' w w_ V- CEREBRAL HEMODYNAMIC AND METABOLIC ALTERATIONS DURING ENDOTOXIN SHOCK IN THE DOG Dissertation for the Degree of Ph. D. MICHIGAN STATE UNIVERSITY JANET LEA PARKER 1975 I .. . 5 LIBRARY mm 0' ~ a .# . This is to certify that the thesis entitled CEREBRAL HEMODYNAMIC AND METABOLIC ALTERATIONS DURING ENDOTOXIN SHOCK IN THE DOG presented by Janet Lea Parker has been accepted towards fulfillment of the requirements for Ph.D. degree in Physiology Mm? W, Major professor ‘ Date 6' Y-7; 0-7639 ? V BINDING BY “ HUM? & SUNS' 800K BINUEIIV INC. LIBRARY BINDERS mum". MICHIGAN |§\r “we... BJIII I CEREBRAL HEMODYNAMIC AND METABOLIC ALTERATIONS DURING ENDOTOXIN SHOCK IN THE DOG BY Janet Lea Parker Cerebral hemodynamics, vascular reactivity, and meta- bolic alterations were studied in spontaneously respiring dogs following intravenous administration of 1 mg/kg (n=5), 2 mg/kg (n=16), and 5 mg/kg (n=6) E. 99;; endotoxin. Cerebral venous outflow was measured directly from the can- nulated confluence of the sagittal, straight, and trans- verse (lateral) sinuses, with the transverse sinuses occluded. This method eliminates the major sources of extracranial vascular contamination of the cerebral venous outflow. All preparations were required to meet criteria of vascular separation and reactivity before considered acceptable for use. Cerebral blood flow and cerebral perfusion pressure decreased immediately upon administration of l, 2 or 5 mg/kg endotoxin, and remained below control values for the entire four hour experimental period. By the fourth hour of shock cerebral blood flow had decreased 37, 48, and 45% of control in the l, 2 ' perfusion pre not equaling; Upon ad: resistance tr Slgnificantly :reased to its at least the tOXin Cer endc above control :93§ectinly . CEIEbral VaSC1 :ecreased per} VdSCular Janet L. Parker in the l, 2 and 5 mg/kg doses, respectively, and cerebral perfusion pressure had increased to levels approaching (but not equaling) control values. Upon administration of endotoxin, cerebral vascular resistance transiently decreased, returned to levels not significantly different from control, and gradually in- creased to levels which were significantly above control by at least the fourth hour of shock. At 240 minutes post- endotoxin cerebral vascular resistance was 25, 55 and 37% above control in the l, 2 and 5 mg/kg doses of endotoxin, respectively. Thus, cerebral blood flow decreased follow— ing endotoxin administration, due initially to the decreased perfusion pressure and subsequently, to the increased cerebral vascular resistance and, to a minor extent, to the decreased perfusion pressure. Vascular reactivity studies indicated that the cere- bral autoregulatory and "venous—arteriolar" responses were well maintained after four hours of endotoxin shock. The cerebral vascular responsiveness to inhalation of 7.5% carbon dioxide, however, was depressed after shock. Prior to endotoxin administration, cerebral vascular resistance decreased to 37% of control in response to the CO2 chal- lenge. After 1, 2, 3 and 4 hours post-endotoxin, resist- ance decreased to only 81, 80, 80 and 75% of control, respectively. I!" Endotox respiratory a": s d P _, e rea e C' the hyperven tiOll, as ind‘ and acid‘base ISagittal Sin Strated inCre Saline c cant alterati or arterial a Janet L. Parker Endotoxin administration generally caused an increased respiratory activity, resulting in increased P02 and decreased PCOZ in the systemic arterial blood. Initially, the hyperventilation did not produce an adequate compensa— tion, as indicated by the early acidosis in the 2 mg/kg and 5 mg/kg doses. Cerebral venous blood pH decreased, and blood gas changes included decreased P02 and increased arterial-venous differences of percent oxygen saturation, total C02, and HCO3’ concentrations. Thus, brain tissue and acid-base balance, reflected in cerebral venous (sagittal sinus) pH and bicarbonate concentrations, demon— strated increased acidosis during endotoxin shock. Saline control experiments (n=8) yielded no signifi— cant alterations in cerebral blood flow, vascular reactivity, or arterial and venous blood gases. in p; CEREBRAL HEMODYNAMIC AND METABOLIC ALTERATIONS DURING ENDOTOXIN SHOCK IN THE DOG BY Janet Lea Parker A DISSERTATION Submitted to Michigan State University in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY Department of Physiology 1975 Tc unfaili father , memo ry DEDICATION To my husband, Paul B. Parker, for his unfailing support and encouragement, and to my father, F. Glenn Smith, Jr.. and the loving memory of my mother, Helen. ii The anti research guic‘ Flierson, Jr. , aer doctoral :‘cierson, Ruc' ful sUrgical Young is alsc The auth late Mr. TOm SChOoll and t state Univers ACKNOWLEDGEMENTS The author gratefully acknowledges the academic and research guidance from her major professor, Dr. Thomas E. Emerson, Jr., and the assistance from the other members of her doctoral committee: Drs. Jerry B. Scott, Donald K. Anderson, Rudy A. Bernard and W. Doyne Collings. The skill- ful surgical and technical assistance of Mr. Robert J. Young is also appreciated. The author wishes to extend special recognition to the late Mr. Tom Talley, former teacher, Thomas Jefferson High School, and to Dr. David R. Redden, Professor, North Texas State University, for their early encouragement of her career . iii fin! hm.“ v.23“ ILR I INT: II son. C F S T E E I. III' METH< Ca P1 IV- Resm He Al Ce V' DISco ' SUlmA A tati: A CHAPTER TABLE OF CONTENTS I . INTRODUCTION 0 O O O I O O O O O O O O C O O O 0 II. SURVEY OF THE LITERATURE. . . . . . . . . . . . Clinical Background of Septic Shock. . . . . Research Models of Septic Shock. . . . . . Systemic Circulatory Effects of Endotoxin Shock . . . . . . . . . . . . . . . . . . Transvascular Fluid Fluxes During Endotoxin Shock . . . . . . . . . . . . . . . . . . Effect of Endotoxin Shock Upon the Myocard- 1um . . . . . . . . . . . . . . . . . . . Effects of Endotoxin on Peripheral Vascular Hemodynamics. . . . . . . . . . . . . . . Involvement of the Central Nervous System and Cerebral Vascular Bed in Endotoxin Shock . . . . . . . . . . . . . . . . . . III 0 METHODS O O C O O O O O O O O O O O O O O O O 0 IV. RES V. DIS Cannulations . . . . . . . . . . . . . . . . Preparation of cerebral circuit. . . . . . . ULTS O O O O O O O O O O O O O O O O O O O 0 Hemodynamic Alterations. . . . . . . . . . . Alterations in Cerebral Vascular Reactivity. Cerebral Metabolic Changes . . . . . . . . . CUSS ION I O O O O O O O O O O O O O O O O O 0 VI. SUMMARY AND CONCLUSIONS . . . . . . . . . . . . BIBLIOGRAP APPENDICES HY O O O O O O O O O O O O O O I O O O O O O A. Statistical Analysis. . . . . . . . . . . . . . B 0 Raw Data 0 O O O O O O O O O O C O O O I O O O 0 iv Page 10 12 14 22 22 23 33 33 55 7O 83 99 102 114 117 LIST OF FIGURES FIGURE Page 1. Response of cerebral venous pressure (P ) to transient occlusion of the outflow tubigg dur- ing unblocked, partially blocked and completely blocked lateral sinuses. . . . . . . . . . . . . 