REGIONAL BRAIN BLOOD FLOW IN ENDOTOXIN SHWK Thesis for the Degree of M. S. MICHIGAN STATE UNIVERSITY WILLIAM 'J. BRYAN. 1976 --M_ I- '— L I I}? R 3% R Y P man 311 1 State UhiVi‘tSity REGIONAL BRAIN BLOOD FLOW IN ENDOTOXIN SHOCK By William J. Bryan A THESIS Submitted to Michigan State University in partial fulfillment of the requirements for the degree of MASTER OF SCIENCE Department of Physiology 1976 INTRODUCTION . TABLE OF CONTENTS SURVEY OF THE LITERATURE . Endotoxin Shock History Blood Brain Barrier. Microspheres Brain Circulation METHODS . . Microsphere Preparation Counting Tubes Brain Samples Calculations Group I, Control (n=6) Group II, Two Hour Shock (n=lO). . Group III, Four Hour Shock (n= 10) Statistical Analysis . RESULTS Control Group I . Group II (2 Hour Shock) Group III (A Hour Shock) DISCUSSION . BIBLIOGRAPHY . APPENDIX . ii LIST OF FIGURES Figure Page 1. Average hemodynamic responses of the cerebral vascular bed before and four hours after intra- venous administration of saline . . . . . . . . 21 2. Average hemodynamic responses of the cerebral vascular bed before and two hours after intra- venous administration of 5 mg/kg of E. coli endotoxin . . . . . . . . . . . . . . . . . . . 23 3. Average hemodynamic responses of the cerebral vascular bed before and four hours after intra- venous administration of 5 .mg/kg of E. coli endotoxin . . . . . . . . . . . . . . 26 INTRODUCTION Mechanisms of death during gram-negative endotoxin shock are unknown. This may account for the high mortality associated with this pathological disorder in man (15, 61). Evidence indicating a significant involvement of the cen- tral nervous system in the pathogenesis of shock has appeared from time to time in literature, beginning perhaps over a century ago (12).. However, a major contributory role of the central nervous system in shock has not been established and does not appear to have received serious consideration by a majority of scientists. Recent work from this laboratory shows that cerebral blood flow is severely decreased (50-60% of control) during endotoxin shock in the artificially (13, 1A) or spontaneously (13, 1N, 35) venti- lated dog; cerebral metabolic disorders, such as increased MR02, PaO2 and decreased Pv02’ cerebral venous and arterial glucose levels, during this condition have also been indi- cated in a preliminary study (M2). Earlier studies by others provide evidence that varying degrees of cerebral ischemia in the dog produces systemic hemodynamic alterations (5) such as decreased cardiac out- put and increased total peripheral resistance, and other abnormalities (32), such as pulmonary edema, similar to those associated with gram-negative endotoxin shock. Further support for an active role of the central nervous system in the pathogenesis of endotoxin shock comes from studies which suggest that endotoxin may act directly to initiate centrally mediated systemic hemodynamic dysfunc- tion (11, H7, 48), such as cardiac valve edema, pulmonary edema, and some intestinal congestion. While it has been clearly demonstrated that total cerebral blood flow is severely compromised during irrever- sible endotoxin shock in the dog (13, 35) and monkey (67), little information is available which related to regional cerebral blood flow particularly to flow in the cardiovas- cular and respiratory regulating centers of the hypothala- mus and medulla during endotoxin shock. Since regions of the brain can regulate their flow independent of total cerebral blood flow during normal conditions (21), it is reasonable to assume that a decrease in total cerebral blood flow during shock need not necessarily be prOportionate. There seem to be no studies in the literature con- cerning regional cerebral blood flow in endotoxin shock. Thus, the current study was therefore undertaken to deter- mine alterations in regional cerebral blood flow, particu- larly blood flow through regions involved in regulation of systemic cardiovascular function and respiration, during irreversible endotoxin shock. SURVEY OF THE LITERATURE I. Endotoxin Shock A. History Involvement of the central nervous system (CNS) in the pathogenesis of endotoxin shock has concerned physio- logists for many decades. As early as 1872 Dubezanski and Naunyn (12) suggested that pyrogenic substances induce fever and "vasomotor collapse" through central action on the vasomotor center. Later, in 1890, Charrin and Gley (8) found depressed responses of vasodilation and vasocon- striction when vasopressor and vasodepressor nerves were stimulated in dogs in endotoxin shock. They claimed the results were due to medullary paralysis. Anderson and Rose- unau (2) in 1909, investigated the responses of anaphalaxsis and endotoxin shock and agreed both were a result of toxic effects on the CNS. But a year later Pearce and Eisenbrey (37) refuted this finding showing that decapitation and mechanical destruction of the spinal cord did not protect against anaphylactic shock. Greater illumination of the positive contribution of the CNS in shock came when Penner and Klein (38) introduced their cross-perfusion technique. In paired dogs, the head of each animal was isolated from its own body circulation and perfused through one carotid artery with arterial blood from the other dog's body. They injected Shigella endo- toxin, producing shock only in the dog whose head was per- fused with blood from the injected dog. Weil gt_al. (66) repeated this experiment using E. coli endotoxin and concluded, along with Penner and Klein, that the CNS was not essential for the initiation of shock, but, participation of the CNS in the pathogenesis of this state cannot be ruled out. Using a similar method, Simmons g£_al. (A7, A8) found that even though differences exist between endotoxins given intra-ventricular or intracisternally, the endotoxin may act on the CNS to accentuate the development of shock. That is, small doses (down to 1.5 mg/kg) of endotoxin, given into the CNS, can induce severe pulmonary edema, congestion in the liver and hyperemic kidneys and intestinal mucosa, which can produce death in the absence of significant amounts of intravascular endotoxin. Recently, Dobbins (16) used a vascularly isolated, neural intact head-trunk preparation, in which a donor dog supplied a constant flow perfusion of arterial blood to the head of the recipient animal. He also found a decrease in blood pressure in the recipient dog head with the adminis- tration of endotoxin to the donor dog. Histamine, acetyl- choline, epinephrine, and other vasoactive drugs elicited blood pressure changes in the recipient trunk when adminis- tered to the perfusion circuit. These responses were abolished following bilateral denervation of the recipient 5 head's carotid sinus-body complex. Dobbins concludes that endotoxin is capable of eliciting centrally mediated hypo- tension responses. Also the severe hypotension can be demonstrated when endotoxin is administered into the trunk of a dog with the CNS excluded. It has been shown that the brain is very sensitive