstitial Space. Edited by A. P. Fishman and H. H. Hecht. Chicago, The University of Chicago Press, 1969, p. 239-268. Szidon, J. P. and J. F. Flint. Significance of sympathetic inner- vation of pulmonary vessels in response to acute hypoxia. J. Appl. Physiol. 43: 65-71, 1977. Taylor, A. E. and R. E. Drake. Fluid and protein movement across the pulmonary microcirculation. In: Lung Water and Solute Exchange-Lupg Biology in Health and Disease. Vol. 7, edited by N. C. Staub, New York, Marcel Dekker, Inc., 1978, p. 129-166. Theodore, J. and E. 0. Robin. Speculations on neurogenic pulmonary edema (NPE). Am. Rev. Resp. Dis. 113: 405-411, 1976. Thilenius, O. G., B. M. Candiolo and J. L. Beug. Effect of . . adrenergic blockade on hypoxia-induced pulmonary vasoconstriction in awake dogs. Am. J. Physiol. 213: 990-998, 1967. Visscher, M. B., F. J. Haddy and G. Stephens. The physiology and pharmacology of lung edema. Pharmac. Rev. 8: 389-434, 1956. Viswanathan, R. S., S. K. Jain and S. Subramanian. Pulmonary edema of high altitude. III. Pathogenesis. Am. Rev. Resp. Dis. 100: 342-350, 1969. Vranova, Z., H. Keszler and M. Vrana. Oxygen tension in the cerebral cortex of dogs in haemorrhagic shock and its changes after oxygen therapy. Acta Anaesth. Scandinav. 12: 171-182, 1968. Warren, M. F. and C. K. Drinker. The flow of lymph from the lungs of the dog. Am. J. Physiol. 136: 207-221, 1942. Weibel, E. R. The Ultrastructure of the alveolar-capillary membrane or barrier. In: The Pulmonary Circulation and Interstitial Space. Edited by A. P. Fishman and H. H. Hecht. Chicago, The University of Chicago Press, 1969, p. 9-27. Weibel, E. R. and H. Bachofen. Structural design of the alveolar septum and fluid exchange. In: Pulmonary Edema. Edited by A. P. Fishman and E. M. Renkin, Bethesda, American Physiological Society, 1979, p. 1-20. Weisman, S. J. Edema and congestion of the lungs resulting from intracranial hemorrhage. Surgery 6: 722-729, 1939. West, J. B. In: Pulmonary Pathophysiology--the essentials. Baltimore, Williams and Wilkins Co., 1977. 242. 243. 244. 245. 246. 247. 248. 249. 250. 251. 252. 197 West, J. B. In: Respiratory Physiology--the essentials. Second edition, Baltimore, Williams and Wilkins Co., 1979. Whayne, T. F., Jr. and J. W. Severinghaus. Experimental hypoxic pulmonary edema in the rat. J. Appl. Physiol. 25: 729-732, 1968. Whisnant, J. P., C. H. Millikan, K. G. Wakim and G. P. Sayre. Collateral circulation of the brain of the dog following bilateral ligation of the carotid and vertebral arteries. Am. J. Physiol. 186: 275-277, 1956. Widdicombe, J. G. and G. M. Sterling. The autonomic nervous system and breathing. Arch. Intern. Med. 126: 311-329, 1970. Wilcox, B. R., W. G. Austen and H. W. Bender. Effect of hypoxia on pulmonary artery pressure of dogs. Am. J. Physiol. 207: 1314-1318, 1964. Wilson, J. W. Mast cell morphologic alterations after acute pulmonary injury or respiratory distress. Chest 65(Supplement 4): SS-9S, 1974. Wood, C. 0., L. D. Seager and G. Ferrell. Influence of autonomic blockade on aconitine induced pulmonary edema. Proc. Soc. Exp. Biol. Med. 116: 809-811, 1956. Yashon, D., G. E. Locke, G. Paulson and C. A. Miller. Brain, CSF, and blood lactate levels during hemorrhagic shock. Surg. Forum 22: 23-24, 1971. Yipintsoi, T. Single-passage extraction and permeability estimation of sodium in normal dog lungs. Circ. Res. 39: 523-531, 1976. Yu, P. N. In: PulmonaryyBlood Volume in Health and Disease. Philadelphia, Lea and Febiger, 1969. Zierler, K. L. Theoretical basis of indicator-dilution methods for measuring flow and volume. Circ. Res. 10: 393-407, 1962. A