26 2. Average hemodynamic responses of the cerebral vascular bed following intravenous administra- tion of 2 mg/kg E. coli endotoxin. . . . . . . . 35 3. Representative tracing from an experiment following intravenous administration of 2 mg/kg E. coli endotoxin. . . . . . . . . . . . . . . . 38 4. Average hemodynamic responses of the cerebral vascular bed following intravenous administra- tion of 5 mg/kg E. coli endotoxin. . . . . . . . 4O 5. Representative tracing from an experiment terminating in death of the animal due to intra— venous administration of 5 mg/kg E. coli endo- toxin. . . . . . . . . . . . . . . . . . . . . . 43 6. Average hemodynamic responses of the cerebral vascular bed following intravenous administra— tion of 1 mg/kg E, coli endotoxin. . . . . . . . 46 7. Average hemodynamic responses of the cerebral vascular bed during control experiments. . . . . 48 8. Average percent change of cerebral blood flow during control experiments (dashed line) and in animals receiving 1, 2 and 5 mg/kg E, coli endotoxin (solid lines). . . . . . . . . . . . . 51 9. Average percent change of cerebral vascular resistance during control experiments (dashed line) and in animals receiving 1, 2 and 5 mg/kg endotoxin (solid lines). . . . . . . . . . . . . 54 FIGURE 1m 12. 13. 14. 15. 16 17. 18 19 Avera: ceivi: for 5‘ Averat ance level hours 2-20 Avera ance befor admir Avere resi: diox; trat Aver resi diOx adgu Aver Sy$t ing AVe: Sys- ing AVe §Ys lug AVE Ce: and 2 n AVQ 3y: CO] FIGURE 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. Average hemodynamic responses of animals re- ceiving 2 mg/kg E. coli endotoxin and followed for six hours . . . . . . . . . . . . . . . . . Average responses of cerebral vascular resist- ance (A) and cerebral blood flow (B) to graded levels of hemorrhage before and after four hours of shock due to administration of 2 mg/kg E. coli endotoxin . . . . . . . . . . . . . . . Average response of cerebral vascular resist- ance to increases in cerebral venous pressure before and after four hours of shock due to administration of 2 mg/kg E. coli endotoxin . . Average percent change of cerebral vascular resistance due to inhalation of 7.5% carbon dioxide prior to and four hours after adminis- tration of 2 mg/kg E. coli endotoxin. . . . . . Average percent change of cerebral vascular resistance due to inhalation of 7.5% carbon dioxide prior to and at hourly intervals after administration of 2 mg/kg E. coli endotoxin . . Average responses of respiratory rate, and systemic arterial and venous blood gases follow- ing administration of 2 mg/kg E, coli endotoxin Average responses of respiratory rate, and systemic arterial and venous blood gases follow- ing administration of 5 mg/kg E. coli endotoxin Average responses of respiratory rate, and systemic arterial and venous blood gases follow- ing administration of 1 mg/kg E. coli endotoxin Average responses of systemic arterial and cerebra1_venous oxygen saturation, total CO and HCO3 content following administration of 2 mg/kg E. coli endotoxin . . . . . . . . . . . Average responses of respiratory rate, and systemic arterial and venous blood gases during control experiments . . . . . . . . . . . . . . vi Page 57 59 62 65 68 72 74 76 79 81 The investiga negative 319:! (l7) duce Shot: disPlayed Tested the ’eCently. fie Carotj spinal flu t0 elicit fliniCal 8. studies Cm Have Shown centrally-n ShOck SiOnl an CHAPTER I ' INTRODUCTION The central nervous system has been implicated by many investigators in the pathogenesis of irreversible gram- negative endotoxin shock. As early as 1890, Charrin and Clay (1?) administered crude pyrocyaneus endotoxin to pro- duce shock in rabbits. They reported that shocked animals displayed depressed responses to neural stimulation and sug- gested that this was due to central "paralysis". More recently, the introduction of small doses of endotoxin into the carotid artery or into the ventricular or cerebro- Spinal fluid systems of dogs (24,90,91,ll4,115) was shown to elicit a shock syndrome which bears many resemblances to clinical septic shock. In addition, cross-circulation studies completed in isolated head-trunk preparations (24) have shown that endotoxin is capable of eliciting a marked centrally-mediated hypotensive response. Shock is often described as "inadequate tissue perfu- sion” and the central nervous system is most sensitive to such deprivation (85). In this regard, few studies have concentrated on possible cerebral vascular involvement in the reSPO- (T Moss .3. induce t?“ interstiti These autl' caused by mothalam been repor acteristic inadequate factor in t measurement is of value The hy W W the response of the central nervous system to endotoxin. Moss 2£.§L- (85) demonstrated that cerebral hypoxemia can induce the "shock lung syndrome", characterized by pulmonary interstitial and intra-alveolar edema and hemorrhage. These authors believed that these symptoms of shock were caused by autonomically mediated venous spasm, mediated by hypothalamic hypoxia. In addition, cerebral hypotension has been reported (14) to produce hemodynamic alterations char- acteristic of shock. Thus, cerebral ischemia subsequent to inadequate blood flow during shock may be a contributory factor in the pathogenesis of the shock state, and the measurement of cerebral blood flow and vascular resistance is of value in evaluating this theory. The hypothesis tested in this study was: Do altera- tions in cerebral hemodynamics, vascular reactivity and cerebral metabolism (arterial and cerebral venous blood gases) indicative of cerebral ischemia occur during endo- toxin shock in the dog? Clin; secondary intensive research, ranse of 5 CaIISatiVe HEgativE t most gram- are CauSed OrganiSm i siflCe the S CHAPTER II SURVEY OF THE LITERATURE Clinical Background of Septic Shock Clinically, the pathogenesis and treatment of shock secondary to sepsis remains a difficult problem, and despite intensive research and improved treatments based on this research, the mortality rates reported