to a decrease in its tissue perfusion and that brain hypoxia can induce shock-like disorders (32). The lowered arterial P02 was accomplished without ischemia, acidemia or hypercar- bia and caused interstitial and intra-alveolar edema, hemorrhage and congestion. Gallo and Schenk (17) found a 50-60% decrease in blood flow, measured with magnetic flowmeters, in the carotid and vertebral arteries of the dog in endotoxin shock. Brown §t_al, (5) show that a decrease in cerebral perfusion pressures by unilateral carotid occlusion, causes the hemodynamic changes (decreased cardiac output, increased total peripheral resistance) similar to that seen in shock. Buyniski §§_al. (6) found that after one hour of endotoxin shock, there is a decrease responsive- ness to increased levels of arterial C02 in the dog. They found that there was a decrease in vascular conductance. Emerson and Parker (13, 1A) found in endotoxin shock, a decrease in cerebral blood flow of approximately 60% in dogs at four hours of shock. Also, the cerebral resistance was significantly higher (approximately 72%) by the fourth hour of shock. These results were later confirmed by Parker (35) showing that there is an increase in resistance occurring at three to four hours of shock. B. Blood Brain Barrier To answer the question of whether endotoxin crosses the blood brain barrier, Trippodo e£_al. (56) used the Limu- lus assay technique to show that no endotoxin was present in the cerebrospinal fluid two hours after intravenous endotoxin administration. They feel that endotoxin cannot cross the blood-brain barrier due to its high molecular weight (106) and thus cannot act directly on the CNS. However, the validity of the Limulus technique is question- able. Tamura (53) states there is evidence that hypotension damages the blood brain barrier. He concludes, "profound oligemic hypotension results in irreversible neural injury and in microvascular changes." Realizing that the permea- bility of the parenchymal capillaries which supply the brain with nutrition possess different characteristics than other organs, and that dyes diffuse through the more permeable vessels supplying the pituitary, tuber cinerum, and hypotha- lamus, Penner and Klein (38) suggested that toxins reach these areas and initiate diffuse sympathomimetic discharge. II. Microspheres Since the time of William Harvey, blood circulation and its measurement has interested scientists and clinicians alike. A number of techniques have been used, but possibly the first to use microspheres was Prinzmetal (Al), who in 19u7 investigated the possibility of arterioarterial and arteriovenous anastamoses in the human heart. Unlabeled glass microspheres of various sizes (IO-MO u) were injected into the arteries of a number of organs of several species. Appearance of microspheresixxthe venous efferent indicated the existence of arteriovenous shunts. Later, in the 1950's, low-density nonradioactive microspheres (wax) were used to study arteriovenous anastamoses in dog hindlegs and lungs (3, 36, HO, HA). However, these techniques proved unsatis- factory due to the complexity of assay and the hemodynamic alterations the microspheres produced. The advantages of the radioactive microsphere techni- que are that relatively slow measuring devices (gamma counter compared to a hand-radiation detector) can be used and the distribution of the radioactivity can be determined by an external radiation detector. But, the problem with the glass microspheres are their high density, the necessity of using a large quantity, and the non-quantitative methods for analysis (63). In 1960 microspheres were radioactively prepared by bombarding the glass beads with neutrons to convert the sodium to the glass to 2“Na (31). This improve- ment was adapted rapidly and Lahr and Ryan developed ceramic micrOSpheres labeled with Ll6Sc, 203 Hg, and other isotopes. By 1966 the ceramic microspheres and a new type of microsphere made cfi' plastic, available in twenty-seven different radio- nuclides (20) and varied from 0.5 to 1.0 to 80 microns withzaspecific activity as high as 50 p curies/gm. At this time the primary use of the microspheres with high specific activity of beta-emitting radionuclides was the treatment of malignant neoplasma; not circulation studies. Ya §t_al. (69) injected ceramic microspheres of uniform size (60 i 5 microns dia.) into particular organs to irradiate them without diffusion of the radioactivity throughout the body. He found that the hepatic and renal arteries trapped 99% 0Y, and 153Sm that were injected intra- of the 51Cr, 9 arterially. Histological examination levels showed little damage to the lungs or liver caused by the microspheres. The problem of the density of the microspheres was solved by Emmennegger gt_al. (16). Previously, the ceramic microspheres had a density of 3.0 gm/cc and the newer plas- tic microspheres were 1.5 gm/cc (blood density is 1.06 gm/cc) (18). Emmennegger prepared them from caunuba wax and phos- phoric acid ester, which were sprayed through a fine mesh and the solid droplets caught in water. The microspheres were then irradiated in a nuclear reactor to convert the 31F to 32F. In 1965 Halpern §t_§;, (23) introduced the first metabolizable radioactive particle for investigational and diagnostic use in patients. It was made of aggregates of human serum albumin for the study of phagocytosis. These particles, about ten microns in diameter, were rapidly absorbed and degraded by the Kupffer cells of the liver and the phagocytic cells of the spleen and bone marrow. These aggregates have advantages over the previously used particles, such as carbon, thorium dioxide, saccharated iron oxide, chromic phosphate, and colloid gold: (1) they do not remain in the phagocytic cells indefinitely; (2) they are metabo- lized rapidly after phagocytosis and excreted in the urine. Taplin et_al. (55), using the small size albumin aggregates (10—20 microns in dia.) scanned the liver, stomach, Spleen, and salivary glands for radioactive distribution. The first to scan human lungs using large albumin aggregates (20-50 u) were Wagner and associates (62). They established the macro—aggregated albumin (MAA), caused no hemodynamic, immunologic, or radiation hazard and were the best aid in the diagnosis of pulmonary embolism. There have been studies using MAA to determine rela- tive regional blood flow in different organs (A, 2A, 3A, 51, 58, 59, 60, 6A). However, a major drawback of the MAA in studying systemic circulation or arteriovenous shunting is the wide variation in particle sizes, which lead to the development of a more uniform sized albumin microspheres (63). Rhodes et_al. (43) developed such a particle although the distribution of sizes (6-65 u) was about the same as the MAA, these microspheres can be separated into more uniform fractions, prepared in large quantities and subjected to stricter quality control measures. The use of microspheres has been widely validated by now. There are three assumptions made when this technique is used: 1. the