are usually in the range of 50 to 90% (27,53,109,120). The most frequent causative agents of septic shock are gram-positive and gram— negative bacteria, although due to antibiotic control of most gram-positive agents, most infections initiating shock are caused by gram-negative organisms. In addition, an organism is presented with gram-negative bacteria more often since these are normal inhabitants of the gastrointestinal tract (92). Altemeir 32 31. (2) reported that the most frequent source of gram-negative infections is the genitourinary tract. Approximately half of the patients presenting gram— negative sepsis showed a history of recent urinary tract operations or instrumentation. Irritations and infections in the respiratory tract (following tracheostomy), repro- ductive tract (due to septic abortion) and gastrointestinal tract (peri of gram-me; (diabetes I crease susc Bactel manifestat: prostation‘ tional syn; I3) tachyp; with Perip': I6) Oligur; have used 1 list“ abox ated with ( Warm Shoe) tract (peritonitis) have also been indicated as primary foci of gram-negative infections, and predisposing conditions (diabetes mellitis, cirrhosis, patient age) may also in- crease susceptibility (109). Bacteremia or septicemia usually begins with clinical manifestations of chills, fever, vomiting, diarrhea and prostation. When the septicemia proceeds to shock, addi- tional symptoms occur: (1) hypotension, (2) tachycardia, (3) tachypnea, (4) cool, pale extremeties, often associated with peripheral cyanosis, (5) mental obtundation, and (6) oliguria (92,111). Several authors of clinical reports have used the term “cold shock", typified by the symptoms listed above, as the more advanced stage of shock associ- ated with catecholamine release (16,110), and the term "warm shock" to describe an earlier stage of shock associ- ated with hypotension and a hyperdynamic circulation (16,80, 110). At this time the skin is dry and pink, extremities warm and urine output usually adequate (16,40,80,110). Although metabolic acidosis is generally considered to be the classic acid-base alteration in shock (80,124), it is usually a late development, and the typical early response of patients in septic shock is respiratory alkalosis (80,92, 124). Progressive pulmonary insufficiency is often reported (5,8,109,113), resulting in deterioration of pulmonary function and severe hypoxia. A myriad of other clinical abnormalities have been described, including coagulation defects, dec nitrogen, or tory data v; Shoema}; shock is a compensatio. 0f compensat drive to mee Greased“, Rssearc‘n MO \ tinal tract Bacterj~Ode 5996a: t0 defects, decreased platelet counts and elevated blood urea nitrogen, creatinine, and lactate (92), although the labora- tory data varies greatly. Shoemaker (111) reported that the natural history of shock is a progression from compensatory responses to de- compensation, the shock syndrome thus resulting from failure of compensatory mechanisms to "provide adequate circulatory drive to meet physiologic needs, be they normal or in- creased”. Research Models of Septic Shock As previously indicated, common organisms causing sepsis are similar to those found in the human gastrointes- tinal tract and include: (1) Escherichia coli, (2) Klebsi- 2112 aerobactor, and (3) Proteus, Pseudomonas, and Bacteriodes species, in decreasing frequency. The abnormal- ities of the shock state are incompletely understood, but appear to be initiated by endotoxins from the cell walls of the gram-negative bacteria listed above (109). An intra— venous injection or infusion of the purified endotoxin of any one of these bacteria into otherwise healthy animals produces signs and symptoms similar to those seen in human septic shock. The purified endotoxin is a lipopolysacchar- ide protein complex, which is obtained by extracting bacteria, commonly E, coli, by various procedures (23). The Boivin trichloracetic acid procedure is a commonly used technique a: achieve a d' the clinics an endotoxi to L090 is , Many a. tories in t Il8-20,28-3 Sheep (7), 96,107,118 the [ESan of the ani A notable ma'T‘cifestm versible : tilted by . technique and yields endotoxin of high potency (23). To achieve a degree of shock corresponding to that reported in the clinical conditions mentioned in the previous section, an endotoxin dosage in the research model yielding an LD 50 to LD is commonly used. 90 Many animal species have been used by various labora- tories in the experimental shock model, including dogs (18-20,28-34,54-57,93,122), cats (50,71), rabbits (86,102), sheep (7), calves (99), mice (127), and primates (59,61,66, 96,107,118). However, clear-cut differences may exist in the responses of the human species of septic shock and those of the animals used in the experimental model (84,106). A notable example is the dog. In this animal, one of the manifestations of what is termed by some authors as irre- versible shock is mesenteric congestion, perhaps precipi- tated by constriction of hepatic sphincters (18-20,42,106). Such congestion, however, is not seen in the primate intes- tine in shock (66,74,96,106). Although direct extrapolation of particular findings to human septic shock may be diffi- cult, as in the area mentioned above, the use of the experi- mental animal model has contributed greatly to an overall understanding of the shock state (109). Systemic Circulatory Effects of Endotoxin Shock Administration of intravenous E. coli endotoxin in anesthetized dogs is immediately followed by a period of rapid system temporary pa and a later 30,67,122) . causes a dec ates followe decrease in in total p91 output (48) in arterial Uttput and eral resiSt @1leng E (33). The: Mean arte: Went rgr rapid systemic hypotension within 5 minutes, succeeded by a temporary partial recovery during the next 20-60 minutes, and a later progressive decline in blood pressure (18,19,42, 50,67,122). In baboons (59,66,96) endotoxin administration causes a decrease in blood pressure over the first 30 min- utes followed by a gradual progressive decline. The initial decrease in arterial pressure must be due either to a fall in total peripheral resistance or to a decline in cardiac output (48). Most authors have reported that the decrease in arterial pressure is accompanied by a decrease in cardiac output and either no change or an increase in total periph- eral resistance (18,63,122). Similar results were obtained following endotoxin administration in unanesthetized dogs (93). These animals demonstrated the classic decrease in mean arterial pressure, accompanied by diminished cardiac output, greatly elevated total peripheral resistance, increased heart rate and decreased stroke volume. Several authors have attributed the initial decrease in cardiac output in the dog to a decrease in venous return (10,11,42,122). Cardiac output and venous return are also diminished in the baboon (59,66) although the time sequence varies in comparison to the dog. A primary decrease in venous return is also supported by recent studies by Hinshaw gg_§1, (55,56) which indicate the absence of a direct toxic action of endotoxin upon the myocardium, at least in the early hypotensive stage of shock. Thus, initially, the de- crease in arterial pressure is due to a decrease in cardiac output subsequent to decreased venous return to the heart. The direct causes for the decreased venous return apparently vary according to the species being considered. In the monkey (59,66), the decrease in venous return occurs slowly, and the sites of venous pooling have not been specifically identified (27,59,66). However, in the dog, Gilbert (42) reported that hepatic venoconstriction with consequent portal hypertension and hepatic venous pooling of blood was the cause of the decreased venous return. On the other hand, Blattberg and Levy (10,11) studied endo- toxin shock in dogs in which the portal venous pressure was prevented from rising via a Servo-pump, or the liver was actually bypassed, and found that significant pooling of blood still occurred. This group also performed ligations of the aorta at various levels and correlated the resulting fall in blood pressure after endotoxin administration with the amount of blood pooled at the various ligations. They concluded that in the early post-endotoxin period, approxi- mately equal volumes of pooled blood are found in and out of the hepatosplanchnic area (11). In summary, the systemic circulatory response to endotoxin in the dog is typified by pooling of blood in the hepatosplanchnic (and other) regions, thus decreasing cardiac output, and decreasing “‘13.: arterial pr cardiac mec systemic hy TIEUSVESC‘JI :_..otox1n E Theore to decreaSI fluid. En! 'ith net t decteases 351191“; i influx I12 fluxes am tratiOn he bEd Of all reported beds, 0ft A arterial pressure. Thus, peripheral rather than direct cardiac mechanisms are primarily responsible for the systemic hypotension of shock, at least in the early stages. Transvascular Fluid Fluxes During Efidotoxin Shock Theoretically, venous return could also decrease due to decreases in blood volume via transcapillary loss of fluid. Endotoxin shock has been shown to be associated with net transvascular fluid flux changes. Initially, decreases in hematocrit, viscosity, and plasma protein, and a slight increase in plasma volume indicate a net fluid influx (18-20). The subsequent direction of the fluid fluxes during shock, however, is controversial. Fluid fil- tration has been shown to occur in the hepatosplanchnic bed of the dog (20,58). In addition, several authors have reported evidence for fluid filtration in other vascular beds, often based upon measurements of thoracic lymph flow, decreases in plasma volume, and increases in hematocrit (19,20). However, when the effects of splanchnic constric- tion are abolished via splenectomy, hematocrit remains unaltered and plasma volume increases (18,19). In addition, the validity of the dye dilution techniques for determina- tion of plasma volume in endotoxin shocked animals may be questionable, due to slower, uneven mixing and sequestra- tion of blood during shock (68). In this regard, measure- ments of blood and plasma volumes in the dog may be of limited sig mique pool tare. If net does occur. been defint reabsorpti. Presented alteration Skeletal n Owens (64) were assoc “Sting tl 10 limited significance when applied to primates, due to the unique pooling of blood in canine hepatosplanchnic vascula- ture. If net fluid filtration in extra-splanchnic vasculature does occur, the specific location of this efflux has not been defined. Indeed, convincing evidence for net fluid reabsorption throughout the entire period of shock has been presented (64,121), based upon organ weight and resistance alterations in several vascular beds, including skin and skeletal muscle. Weidner 33 El. (121) and Hinshaw and Owens (64) reported that canine forelimb weight decreases were associated with decreasing vascular resistance, sug- gesting that the weight loss was due to extravascular fluid reabsorption. Limb weight losses in the cat and monkey (58) during endotoxin shock also support this concept. Thus, blood volume decreases subsequent to transcapillary fluid loss cannot explain the decrease in venous return in the dog or primate. Effect of Endotoxin Shock Upon the Myocardium In the intermediate and preterminal stages of shock, the question of a direct effect of endotoxin shock upon the myocardium and the involvement of the failing myocardium in the development of irreversibility is controversial. Most authors agree that the heart will ultimately fail in shock (1,12,13,5‘ mechanism c reported t1 suggested coronary p- lated, WOI‘ trate fun first 3-4 conSiStent PEIIOG p05 a direct c failure 0c animals I V splanchni‘ t0 shOCk ‘ Hinshaw e Clearl: "PQar to 5 «action animals 11 (1,12,13,54-57,75-77), although reports vary as to the mechanism of the myocardial dysfunction. Alican SE.E$° (1) reported that heart failure was seen late in shock, and suggested that this was the result of prolonged inadequate coronary perfusion. Hinshaw et_al. (55,56) used an iso- lated, working heart preparation and were unable to demon- strate functional alterations in the myocardium in the first 3-4 hours after endotoxin administration, but found consistent signs of myocardial failure in the 4-6 hour period post-endotoxin. Their results did not demonstrate a direct cardiotoxic action of endotoxin, and suggest that failure occurs only after prolonged hypotension. However, Brand and Lefer (13) reported the discovery of a myocardial depressant factor (MDF) obtained from the blood of shock animals, which was later reported as having cardiodepressant (negative inotropic) action (75,76). This group reported that plasma MDF activity is inversely related to the splanchnic blood flow in shock and that pancreatectomy prior to shock abolishes MDF production (77). In contrast, Hinshaw gt_al. (60) reported that although heart failure can be clearly demonstrated 4-6 hours following endotoxin admin- istration, the blood of