microspheres mix uniformly with the blood and have essentially the same rheology as red blood cells 10 2. the microspheres themselves do not alter blood flow 3. the microspheres are removed from the blood stream by becoming impacted in the microcircula- tion. Schibata and MacLean (A9) showed uniform distribution of microspheres in the lung after intravenous injection. Wagner gt_a1. (63) also found that albumin microspheres are evenly distributed in lungs, compared to simultaneously- administered macroaggregated albumin. Mixing of the microspheres was evaluated by Phibbs §t_al. (39) by rapidly freezing the artery in which micro- spheres were in transit. He found no settling of microspheres even when the specific gravity was slightly higher (1.1 to 1.6 gm/cc) than that of whole blood. Kaihara g§_al. (28) injected two different radionuclides five to eight minutes apart into the left atrium and found no difference in distri- bution through the heart, kidney, spleen, liver, or lung. But, when injected into the left ventricle or origin of the aorta there was a difference (m 50%) of activity in the coronary circulation. Rudolf and Heymann (A6) also showed that microsphere distribution is prOportional to blood flow rates and extraction by the pulmonary and systemic capillary beds is virtually complete. Also, the quantity of macro- aggregated albumin (5A) and albumin (A3) microspheres required to alter circulation significantly is far higher than the quantity used in tracer studies. 11 The assumption that micrOSpheres are removed only by mechanical trapping has not been proved for the systemic circulation, but has for the pulmonary circulation (A3, 51). However, an increase in activity has been noted in areas of trauma to veins after injection of microspheres into the femoral veins of dogs (63), suggesting that they could be removed by adhesion to the vessel walls. The microsphere technique has been tested against direct blood flow measuring techniques and the ability to reproduce data. Rudolf §t_al. (AA) used 50 u microspheres to measure organ blood flow in dogs and found excellent correlation between head and brain flow measured directly by flowmeters and that calculated from the nuclide distribution. Tschetter §t_al. (57) studied the cerebral blood flow of lightly anes- thetized dogs using radioactive microspheres labeled with 169Yb (25 i 5 microns in dia.) and cardiac output by inject- “2K or 86Rb. The blood ing intravenously a known activity of flow of cerebral cortex, white matter, thalamus, cerebellum, and caudate nucleus was determined in ml/min'gm and compared to values reported by others (33, A6). The flow values, when compared, showed no significant difference. In 1962, Ingvar and Lassen (27), using the 85Kr-clearance technique, found the average blood flow in the sensory cortex to be 65 ml/min'lOO gm in nembutalized dogs. Harper (2A) found a mean cortical blood flow of 99 ml/min'lOO gm (standardized to A0 mm Hg P002) in dogs anesthetized with nembutal using the same technique. He also found no significant difference between occipital cortex, frontal cortex, and parietal cortex. 12 The values obtained via the same 85Kr-clearance technique are, in general, higher than values found using microspheres. Also, there is no general agreement on the interpretation of the 85Kr-clearance curves or on the physiological meaning of the components of the curves. Another problem is that in calculating tissue flow from the clearance curves, it must be assumed the white and grey tissue are homogenous. Haggendal §t_a1. (22) found the ratio of grey to white matter flows in the dog brain to be 5.2:1, which agrees with the microsphere data of Roth g§_al. (A5). If several perfusion rates are represented in a section of cortex, using only the first exponential component of the whole section and ignoring the other exponential components of the slower rates, the entire flow would be calculated too high. There are several technical disadvantages to the clear- ance technique. The volume of tissue whose radioactivity is measured is not really known and extra-cerebral tissue which receives part of the injectate will be included in the flow measurement. Also, artifacts may result from the posi- tioning of the probe. Roth g§_al, (A5) ran experiments to see if varying quantities of micrOSpheres changes brain circulation. They varied the doses of microspheres to unanesthetized dogs. One injection was 0.35 ml of microsphere suspension (8.0 x 105 microspheres) and the other was 0.1 ml of suspended micro- spheres (2.3 x 105 microspheres). The reaction to the latter 13 injection was mild and the determined cerebral blood flows were not significantly different. The application of labeled microspheres to the study of shock requires some special consideration. First of all is the hemodynamic effect of the microspheres. These microspheres have been shown to have no hemodynamic effect in normal dogs (28) if the number injected is not over 106. However, in sick dogs Kaihara §t_al. found the microspheres decreased cardiac output and coronary blood flow (unpublished observations). But, Kaihara gt_al. (29) found that neither the amount of hemorrhage required to lower pressure to A0 mm Hg nor the subsequent course of hemorrhagic shock in the dog was affected by microsphere administration. Any effect the microspheres may have on the hemodynamic state of some dogs, would occur after the distribution of the microspheres and thus the measurement of regional blood flow during shock is considered valid. III. Brain Circulation The cerebral circulation can be divided into two gen- eral categories, superficial or conducting arteries and the penetrating or nutrient arterioles (18). Cerebral arteries can also be classified into several orders on the basis of size (1, 30), the largest (7th order, 2-3 mm), beginning with the internal carotids and vertebral arteries. The con- ducting arteries also include the Circle of Willis (6th order, 2-1.5 mm) and its branches over the cerebral cortex (5th 1A order, 1.5-100 u). Fifth order arteries have anastomoses and are considered flood gates of the cerebral circulation. Nutrient arterioles (Ath order, 51-16 A) originate from large and small pial arteries, and penetrate the cortical surface to supply six layers of the cortical mantle and white matter. These narrow in diameter (3rd order, 25-12 H) until the precapillary segment is reached (2nd order, 12-8 u). They have anastomoses only when connections are made with penetrating arterioles (7). There are approximately 1,000 mm of capillaries/mm3 (lst order, 6 n) which nourish the 3 cortical grey matter and about 300 mm of capillaries/mm which nourish the water matter (9). METHODS Twenty-six mongrel dogs of both sexes weighing 8.2 i 3 kg were anesthetized with sodium-pentobarbital (30 mg/kg). An endotracheal tube was inserted and animals were respirated with a positive pressure respirator. Left ventricular, arterial and venous pressure recordings were made using Statham pressure transducers (Model P23Gb) connected to a Sanborn direct—writing recorder (Model 60-1300). Regional cerebral blood flows were calculated by the radioactive labeled, particle distribution technique (63). Polyethylene Cannulae (PE2A0) were placed into the right femoral artery (for the measurement of systemic arterial pressure), into the aorta through the left femoral artery (for withdrawal of reference blood after each microsphere injection), into the left femoral vein (for the infusion of endotoxin and drug administration), and into the left ven- tricle via the left carotid artery (for microsphere injection and measurement of heart rate and ventricular pressure). Purified E. coli endotoxin (Difco), 2 mg/kg, was diluted in normal saline to yield a 20 ml suspension. Follow- ing the cannulations and first microsphere injection, the endotoxin was administered over an approximate five minute period at a rate of 3.88 ml/minute. 