endotoxin shocked animals does not appear to possess a toxic factor which impairs myocardial function. More recently, this group used pancreatectomized animals and found that the presence or absence of the pancreas bc failure fol esis of a p primary fac They Sugges diastolic f tion of myo Selective c proved card | EffeCtS of Visoul % 12 pancreas bore no relationship to the degree of myocardial failure following endotoxin (54), thus weakening the hypoth- esis of a pancreatic myocardial depressant factor as a primary factor causing the heart failure of endotoxin shock. They suggested thatinadequahecoronary perfusion and abnormal diastolic filling were the potent factors in the precipita- tion of myocardial failure (57). Elkins 33 a1. (26) used selective coronary artery perfusion during endotoxin shock and demonstrated that high coronary artery perfusion im- proved cardiac performance during shock, supporting the importance of adequate coronary artery perfusion, and the absence of a circulating MDF. Effects of Endotoxin on Peripheral Vascular Hemodynamics As previously indicated, the hepatosplanchnic vascular bed in dogs is greatly affected by systemic endotoxin admin- istration, manifested primarily by intravascular sequestra- tion of blood within the capacitance vessels (10,11,18,42, 106). In addition, Blattberg and Levy (10,11) presented evidence for considerable blood pooling in extra-hepato- splanchnic vascular beds of the body. Emerson gt_gl. (34) demonstrated that following intra- venous administration of a lethal dose of endotoxin, renal resistance in natural flow kidneys increased transiently, and then began to decline toward control levels by 30 minutes po decreased in proport results we drinistra flow decre :ained Sig Penal VaSC turned to Intra 51‘3““ to c‘ Md skelet Tracilis n hCurs fOl] transieht] In the iSC (2 or 5 mg and inCree in the 3k; 13 minutes post-endotoxin. Renal blood flow was markedly decreased for the two hour experimental period, apparently in proportion to the decreased perfusion pressure. Similar results were reported in the baboon (96) following systemic administration of coliform endotoxin. Renal arterial blood flow decreased 79% at three minutes post-endotoxin and re- mained significantly depressed for the next two hours. Renal vascular resistance peaked at three minutes and re— turned to near control values at 60 minutes post-endotoxin. Intravenous endotoxin administration (2 mg/kg) has been shown to decrease blood flow and increase resistance in skin and skeletal muscle vasculature (89,121). In the isolated gracilis muscle, blood flow remained depressed for four hours following endotoxin; vascular resistance increased transiently and then began to wane toward control levels. In the isolated canine forelimb (121), E. ggli endotoxin (2 or 5 mg/kg, intravenous) decreased forelimb blood flow and increased segmental vascular resistances, especially in the skin vascular bed. Local (intra-arterial) administration of endotoxin elicits varying responses, depending on the vascular bed being studied. Isolated intestinal loops and hepatic vas- culature demonstrate increased resistance when perfused with endotoxin (63). The splenic artery, however, dilates in response to intra-arterial injections of E. 921; endo— toxin (47). Arterial and venous resistances increase in the isolate occur when lack Of a d ture (62)- nary arteri (38). Howe lature, loc transient e lnvolvemen1 9 stem and in Endotox Sever the centre 1890, Char PYOCyaneus Ported the respohses They sugg. .Paralysi 14 the isolated pulmonary circulation, although this does not occur when whole blood is not the perfusate, indicating the lack of a direct effect of endotoxin upon pulmonary vascula- ture (62). Administration of endotoxin directly into coro- nary arteries of dogs decreases coronary vascular resistance (38). However, in the renal, hindlimb, and forelimb vascu- lature, local administration of endotoxin causes minimal transient effects (43,67). Involvement of the Central Nervous System and Cerebral Vascular Bed in Endotoxin Shock Several studies have reported possible involvement of the central nervous system in endotoxin shock. As early as 1890, Charrin and Gley (l7) administered intravenous crude pyocyaneus endotoxin to produce shock in rabbits, and re- ported that the shocked animals displayed depressed ' responses to vasopressor and vasodepressor nerve stimulation. They suggested that this was due to central (medullary) "paralysis". Penner and Klein (91) employed a cross circu- lation technique in which the cephalic blood flow of one animal was derived from the systemic circulation of another animal, and vice versa. Shigella toxin was injected into the latter animal, and visceral lesions, hemoconcentration, hyperglycemiaand other symptoms of shock appeared in the former dog whose head received blood containing the toxin. They suggested that the Shigella toxin acts directly on the hypothalamu changes res: was critize: siderable le preparation Weill gt a. rePOIted the the animal r they did no: nervous sys: Went cours I titted. Ptanner tIOn of Verf trifle resu ‘1 tract a: which they gested the Central m autOnOmi C studies We U141/15), WtGrin we 381/ fOund If [4de ac Ont J :57 15 hypothalamus of the brain to cause diffuse sympathomimetic changes resulting in the pathologic alterations. This study was critized by other workers (25) on the grounds that con~ siderable leakage of tagged materials from the isolated head preparation to the systemic circulation occurs. In addition, Weill 35 El. (122) repeated the work of Penner and Klein and reported that the pathologic changes were not limited to the animal whose brain was perfused with endotoxin. However, they did not rule out the possibility that the central nervous system may have an important influence on the subse- quent course and lethality of shock once it had been ini- tiated. Penner and Bernheim (90) reported that the administra- tion of very small doses of endotoxin into the third ven- tricle resulted in ulcerative lesions of the gastrointestin- al tract and other symptoms suggestive of shock. The doses which they used were ineffective systemically. They sug- gested that a prime site of systemic endotoxin action may be central nervous centers and that these centers might mediate autonomic activity leading to pathologic changes. These studies were confirmed by the experiments of Simmons et 31. (114,115), in which relatively small doses of E. 991i endotoxin were injected into the lateral ventricles in dogs. They found that hyperventilation occurred immediately, and cardiac output, heart rate, and arterial pressure decreased more gradua prior to de Recent larly isola supplied cc 0f the reci donor anima resulted in addition, a did other v‘ in the reciv' - I Circait. the. demervation PIEXES, In Similar debt again fell II endotoxin is tenSiVe res; HOWever at fth UPOn mt °0ntact Tri _ Pyodo 9t \ 553 337: detEr lstration. a 30:1 16 more gradually. Total peripheral resistance rose sharply prior to death of the animals. Recentlyn .Dobbins (24) used a neurally intact, vascu- larly isolated head-trunk preparation, in which a donor dog supplied constant-flow perfused arterial blood to the head of the recipient animal. Endotoxin administration to the donor animal or into the perfusion circuit to the head resulted in marked hypotension in the recipient trunk. In addition, although histamine, acetylcholine, epinephrine and other vasoactive drugs elicited blood pressure changes in the recipient trunk when administered into the perfusion circuit, these responses were abolished following bilateral denervation of the recipient dogs carotid sinus-body com- plexes. In contrast, when endotoxin was infused into a similar debuffered preparation, recipient systemic pressure again fell markedly. These studies support the concept that endotoxin is capable of eliciting centrally mediated hypo- tensive responses. However, in order for endotoxin to exert a direct effect upon the central nervous system, the molecule itself must contact the elements of the central nervous system. Trippodo gt El: (117), using the Limulus in_!it£g_endotoxin assay, determined that up to 2 hours after endotoxin admin- .istration, and even during the period of highest plasma concentration of endotoxin (5-30 min) no endotoxin was detected i cause of i does not c during the not exert nervous s; tainly doe causes the to exert : the time 1 is ROI in Shocf Sion" and such dEpr COnCentra the respo M03$ et a \ - three the 17 detected in cerebrospinal fluid. They theorized that be- 5 cause of its high molecular weight (10 to 106) endotoxin does not cross the blood brain barrier. This suggests that during the early stage of endotoxemia, the endotoxin does not exert its effects by acting directly on the central nervous system. This, however, is controversial, and cer- tainly does not exclude the possibilities that endotoxin causes the release of a substance or reacts with a substance to exert its effect upon the central nervous system, or that the time course of the reactions (or transport) of endotoxin is not in concert with the plasma endotoxin levels. Shock is often described as "inadequate tissue perfu- sion" and the central nervous system is most sensitive to such deprivation (85). In this regard, few studies have concentrated on possible cerebral vascular involvement in the response of the central nervous system to endotoxin. Moss 23 31. (85) demonstrated that cerebral hypoxia can in- duce the "shock lung syndrome", characterized by pulmonary interstitial and intraalveolar edema and hemorrhage. These authors believe that these symptoms of shock were caused by autonomically mediated pulmonary venous spasms, initiated by hypothalamic hypoxia. In addition, Brown SE.2£° (14) reported that cerebral hypotension produced hemodynamic changes characteristic of shock. Thus, cerebral ischemia subsequent to inadequate flow during shock may be a contribu- tory factor in the pathogenesis of the shock state, and the measurement is of value The ce late and s1 '0 Vs ' ‘_ . CCLEuIllCdt. have been ; Pick Princ the KEI-Y‘S tions of t 91185 to t COTmunicat For exampl the Of a] extracerek net than 1' 18 measurement of cerebral blood flow and vascular resistance is of value in testing this hypothesis. The cerebral vascular bed is a difficult one to iso- late and study, due to the extensive arterial and venous communications with extracerebral circuits. Indirect methods of measuring and calculating cerebral blood flow have been reported, and most involve applications of the Pick principle. Popular techniques use modifications of the Kety-Schmidt nitrous oxide method (69), but the limita- tions of this technique are especially vulnerable when ap- plied to the cerebral vasculature, due to the extensive communications and collaterals with extra-cerebral circuits. For example, the venous blood sample should be representa- tive of all parts of the brain and be uncontaminated by extracerebral blood. In man, this criterion is more easily met than in other animals, as the internal jugular venous bulb is situated so that only cerebral venous blood enters it. This, of course, is not the situation in the dog. In addition, it is difficult to be certain of even mixing of the indicator in the arterial blood, due to the number of arterial supplies entering the brain. Other authors have used radioactive indicators or dyes injected into one or both internal carotid arteries, but again, the limitations described above render the quantitative measurement of cerebral blood flow difficult, complicated, and uncertain (44). Thus, although the nontraumatic nature of most of these tech: ation, the; canine cert in the exp- Some cerebral b tures (ski uPERI neck have simpl Artery blo out teking mullicatiOD limitatiOn AbOVe, the the cereb: with the 1 This teChn Gwen (94) dorsal Sag technique Gallc CarOtid an Never, t of CErebra extraQEreb 5e 19 these techniques is certainly of value in the clinical situ- ation, they present certain limitations when applied to the canine cerebral vascular bed, and when accuracy is demanded in the experimental situation. Some authors have reported techniques of measuring cerebral blood flow in which all of the extracranial struc- tures (skin, muscle, etc.) were removed from the skull and upper neck region, a very traumatic procedure. Others (73) have simply tied the vertebral arteries and reported carotid artery blood flow as representing cerebral blood flow with- out taking into consideration the extensive arterial com- munications. In consideration of these factors, and the limitations of the other direct and indirect methods reported above, the method used in this study (direct collection of the cerebral venous outflow) appears to yield the best data with the least amount of trauma to the experimental animal. This technique has been previously described by Rapela and Green (94), and involves collection of the outflow of the dorsal saggital and straight sinuses. Details of this technique will be discussed