15 l6 Microsphere Preparation 1A1 0.1: — 0.5 ml of 85 Ce and Sr microsphere (15 i 5 u; 3M Company) were withdrawn from 10 ml vial using a 1 ml syringe, and thoroughly mixed in a glass tube with 3 ml of 20% dextran using an ultrasonic mixer. Approximately 2 x 105 microspheres were injected at 0.1 mC/ml. The suspension was then drawn from the glass tube into a 3 m1 syringe (mak- ing sure no air was left in the syringe) and then injected into the left ventricle as a bolus. Counting Tubes Each microsphere injection was flushed with 3-5 ml of saline to make sure all microspheres reached the left ven- tricle. At the same time a three minute reference blood sample was withdrawn (3.88 ml/minute) from the left femoral artery using a Harvard withdrawal pump. The reference blood sample was then divided into 12 gamma counting tubes (5 ml capacity) at approximately 1 ml/tube. The tubes were placed into a Searle (Model 1185) gamma counter and counted with a lAlC base of 1A0 and a window of 30 for e, and a base of A80 and window of 80 for 85Sr. Brain Samples After the animal was sacrificed, the skin and muscle layers of the head were cut from the skull. The brain was exposed by removing the t0p half of the skull with a Stryker saw, the attachment to the spinal cord was carefully cut, and the entire brain was removed. Duplicate tissue samples, 17 approximately one gram each, were taken from the medulla, pons, thalamus, hypothalamus, cerebellum, cortex, and pitui- tary, placed into plastic gamma counting tubes, and weighed. The samples were located and removed by pre-set boundaries found on the brain. Calculations Regional cerebral blood flow (rCBF) was calculated by dividing counts per minute (cpm) per gram of brain tissue by counter per minute of the three minute reference blood sample and multiplying by the withdrawal rate of the refer- ence blood sample (RBWR, a constant) (10). cpm/gm brain cpm ref. bld. rCBF = x RBWR This calculation was made for each tissue sample taken. The two blood flows calculated from the two samples of each area were averaged into one regional blood flow; regional resistance was calculated by dividing the arterial pressure (Pa) by the regional blood flow. rR = Pa/rCBF The animals were divided into three groups: Group I, Controls (n=6). These animals were infused with 20 mlcfl‘saline, which contained no endotoxin, over a five minute period following the first microsphere injection. This was followed by a four-hour period in which systemic arterial blood pressure and heart rate was recorded every 30 minutes. 18 After this four-hour period, the second microsphere injec— tion was given, the animal sacrificed with sodium pentobar— bital, and brain samples taken. Group II, Two Hour Shock (n=10). Two mg/kg endotoxin were infused intravenously over a five minute period following the first microsphere injection. Heart rate and blood pressure were subsequently recorded every 30 minutes for four hours. The second microsphere injection was given two hours later, the animal was then sacrificed and brain tissue removed. Group III, Four Hour Shock (n=10). The animals received 2 mg/kg of endotoxin over a five minute period following the first microsphere injection. Heart rate and blood pressure were subsequently recorded every 30 minutes for four hours. At this time the second microsphere injection was given. The animal was then sacrificed and brain tissue taken. Statistical Analysis The regional blood flows, cerebral vascular resistances, and arterial pressures were analyzed using the Student's "t" test modified for paired replicates. A "p" value less than 0.05 was considered significant. RESULTS Control Groung Average values for arterial blood pressure, regional cerebral blood flows and vascular resistances before and four hours after saline infusion are shown in Figure l. The open bars denote values four hours after saline infusion. In all regions of the brain sampled (pons, medulla, thalamus, hypothalamus, cerebellum, cortex, and pituitary) the mea- sured parameters did not change significantly (p > 0.05) over the four hour observation period. It is again empha- sized that arterial pressure is aortic pressure, and is always shown as the same value for each brain region. GrouijI (2 Hour Shock) Average values for arterial blood pressure, regional cerebral blood flows and vascular resistances before and two hours after endotoxin infusion are shown in Figure II. The solid bars denote values before the 2 mg/kg of endotoxin and the open bars denote values two hours after the infusion. In all regions of the brain sampled the average blood flow two hours after shock induction was signifiCantly lower than pre-shock values (P < 0.05). The percent decrease in blood flows in the regions sampled were: pons, 31.9%; medulla, l9 20 .mcfiamw mo coaumpumficfisnm msoco>mnucfi nouns mason snow cam whommn pop amazomm> Hmnnmpmo on» no mmmcoommp anmcmooEmn owmhm>< .H mpswfim 21 H 3.1 E .88 too .22: Sf .82 .1 DVHmN_NHNO_ 6.2 .828 . I C; S .828 aooe >o n. 22 .cHxOpopcm Haoo AM no wx\we m mo COHQMApchHEpm msocm>wmch hwpmm mason 03p paw mLOan own amazomm> Hmmnmnoo map mo momCOQmoh oHEmchoEms owmhm>< .m opswfim 23 HH .mfi .928 :8 .22: .oef .82 + 4 EN 59%. D 9.2 .058... I x85 ExSoucm .LI N + 2A 28.9%; hypothalamus, 23.0%; thalamus, 36.0%; cortex, 32.A%; cerebellum, 30.7%; and pituitary, 38.9%. Cerebral vascular resistance was decreased significantly after two hours of endotoxin shock in the regions of the pons, hypothalamus, and medulla (P < 0.05) by 36.5%, 39.A%, and 38.A%, respectively, but was not significantly different from control in the thalamus, cortex, cerebellum, and pituitary (P > 0.05) with decreases of 2A.5%, 19.1%, 12.8%, and 2A.l%, respectively. Average systemic arterial blood pressure was significantly lower after two hours of endotoxin shock (P < 0.05) by 53.A%. Group III (A Hour Shock) Average values for arterial blood pressure, regional cerebral blood flow and vascular resistances before and four hours after endotoxin infusion are shown in Figure III. The solid bars denote values before endotoxin and the open bars denote values four hours after endotoxin infusion. In all regions of the brain sampled average blood flows four hours after the infusion of endotoxin were decreased significantly (P < 0.05). The percent decrease in blood flows of the re- gions sampled were: pons, 38.8%; medulla, A3.0%; hypothala- mus, 19.2%, thalamus, 50.5%; cortex, 50.6%; cerebellum, 50.2%; and pituitary,51.8%. Cerebral vascular resistances were decreased significantly after four hours of endotoxin shock in the pons, medulla nad hypothalamus (P < 0.05) by 68.2%, 18.2%, and 32.A%, respectively but was not significantly different from control values in the thalamus, cortex, 25 .QHxOpoocm HHoo AM no wx\wE m go coapmhpmficfispw msocm>mnpcfi pmpmw mason hsom 6cm whommn cop amazomm> Hapnmmoo map uo mmmcoamop oHEmcmooEmn mwmnm>< .m mhswfim 26 HQ .3“. may: E 8.8 :8 99: .8»: 8.2 eon. oIEE < d 80_\c_E\_E\oIEE >o m O ._ ooo_\c_e\_e . >0 , Om .H_ NEHMS .Ev £82m u D 072 .2250 .... I 0.85 5.38291? 