further in this thesis. Gallo and Schenk (39) reported marked decreases in carotid and vertebral artery flows during endotoxin shock. However, these arterial flows are not truly representative of cerebral blood flow as they widely communicate with extracerebral vascular beds. Also these workers reported a decreased "cerebral" vascular resistance whereas in the monkey, W resistanc measured following cerebral longer th used arti and meC Control, fourth hr fliEICant Shock, of a Sin confirms third tr Frasent Ve vasculE ing em markeé 20 monkey, Weiner (123) reported an increased cerebral vascular resistance during entodoxin shock. Buyniski 35 El- (15) measured cerebral outflow directly after only one hour following endotoxin administration. They found an increased cerebral vascular conductance but did not follow this for longer than one hour. Recently, Emerson and Parker (30,31) used artificially respired dogs in a similar preparation, and found an initial decrease in cerebral blood to 63% of control, which decreased further to 39% of control by the fourth hour of shock. Cerebral vascular resistance was sig- nificantly higher than control values by the fourth hour of shock. Parker and Emerson (88) reported preliminary results of a similar study in spontaneously respiring dogs, which confirmed the increased vascular resistance occurring in the third to fourth hour of shock. Results of this study are presented in this thesis. Very few studies have reported on changes in cerebral vascular reactivity or cerebral metabolic alterations dur- ing endotoxin shock. Buyniski gt_al. (15) have reported a marked reduction in the cerebral vascular responsiveness to increases in arterial PCO2 levels, although this was studied for only one hour post-endotoxin. Mamo gt 21, (81) reported reduced reactivity of cerebral vessels to hypercarbia in 21 out of 27 patients with cerebrovascular disease. Clinically, several authors have suggested that an increase in cerebral P fol CO2 thus in; increase ing acut levels 0 The dEpends Vascular SPCI‘ses hit Ce] bral me. may the. tal gra 21 PCO following ischemia causes prior maximal dilation, and 2 thus impairs the reactivity of the cerebral vessels to increased arterial PCO during the hours immediately follow— 2 ing acute experimental ischemia (35,81,82). Cerebral venous levels of P€02 or other blood gases have not been measured. The function of the central nervous system during shock depends greatly upon its nutritional supply through its vascular bed. Thus, measurement of cerebral vascular re- sponses during shock is of value in testing the hypothesis that cerebral hemodynamics, vascular reactivity, and cere— bral metabolism are altered or impaired during shock, and may therefore play a role in the pathogenesis of experimen- tal gram—negative endotoxin shock. Mong: testhetiz tIac‘neal t IESPI-Ie s; SWQEry, I Sodium he; EXpeI-imem Tile SOdiu) thh‘YleI-XE CHAPTER III METHODS Mongrel dogs of both sexes weighing 20:2 kg were anesthetized with sodium pentobarbital (30 mg/kg). An endo- tracheal tube was inserted and the animals were allowed to reSpire spontaneously. Following completion of the required surgery, the animals received intravenous injections of sodium heparin (10,000 units) supplemented during the experiment at a rate of approximately 500 units per hour. The sodium pentobarbital was also supplemented throughout the experiment at a rate of about 50 mg per hour. All pres- sure recordings were made from appropriate pressure trans- ducers (Statham Laboratories, Model P23Gb) connected to a Sanborn polygraph (Sanborn Co., Model 60-1300). Blood gas determinations were made using a Radiometer blood gas system. Cannulations The right femoral artery was cannulated with poly— ethylene tubing (PE 240) for measurement of systemic arteri- al pressure. The right femoral vein was cannulated (PE 240) to allow replacement of the cerebral blood flow into the systemic circuit and for introduction of intravenously 22 aministei oftubing vein in t. taken and portion 0 initially ROI. in us aPPl‘Ox in". extendir Skull we Cleaned niche \; 23 administered drugs throughout the experiment. A closed loop of tubing (PE 320) connected the left femoral artery and vein in the animals in which arterial blood samples were taken and in animals which were bled for the hemorrhage portion of the experiment. This arterial-venous shunt was initially filled with normal saline and was clamped when not in use. Preparation of cerebral circuit A midline incision was made in the skin from a point approximately one inch posterior to the cisterna magna extending approximately to the level of the orbit. The skull was exposed and the muscle and connective tissue was cleaned from the dorsal and posterior surfaces. The tech- nique used to isolate the cerebral venous circulation has been previously described by Rapela and Green (44,94). The outflow from the sagittal and straight sinuses, which drain only cerebral structures, was collected by drilling a hole in the skull at the level of the confluence of the sinuses, and fitting a snug cannula into this Opening. Communica- tions of the confluence of the sinuses with the extracere- bral circulation of the lateral and sigmoid sinuses was abolished by drilling an additional hole on each side of the skull so as to intercept the lateral sinus and occlude it with softened bone wax injected into the sinus cavity. The amount of bone wax injected was approximately 0.1-0.2 ml on each side. F: the ce: region in the: pressux tubing. inserts to the tight, nigh ri or grea ‘35 being Vascula; illustre Separat; aPPl’Oac} the Out; tight S} 599 Of 1 loges' The p911 during E carbOn C during t c 0L heme: 24 Following blockage of the lateral (transverse) canals, the cerebral and extracerebral vascular circuits of this region are almost completely separate. This was verified in these experiments by observation of the venous outflow pressure response during transient clamping of the outflow tubing. Cerebral venous pressure was measured by a needle inserted into the outflow tubing near the skull and proximal to the area to be clamped. The pressure response in a tight, relatively communication-free circuit is a rapid, high rise; a rapid rise in the venous pressure to 40 mm Hg or greater was considered to represent the cerebral circuit as being free of any major connections with extracerebral vascular compartments (94). An example of this response is illustrated in Figure 1. In a theoretical case of perfect separation of the vascular circuits, venous pressure will approach arterial pressure during transient occlusion of the outflow tubing. Failure to achieve verification of a tight system was usually caused by either incomplete block- age of the transverse canals or abnormal cerebral anasto- moses, and these animals were not used in the experiments. The performance of the cerebral vascular bed was observed during graded levels of hemorrhage and during inhalation of carbon dioxide to test the degree of cerebral reactivity during these conditions in the preparations. Graded levels of hemorrhage were used to exhibit characteristic cerebral 25 ZO_WD._UUO \SOJthDO WDOZW> .mmmncflm HonoumH wmxooab mamuonEoo cam .coxooan >HHMHuHmm .pmxooan: mafiusp uneasy 30Hmuso onu mo cowmsaooo ucmflmcmuu ou A>omv whammmum msoco> Hmunmuoo mo omcommom .H ousmam 26 a musmflm .o.. n... .... .... .... .o.. .... ... u... ..»v ... . . .o o..: . . .. . .. a... c .