27 cerebellum, and pituitary (P > 0.05) showing decreases of A.5%, 5.6%, 9.2%, and 15.6%, respectively. Average systemic blood pressure was significantly lower after four hours of endotoxin shock (P < 0.05) by 62.7%. DISCUSSION The current study provides information describing regional cerebral blood flows under normal conditions as well as during endotoxin shock in the dog. By pooling flow and resistance values obtained at the first microsphere injection (control values) of all three series of dogs (N=26), we calculated the following average values which are presented in ml/min/lOO g brain (see Table I for individual values): pons = 15.0; medulla = 20.8; hypothalamus = 18.3; thalamus = 2A.l; cortex = 2A.9; cerebellum = 29.7; pituitary = 120.2. E. coli endotoxin shock produced decreases in aortic pressure, (SO-60%) and cerebral blood flows (30-50%) in all areas of the brain sampled (pons, medulla, hypothalamus, thalamus, cortex, cerebellum, pituitary) at both two and four hours of endotoxin shock. Cerebral vascular resistance (CVR) was decreased significantly (approximately 35%) in the pons, medulla and hypothalamus at two and four hours of shock but did not change significantly in the thalamus, cortex, cerebellum, and pituitary. The four hour shock exper- iments produced slightly greater changes in all parameters measured. Investigators have reported a decrease in cerebral blood flow during endotoxin shock in several species. Parker and Emerson (35, 13, 1A), using the Rapela—Green tecnique, 28 29 found at four hours post-endotoxin infusion of l, 2, and 5 mg/kg E. coli the cerebral blood flows were reduced by 63, 52 and 55%, respectively. They also found that cerebral vascular resistance initially fell, but by A hours of shock was significantly higher than control. Wyler §t_al. (68) using monkeys, found that there was a 30% decrease in cerebral blood flow, but the blood flow in the cerebellum and brain stem was maintained after one hour of endotoxin shock. Weiner (67) administered endotoxin I.V. (10 mg/kg; LD88) into rhesus monkeys and found that cerebral blood flow measured by 131 Cs uptake had decreased from 35.2 to 7 ml/min after 3 hours. This is a significantly greater decrease in cerebral blood flows than found by Parker (35). The magni- tude of decrease in regional cerebral blood flows in the present study agrees with the decreases found by Parker (35) when administering 2 or 5 mg/kg endotoxin systemically. Few studies are available in which blood flows were calculated for several discrete areas of the brain, under normal conditions or during shock. Roth gt_al. (A5) used 50 u microspheres labeled with 85Sr, 51Cr, and lulCe to calculate blood flow from some regions of the brain. He used unanesthetized dogs and found the cerebellum blood flow to be 3A.99 ml/min/lOO g and brainstem blood flow to be 20.77 ml/min/lOO g. In the present study cerebellum blood flow was found to be 29.7 ml/min/lOO g and medullary blood flow to be 20.8 ml/min/lOO g. They also found the right temporal lobe (this is approximately where the cortex sample was taken in 30 the present study) to have a flow of 2A.65 ml/min/lOO g, and in the present study, the cerebral cortex blood flow was 2A.9 ml/min/lOO g. Somecfi‘the blood flow values in the pre- sent study are a little lower than the values of Roth §§_§;, This is probably due to the use of anesthesia in the present study. With anesthesia, all animals show changes in myocar- dial function, tissue metabolism and output of catecholamines, that can act together with complex effects on neurogenic vasomotor control which mayeffect the blood flow (33). Neutze §t_al. (3A), using 50 u microspheres, found a decrease in cerebral vascular resistance, as with the heart, and a slight decrease in total cerebral blood flow in hemorrhagic shock. Green and Rapela (19) measured cerebral venous outflow in dogs during a stepwise decrease in sys- temic pressure. After two hours of hypovolemic hypotension the cerebral blood flow remained near control level due to the decrease in cerebral vascular resistance. This was also found to be true in the subsequent hemorrhagic shock phase. It seems that the differences seen in the response of the cerebral vasculature in hemorrhagic and endotoxin shock suggests there may be a difference in the involvement of the cerebral vasculature in the pathogenesis of the two types of shock. Blood flow to the brain, as in other organs, depends on perfusion pressure (arterial pressure minus venous pres- sure) and vascular resistance to blood flow. In this study, the decrease in blood flow in all regions occurred with a 31 decrease in systemic arterial pressure, and a decrease or no change in regional vascular resistances. Thus, it can be concluded that, on the average, the decrease in blood flow in all regions sampled was due to the decrease in cerebral perfusion pressure. There are also important alterations in the regional cerebral vascular resistance at two and four hours of endo- toxin shock. The cerebral vascular resistance of the pons, medulla and hypothalamus, decreased significantly while the cerebral vascular resistance of the thalamus, cortex, cerebellum, and pituitary did not change significantly from the control values. This decrease in the pons, medulla, and hypothalamus occurred concomitantly with the decrease in the systemic arterial pressure. It is interesting to compare these results to earlier studies from this lab that measured total cerebral vascular resistance in dogs (35, 13, 1A). Emerson and Parker used the Green-Rapela technique which measured blood flow from approximately 50-67% of the brain (cortex and thalamic regions). They found that cerebral vascular resistance decreased significantly during the first hour of endotoxin shock, increased to a level no different from control, between one and three hours, and increased above control by the fourth hour of shock. The data at two hours of endotoxin shock, in the present study, agrees with these earlier findings, but, there is no evidence in the current study of any regional cerebral vascular resistances increasing. There is no controversy here, however, because 32 the regions that exhibited resistance decreases inthe present study (pons, medulla, hypothalamus) were not included in the earlier studies. In the present study, the decreased perfusion pressure resulted in an active increase in vessel caliber because the decrease in systemic arterial pressure would lower transmural pressure and produce a passive decrease in vessle caliber. This decrease in cerebral vascular resis- tance of the pons, medulla, and hypothalamus is possibly due to a cerebral autoregulatory phenomenon, because other authors have reported that endotoxin has no direct action upon the cerebral vasculature (6, 67). The pons, medulla and hypothalamic regions appear more able to adjust their vascular resistance than other areas of the brain, possibly because of their importance to the survival of the animal in endotoxin shock. These three regions, acting as Integrating-Control centers, are used in the transmission of impulses to control many of the body functions. The hypothalamus is involved in regulation of temperature, certain autonomic functions, and certain metabo- lic processes. The pons and medulla control many different autonomic functions. Thus, to a great extent the lower brain stem centers are Integrating-relay stations for control activities integrated at higher levels of the brain (21). But even with the decrease in cerebral vascular resis- tance in the pons, medulla, and hypothalamus, the blood flows were significantly lower after two and four hours of endo- toxin shock. Also, the cerebral vascular resistances did not 33 change significantly in the thalamus, cortex, cerebellum and pituitary, even with a severe decrease in blood flow in these regions. However, the decrease in regional blood flow is not as great as is found in other organs. Hinshaw gt_a1. (25) has shown that after giving dogs a lethal dose of endotoxin the renal plasma flow decreases 80-90%. It has also been shown that in endotoxin shock splenic and hepatic blood flows are decreased to almost 100% and mesenteric blood flow is decreased over 80% (52). Coronary blood flow was found to decrease approximately 30% when endotoxin was injected intravenously to dogs (50) and the total forelimb blood flow has been found to decrease 85% in endotoxin shock (65). While the decrease in blood flow is great in most regions of the brain, it is not as great of a decrease as found in other organs. The difference must be due to resistance differ- ences, because all of the organs are subjected to the same pressure gradient. In the present study there was not an increase in cerebral vascular resistance like that seen in other organs of the dog. This may be due to the brain not being as sensitive to sympathetic discharge, circulating vasoconstrictive agents or possibly the fact that the brain can autoregulate under normal conditions. This study has shown that there is an extreme decrease in blood flow in all regions of the brain sampled and that this decrease in blood flow is due to a decrease in cerebral perfusion pressure. It is felt that, with these areas 3A ischemic and needing nutrients, the function of these regions may be compromised, and this may contribute to the pathogenesis of endotoxin shock. BIBLIOGRAPHY 10. ll. BIBLIOGRAPHY Alexander, L.; Putman, T. J.: Pathological alterations of cerebral vascular patterns, Assoc. Res. Nerv. and Ment. Dis 18:A71, 1938. Anderson, J. F., and Rosenau, M. J.: Anaphylaxis, A.M.A. Arch. Int. Med. 3:519, 1909. Bostroem, B., and Schoedel, W.:" Uber die Durchblutung der arterioverosen Amastomosen in der hinteren extremitat des hundes, Pflueger Arch Ges. Physiol. Bradfeld, P. A.; Lopex-Majano, V.; and Wagner, H. N.: Measurement of the regional distribution of arter- ial blood flow in the human forearm and hand, J. Nuc. Med. 8:5A2, 1967. Brown, R. S.; Mohn, P. A.; and Shoemaker, W. 0.: Effects of cerebral hypotension on the neural regulation of the cardiovascular system, Surg. Gynecol. Obstet. 131:A36, 1970. Buyniski, J. P.; Smith, M.; and Bierwagen, M. E.: Effects of gram-negative endotoxin on cerebral circulation, Europ. Neurol. 6:253-258, 1971/1972. Campbell, A. C. P.: The vascular architecture of the cat's brain, Assoc. Res., Nerv. Ment. Dis. 18:69, 1938. Charrin, A. and Gley, E.: Recherches experimentales sur l'action des productis secretes par 1e Bacille pyocyanique. Arch. de Physiol. 3. 5, 2: 72A. In: J. Lab. Clin. Med. A8:661, 1956. Cobb, S., and Talbott, J. H.: Quantitative study of the cerebral capillaries, A. Amer. Physicians. A2:225, 1927. Raymond, R.: Cerebral metabolism during endotoxin shock, Fed. Proc. 35:79A, 1976. Dobbins, D. E.: Centrally mediated cardiovascular ef- fects of systemically administered E. coli endo- toxin, Ph.D. Dissertation, Michigan State Uni., 1975. 35 12. 13. 1A. l5. 16. 17. 18. 19. 20. 21. 22. 36 Dubezanski, V.; and Nauyn, B.: Beitrage zue Lehre von der fiedehaften durch pyrogeno substanzen bewirk- ten temperatureerhohung. Arch. Exper. Path. u. Pharmakol. 1:181, 1872. Emerson, T. E.; Jr., and Parker, J. L.: Cerebral hemodynamics and endotoxin shock in the dog, Circ. Shock, in press, 1972. Emerson, T. E., and Parker, J. L.: Cerebral hemodyna- mics during endotoxin shock in the dog, Physiolo- gist 15:125, 1972. Emerson, T. E., Jr.: Participation of endogenous vasoactive agents in the pathogenesis of endotoxin shock. In: The Fundamental Mechanisms of Shock, Edited by L. B. Hinshaw and B. G. Cox. New York, Plenum Press, 1972. Emmennegger, H.; Hurlomann, A.; and Bucher, K.: A simple method of producing radioactive spheres for the investigation of circulation problems. Helv. Physiol. Acta. 9:25A, 195A. Gallo, E., and Schenk, W. G.: Endotoxin shock: Alterations in cardiac output and regional blood flow with modifications resulting from whole blood transfusion with and without ganglionic blockage, J. Trauma 7:667, 1967. Gillilan, L. A.: General principles of the arterial blood vessel patterns to the brain. Trans Amer. Neurol. Assoc. 82:65, 1957. Rapela, C. E., and H. D. Green: Autoregulation of canine cerebral blood flow. Circ. Res. 1A & 15: 1-205, 196A. Grothenhuis, I. M.: Properties and uses of radiating microspheres. In: Radioactive Pharmaceuticals, G. A. Andrews, R. M. Kniseley, H. N. Wanger, Jr., and E. B. Anderson, eds. CONE-651111, United States Atomic Energy Commission, Division of Technical Information, Springfield, Virginia, 1966: p. 205. ‘ Guyton, A. C.:' Textbook of Medical Physiology. Phila., W. B. Saunders Co., 1971. Haggendal, E.; Nilsson, J. J.; and Norback, 8.: Effects of blood corpuscle concentration oncere- bral blood flow, Acta. Chir. Scand. Suppl. 36A: 13, 1966. 23. 2A. 25. 26. 27. 28. 29. 30. 31. 32. 33. 37 Halpern, B. N.; Biozzi, G.; Benacerrat, B.; Stiffel, C.; and Hillmand, B.: Cinetique de la phagocytose d'une serumalbumine humaine specialment traitee et radiomarzuee, et son application 'al' etude de la circulation bepatique chez l'Homme, C. R. Soc. B121. 150:1307, 1956. In: J. Nuc. Med. 6:150, 19 5. Harper, A. M.; Glass, H. 1.; and Glover, M. M.: Measurement of blood flow in th cerebral cortex of dogs by the clearance of Kr' 5, Scot. Med. J. 6:12-17, 1961. Hinshaw, L. B.; Spink, W. W.; Vick, J. A.; Mallet, E. and Finstad, J.: Effect of endotoxin on kidney function and renal hemodynamics in the dog, Am. J. Physiol. 201:1AA, 1961. deer, R.; Orgis, E.; and Pfeiffer, G.: Splenoporto- scientigraphy to demonstrate prosystemic shunts. J. Nuc. Med. 9:62A, 1968. Ingvar, D. H. and Lassen, N. A.: Regional blood flow of the cerebral cortex determined by Krypton 85, Acta. Physiol. Scand. 5A:325, 1962. Kaihara, S.; vanHeerden, P. D.; Migita, T.; and Wagner, H. N.: Measurement of distribution of cardiac output, J. Appl. Physiol. 25:696, 1968. Kaihara, S.; Rutherford, R. B.; Schwentker, E. P.; and Wanger, H. N.: Distribution of cardiac output in experimental hemorrhagic shock in dogs, J. Appl. Physiol. 27:218, 1969. Kennedy, J. C.: Shunting in cerebral microcirculation, Am. Surg. 35:763, 1967. MacLean, L. D.; Kim, Y. S.; and Henderstrom: Distribu- tion of B. F. in canine and human hearts. Surg. Forum 12:221, 1961. Moss, G.; Staunton, C.; and Stein, A. A.: Cerebral etiology of the "shock lung" syndrome, J. Trauma. 12:885-890, 1972. Neutze, J. M.; Wyler, F.; and Rudolph, A. M.: Use of radioactive microspheres to assess distribution of cardiac output in rabbits. Am. J. Physiol. 215:A86, 1968. 3A. 36. 37. 38. 39. A0. A1. A2. A3. 38 Neutze, J. M.; Wyler, F.; and Rudolph, A. M.: Changes in distribution of cardiac output after hemorrhage in rabbits. Amer. J. Physiol. 215:857, 1968. Parker, J. L.: Cerebral hemodynamic and metabolic alterations during endotoxin shock in the dog. Ph.D. Dissertation, Michigan State University, 1975. Parker, B. M.; Anderson, 0. C.; and Smith, J. R.: Observations on arteriovenous communications in lungs of dogs. Proc. Soc. Exp. Biol. Med. 98: 306, 1958. ' Pearce, R. M.; and Eisenbrey, A. B.: The physiology of anaphylactic "shock" in the dog, J. Infect. Dis. 7:565, 1910. In: J. Lab. Clin. Med. 661, 1956. Penner, A.; and Klein, S. H.: The pathogenesis of experimental dysentery intoxication, J. Exptl. Med. 96:59-69, 1952. Phibbs, R. H.; Wyler, F.; and Neutze, J.: Rheology of microspheres injected into circulation of rabbits. Nature 216:1339, 1967. Piiper, J.; and Schoedel, W.; Untersuchungen uber die Durchhutung der arteriovenosen anastomosen in der hinteren extremitat des hindes mit Hilfe von Jugelm verschieldener Grobe, Pfluger. Arch. Ges. Physiol. 258:A89, 1953. In: J. Nuc. Med. 6:150, 19 5. Prinzmetal, M.; Simkin, B.; Gergman, J. C.; and Kruger, H. E.: Studies on the coronary circulation, II. The collateral circulation of the normal human heart by coronary perfusion with radioactive erythrocytes and glass spheres, Amer. Heart J. 33: A20, 19A7. Cox, J. W.; Boehler, R. W.; Sharma, T.; O'dorisis, T.; Osgood, R. W.; Stein, J. H.; and Ferris, T. F.: Studies on the mechanism of oliguria in a model of unilateral acute renal failure. J. Clin. Invest. 53:1516, 197A. Rhodes, B. A.; Zolle, I.; Buchanan, J. W.; and Wagner, H. N.: Radioactive albumin microspheres for stu- dies on the pulmonary circulation. Radiology 92: 1A53, 1969. an. 45. A6. A7. A8. A9. 50. 51. 52. 53. 39 Ring, G. C.; Blum, A. S.; Karbatox, T.; Mass, W. G.; and Smith, W.: Size of microspheres passing through pulmonary circuit in the dog. Amer. J. Physiol. 200:1191, 1961. Roth, J. A.; Greenfield, A. J.; Kaihara, S.; and Wagner, H. N.: Total and regional cerebral blood flow in unanesthetized dogs, Am. J. Physiol. 219:96, 1970. Radulph, A. W.; and Heymann, M. A.: Methods of study— ing distribution of blood flow, cardiac output and organ blood flow. Circ. Res. 21:163, 1967. Simmons, R. L.; Ducker, T. B.; Martin, A. M.; Anderson, R. W.; and Noyes, H. E.: The role of the central nervous system in septic shock. I. Pathologic changes following intraventricular and intracis- ternal endotoxin in the dog. Ann. Surg. 167: 1A5-157, 1968. Simmons, R. L.; Anderson, R. W.; Ducker, T. B.; Slee— man, H. K.; Collins, J. A.; and Boothman, K. P.: The role of the central nervous system in septic shock. II. Hemodynamics, respiratory and meta- bolic effects of intracisternal intraventricular endotoxin. Ann. Surg. 167:1A5-156, 1968. Shibata, H. H.; and MacLean, L. D.: Blood flow in tumors. (In progress in Clinical Medicine, I. M. Ariel, ed., 1966, p. 33). Spitzer, J. J.; Spitzer, J. A.; Weng, J. T. and Scott, J. B.: Myocardial metabolism during acute hemorr- hagic or endotoxic shock, In: The Fundamental Mechanics of Shock, edited by L. B. Hinshaw and B. G. Cox. New York, Plenum Press, 1972. Strauss, H. W.; Hurley, P. J.; Rhodes, B. A.; and Wagner, H. N.: Quantification of right to left transpulmonary sunts in man, J. Lab. Clin. Med., 1969. Swan, K. G.; Barton, R. W. and Reynolds, D. G.: Splanchnic blood flow in experimental shock, In: The Fundamental Mechanisms of Shock, edited by L. B. Hinshaw and B. G. Cox. New York, Plenum Press, 1972. Tamura, H.; Witoszka, M. W.; Hopkins, R. W.; and Simeone, F. A.: The nervous system in experimental hemorr- hagic shock: Morphology of the brain, J. of Trauma 12:869, 1972. 5A. 55. 56. 57. 58. 59. 60. 61. 62. 63. A0 Taplin, G. V.; Johnson, D. E.; Kennedy, J. C.; Dore, E. K.; Poe, N. D.; Swanson, L. A.; and Greenberg, A.: Aggregated albumin labeled with various radio- isotopes. In: Radioactive Pharmaceuticals. G. A. Andrews, R. N. Kinsely, H. N. Wagner, and P. B. Anderson, eds. CONF-651111, United States Atomic Energy Commission, Division of Technical Informa- tion, Springfield, Virginia, 1966. Taplin, G. V.; Johnson, D. E.; Kennedy, J. C.; Dore, E. K.; and Kaplan, H. S.: Suspensions of radio- albumin aggregates for photoscanning the liver, spleen, lung and other organs. J. Nuc. Med. 5:259, 1954. Trippodo, N. C.; Jorgensen, J. H.; Priano, L. L.; and Traker, D. L.: Cerebrospinal fluid levels of endotoxin during endotoxemia. Proc. Soc. Exptl. Biol. Med. 1A3:932, 1973. Tschetter, T. H.; Klassen, A. C.; and Resch, J. R.: Regional cerebral blood flow in dogs using a particle-distribution method. Proc. Sco. EXptl. Biol. Med. 131:12AA-12A9, 1969. Ueds, H. S.; Kitani; Kameda, H.; Yamada, H.; and Iio, J.: Detection of portal systemic shunts by hepa- topulmonary scanning using I131 MAA: Study and applicagion of macroaggregated albumin labeled with I (III). Jap. Heart J. 6:115, 1965. Ueda, H. S.; Kitani, K.; Kameda, H.; Yamada, H.; and Iio, M : Regional myocardium blood flow measured by 1131 Macroaggregated albumin. Gastroenterology 52:A80, 1967. Ueda, H.; Kaihara, S.; Ueda, K.; Sugishita, Y.; Sasaki, Y.; and Iio, Y31 Regional myocardium blood flow measured by 1 labeled macroaggregated albumin (1 31-MAA), Jap. Heart J. 6:53A, 1965. Visscher, M. B.: Introduction: an overview of the shock problem. In: The Fundamental Mechanisms of Shock, Edited by L. B. Hinshaw and B. G. Cox. New York, Plenum Press, 1972. Wagner, H. N.; Sabiston, D. C.; Iio, M.; McAfee, J. G.; Mayer, J. K.; and Langer, J. K.: Regional pulmo- nary blood flow in man by radioisotOpe scanning, JAMA 187:601, 196A. Wagner, H. N.; Rhodes, B. A.; Sasaki, Y.; and Ryan, J. P.: Studies of the circulation with radioactive microspheres. Invest. Rad. 1:37A, Nov—Dec., 1969. 6A. 65. 66. 67. 68. 69. A1 Wagner, H. N.; Jones, E.; Tow, D. E.; and Langer, J. K.: A method for the study of the peripheral circula- tion in man. J. Nuc. Med. 6:150, 1965. Weidner, W. J.: Transvascular fluid movement and seg- mental vascular resistances in response to trans- fusion in endotoxin shock, Ph.D. Dissertation, Michigan State University, 1975. Weil, M. H.; MacLean, L. D.; Spink, W. W.; and Visscher, M. B.: Investigations on the role of the central nervous system in shock produced by endotoxin from gram—negative micro-organisms. J. Lab. Clin. Med. A8:661, 1956. Weiner, D.: Effects of endotoxin on cerebral blood flow in the monkey. Am. J. Physiol. 218:160, 1970. Wyler, F.; Forsyth, R. P.; Nies, A. S.; Neutze, J. M.; and Melmon, K. L.: Endotoxin-induced regional circulatory changes in the unanesthetized monkey. Circ. Res. 2A:777, 1969. Ya, P. M.; Guzman, M. T.; Loken, M. D.; and Perry, J. F.: IsotOpe localization with tagged micro- spheres. Surgery A9:6AA, 1961. APPENDIX A2 55.0 0.0mH mm.m 0.mm :0.5 m.mH 05.0 0.5H 00.5 0.0H :0.0 m.5H 00.HH H.0H QNH 0H 5:.0 0.m0m 0:.0 :.5H 00.0 :.:H HH.0H 5.00 00.5 0.HH 05.0 0.:H m0.5 0.mH 00 :H :5.0 0.H:H 00.H m.m0 m0.m 0.mm 0m.m 5.5: H:.m 0.0m :0.m 0.:m 0m.0 0.0a 00H ma 0:.0 0.55H 00.m m.0m M5.: 0.0a 00.: 5.0a 00.0 :.:H 0a.: H.0H mm.m m.mH 00 NH :0.m H.0: 0m.: m.mm Hm.0 H.0H 00.: 0.0m 00.0 0.0a 0m.0 0.0H 5m.0 H.NH 00H Ha 0m.H m.0:H mm.m :.0: mm.m 0.Hm m0.m m.0m 5:.: 0.:m mw.m 0.0m 00.m 0.0m OHH 0H 0:.H 5.:0 H0.0 5.MH 00.0H :.50 00.0 0.0a 00.HH m.HH m5.5 m.0H ::.MH m.00 mma 0 m0.0 0.0ma :m.: 0.mm 5m.: :.mm :0.: N.0m 00.0 m.mH mm.0 0.0a 5m.0 5.0H 00H 0 00.H 0.00 m0.m 0.0m 50.: m.mm 00.0 H.0H N5.5 0.MH 0:.0 :.0H H0.0 N.NH 00H 5 00.H m.00 :m.: 0.0m 05.0 0.0a 0:.0 0.5a m0.HH 0.00 00.HH 0.00 m5.HH 0.00 mad 0 mm.H :.00 :5.: m.mm mm.0 :.0H mm.m 5.Hm m0.:H H.00 5H.0H 0.HH om.mH 0.00 oma m 05.0 0.HNH 00.: :.ON 5H.0 :.ma 05.0 0.:H H0.ma m.50 00.0 0.00 50.0H 0.00 00 : 00.H 0.mmH m5.0 0.0a 0m.0 m.ma 0:.0 m.ma 0m.5H N.50 5:.HH 0.0a 00.0H 5.00 mma m 0m.m m.mm 00.0 5.0H m:.5 0.:H :0.5 :.:H 50.:H 0.50 0H.0 m.ma mm.ma H.00 OHH m :0.0 0.HOH 0H.: 0.5m mm.m 0.NN H0.5 :.0H 50.5 m.mH 00.0 0.mm mm.5 0.0a mad H m .m m m «H 0 MM 0 Mm m Mm 0 MM 0 .pfizpfim .Hawnmhmo sophoo demHmnB .Suoazm maazvo: mcom mm A0mucv .Aw 00H\:fip\dghwmcbc mmocmpwfimmm 0cm Aw 00H\:HE\HEV mzofim Hoapcoo .H mqmH 0150.0 A7 303.03 33 0 .0 300.0 030...“ 530.0 000.0 030..... 000...0 030.03 030...“ 030.0 000...3 330.0 030.0 005. .H 353. 050. 503. 030. 003. 000. 003.. 053. 033. 300. 300. 333. 0003 30003 0000 mm0. 5:.H 00H. 00H. 0:H. mom. NHH. 00H. :mH. 05H. 0HH. :5H. M00. m00. 00 omH 0H 0m.H 00.H 00m. 00m. 0:H. 00H. 00H. 05H. :00. 00H. H5H. 05H. 000. HOH. 00 omH 0 HMH. mo.m 000. :5H. 5H0. ::H. mmo. 500. 5:0. 0HH. H00. 0:H. 000. 0mH. 0m 00 0 0H0. H:.H 00H. mm0. HmH. 0N0. ::H. 55:. HOH. 000. 00H. 0:0. 0HH. 00H. 0m 00H 5 005. 055.H 05H. mom. 0:H. 00H. HMH. 50H. 5MH. ::H. H:H. HOH. :NH. 00H. om 00 0 0::. H0:. 00H. 0mm. :0H. HOH. HOH. 00m. H0m. 00H. mmH. 00H. 000. HNH. 0: 00H 0 m00.H 50:.H 000. 30:. HOH. 0H0. 0HH. mom. 0:H. 0:m. H5m. 00m. 0:H. 00m. 05 0HH : 000. 5:0. 0:m. 5MH. mmH. :50. 0Hm. 00H. 000. 0HH. 0NH. NOH. N00. M00. 00 0NH m 5N5. 00m.H 0Hm. 0mm. 50H. :mm. 50H. mom. 00H. NOH. 00H. 00H. 50H. 50H. 00 00H N 500. 000. m0H. 00m. 000. 0mm. H50. HOH. NOH. 0MH. 550. :0H. 000. NNH. 0: 00H H 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 03330 3308.300 000.300 000303000 000000 0333002 0000 00 0 .0030 30300 30033535 3030 I 00000 .3000 03.3 .> 333000 M8 00.00 03.00 30.00 50.00 00.0“ 00.0“ 33.0“ 00.0“ 53.0+ 00.3+ 0000 53.30. 0.00 0.0“ .u .u .n 00.0 30.0 00.0 00.0 00.3 00.3 00.0 00.0 50.0 00.5 30.3 05.0 33.0 03.0 3+m3 3+m33 mm +.m 05.3 55.0 35.0 30.3 03.3 00.3 03.0 03.3 00.0 00.5 00.0 00.0 50.0 00.0 03 003 03 03.0 33.0 05.0 03.0 03.0 00.0 00.0 50.5 50.0 05.0 53.0 05.03 00.0 00.03 00 033 0 53.0 30.0 03.0 00.0 05.0 00.5 53.0 50.0 30.0 05.0 00.0 00.03 03.0 50.03 mm 003 0 35.0 00.3 00.3 03.0 00.3 50.3 30.0 00.3 00.0 50.0 00.0 00.0 35.0 00.5 50 033 5 03.0 00.0 03.0 03.0 50.0 00.0 50.0 00.0 00.0 03.3 53.0 00.0 00.3 50.0 03 003 0 00.3 00.0 30.0 53.3 30.0 33.3 03.0 03.3 00.0 33.3 00.0 00.3 05.0 03.3 00 003 0 00.0 53.0 00.0 03.3 30.0 03.0 03.0 00.3 00.0 00.0 03.0 55.3 03.0 50.0 00 00 3 53.0 30.3 00.0 00.0 00.0 03.3 30.0 35.0 00.3 00.0 00.0 00.0 05.5 50.0 00 003 0 50.0 00.0 00.0 03.3 00.0 30.0 00.0 05.5 00.0 00.0 00.0 00.03 00.3 30.03 00 003 m 03.0 05.0 30.0 00.0 00.3 50.33 00.0 33.03 00.0 00.03 30.0 00.0 00.0 00.03 00 003 3 3030 30 303.0 30 3 030 .030030 .33000000 000000 005030g0 .000000 0330002 0000 00 0 .000 303n00 .300 003xcugxus0w00=0 00000003000 .g. 00000 0000 0000 .3>.mqm¢e 149 003.“ 0530.03 000.“ 050.3 330.0 050.0 000.33. 350.0 500...“ 030.“ 000....3 350.0 000.“ 000.0“ I .. I 030. 03.3 033. 550. 033. 300. 033. 000. 033. 530. 003. 300. 000. 003. 3+m3 3+m33 3.30.00 Hmm. m0.H 0:H. H5m. 500. HHm. 0NH. 0mm. mmH. :0H. 50H. 5mm. 00H. 00H. 0: 0mH 0H 0mm. 0:0. 0:H. 0mm. HOH. 0Hm. 00H. 00H. 50H. 0HH. 000. 50H. 500. 000. 00 0HH 0 H5.H :m.m 000. 00m. H00. 00H. 000. 00H. 000. 50H. 000. 0HH. 000. N50. 00 0mH 0 Hm0. 000. 00m. 00m. mom. m:m. 0NH. 0mm. m0H. 00H. HNH. 0NH. 00H. 00H. 50 0HH 5 0m0. :N.H HmH. mm:. 00H. 50:. 00H. 0:0. 00H. 5mm. mmH. 000. 00H. 05H. 0: 00H 0 0:H. 00.m 000. H00. 000. 000. 000. 000. 00H. 0N0. 050. 505. 000. 505. 0m 00H 0 00:. H0.H :mH. 050. 50H. 00m. HOH. H::. 000. 050. 00H. 05:. 000. 000. 0m 00 : mom. H0.H :00. :00. H00. m0m. 50H. 0HN. 00H.. 00H. :00. NHN. 550. 5NH. 00 0mH m 00.H 0H.H 00H. :mm. 0:H. 00H. :mH. 00H. HNH. 0mH. 00H. 500. 000. 000. 00 00H m 0H.H 05.H 05H. 00H. 00H. 00H. 0:H. 000. N00. 5HH. N50. 00H. 000. N00. 00 00H H 3 o 3 o 3 o 3 o 3 o 3 o 3 o 3 0 03330 33000000 000.000 02:03.05 000930 0330002 0000 00 0 .me 303.05 .3§\:E\35 203.0 I. 030000 .3000 .3000 30> 0.3033. 003.0 053.00 000.00 050.0 330...“ 050.0 000.0 350.0 500...... 030.0 000.0. 350.00 000.0 000.0 I I I 030. 03.3 033. 550. 033. 300. 033. 000. 033. 530. 003. 300. 000. 003. 3+03 3+033 mm+m ‘49 HON. N0.H 0:H. H5N. 500. HHO. 0NH. 000. NNH. :0H. 50H. 50N. 00H. 00H. 0: 0NH 0H 00N. 0:0. 0:H. 0NN. HOH. 0HN. 00H. 00H. 50H. 0HH. 000. 50H. 500. 000. 00 0HH 0 H5.H :N.N 000. 00N. H00. 00H. 000. 00H. 000. 50H. 000. 0HH. 000. N50. 00 0NH 0 HNO. 000. 00N. 000. NON. N:N. 0NH. 0NN. NOH. 00H. HNH. 0NH. 00H. 00H. 50 0HH 5 0N0. :N.H HOH. N03. 00H. 50:. 00H. 0:0. 00H. 50N. 0NH. 000. 00H. 05H. 0: 00H 0 0:H. 00.N 000. H00. 000. 000. 000. 000. 00H. 0N0. 050. 505. 000. 505. 0N 00H 0 003. H0.H 3N3. 050. 503. 000. HOH. H33. 000. 050. 00H. 053. 000. 00N. mm 00 3 000. H0.H :00. :00. H00. 00N. 50H. 0HN. 00H.. 00H. :00. NHN. 550. 5NH. 00 0NH 0 00.H 0H.H 00H. :0N. 0:H. 00H. :NH. 00H. HNH. 0NH. 00H. 500. 000. 000. 00 00H N 0H.H 05.H 05H. 00H. 00H. 00H. 0:H. 000. N00. 5HH. N50. 00H. 000. N00. 00 0NH H 3 o 3 o 3 o 3 o 3 o 3 o 3 o 3 0 000030 33000.00 000.300 083030033. 00.0830 0330002 0000 00 0 .930. 30305 00003535 3030 .I 00000 .3000 .300 30> 0300.3. “‘"II’I‘I'IIII’IIIIIIILIII’I’LIIIIIIIII’I'IIIII'IIII‘II'“