- .... .... . . n», ... .u.‘ ... ... .... .. .... .-.. nun. .. too. .4. . .. ,. . . . n.-. .. V... ..-. .... t . . . .. H.-?t .. o... .... . . . . .... v-ao. .. c... .... . . ... ~..I .-.o a... .... .. .A .. .n»1..utvnu.. .... .. . . , . . ...vx....vo... .u. l. i... n... c... o... . , . . ... .u«. u... .. .. . .A u... .... c... ... .V . . o ...y . . ‘... a... u.. . . a... ... urn. 1 .... .. . . y . .. ...o .9.. .. .,.. . . .... ... .... ... . . ( . .. .u.. v... ‘..- .n. . . .. ‘ .... .uu; .... ..- .. . . .. l... ..., .... .e. . . . . .n A... .... .... .o,. . . . . ...q ... ..a. .... .. . .. .-.u ... u... ....A... . ‘ . .. ...a .:.c a... u... .... ‘ 0 » 1a.! .1». In. .n.. .o. . ,. .n-n .... ,... .nt.... . ‘. . . u... . .. .... a... . l I. . .‘ ...u 0-.. .... non. ..‘. .. .u‘v .ovo s... ...- u... ... . , . .A.. .n.t u... .... .nuc ... . . .. .... p... .v.'.... .... .... . .. .... _ou. o~.. ...-. .¢.. .... ... o ,- .... .c.- o... n.... .‘u. ...u . .. ... ‘cu. ... .... .04. a... .4L .u. .... u.. n..- I... {In I- so... 3... 03383 .3951 v9.3.3.5 ZO_mD._UUO. 301.50 mDOZm> vascular a used to de hypercarbi istics of ‘ Green and regulatory fusion pres art(”Y 0ch 60% 0f the licontamim Cereb: a Constant. dOg Via the reserVOir ‘ no fluid We a needle We measure-mam Probe exter tinal ' 3 t1 the polygre 27 vascular autoregulation, and inhalation of 7.5-10% CO2 was used to demonstrate the cerebral reactivity to systemic hypercarbia, as these are considered exceptional character— istics of the cerebral vascular bed (9,35,70,94). Also, Green and Rapela (44) have demonstrated that similar auto- regulatory responses are elicited by reducing cerebral per— fusion pressure either by hemorrhage or by graded carotid artery occlusion. Thus, with the procedure described, about 60% of the cerebral venous outflow was collected, relatively uncontaminated by extracerebral blood flow (94). Cerebral outflow from the tubing flowed by gravity into a constant-volume 30cc reservoir and was returned to the dog via the femoral vein. Initially, the constant—volume reservoir was filled with 25cc Dextran (60,000 MW) so that no fluid was lost by the animal in filling the reservoir. A needle was inserted into the cisterna magna to allow measurement of cerebral spinal fluid pressure. A thermal probe extension was attached to the exposed end of the animal's tracheal tube to allow respiratory recordings on the polygraph paper. Blood flows were measured with a graduated cylinder and stopwatch. A rectal thermometer was used to monitor body temperature. Pufified E. ggli_endotoxin (Difco) was diluted to the appropriate dosage in normal saline to yield 20cc suspension solution. Following a suitable control period, the endo- toxin was administered intravenously over a five minute m. 1.. period. P iv Grou I dose of 2 mg four hours w Primary grou this dose of blood flow, minus cerebr ance (cerebr cerebrospina ltd body tem tion, every minute inter In eight of ulOOd pH, PC PGriod and a lental perio (“Cl/l Plas calculated f at‘ure and the Ile font 1) 28 period. Five groups of animals were studied: Group I (N=12) consists of animals which received a dose of 2 mg/kg endotoxin and were subsequently followed for four hours without interruption. This group serves as the primary group used for studying the cerebral responses to this dose of endotoxin. Control values of the cerebral blood flow, cerebral perfusion pressure (aortic pressure minus cerebral venous pressure), cerebral vascular resist- ance (cerebral perfusion pressure/cerebral blood flow), cerebrospinal fluid pressure, respiratory rate, heart rate and body temperature were taken before endotoxin administra— tion, every 5 minutes for the first 15 minutes and at 30 minute intervals for the remainder of the four hour period. In eight of the twelve animals, arterial and cerebral venous blood pH, P and P values were taken during the control C02' 02 period and at 30 minute intervals throughout the experi- mental period. Percent oxygen saturation, total CO2 (mMol/l plasma), and HCO — content (meg/l plasma) were 3 calculated from these values using the animal's body temper- ature and the Siggaard—Anderson alignment nomogram. During the four hour observation period, the animal's progress was not interferred with except to add pentobarbital and heparin periodically, as described above. Either one or two tests were performed, before and again after the four hour experi— mental period. In half the animals (N=6), the cerebral vascular response to graded levels of hemorrhage was completed to before and a: toe test givt. murals were control le‘dt ”=5), the C; carbon dioxi: bral venous E tested befor diOXide was 5' “annular on other Side 01 inhalatiOn 01 Pressure was tubing to gr.- increaSeS 1 Six 0f hmu 0f ShOc ; .or a total «A . . Ca: fl diOXi S‘I 29 completed to test the cerebral autoregulatory response before and after four hours of endotoxin shock. Following the test given before administering the endotoxin, the animals were allowed to stabilize before obtaining the base control values for the experiment. In the remaining half (N=6), the cerebral vascular responses to: l) 7.5 or 10% carbon dioxide inhalation, and 2) graded increases in cere- bral venous pressure (venous-arteriolar response) were tested before and after the four hours of shock. The carbon dioxide was administered via respiratory bags attached to a Y cannula on the end of the animal's tracheal tube; on the other side of the Y a flutter valve was attached to prevent inhalation of room air during inspiration. Cerebral venous pressure was increased by raising the tip of the outflow tubing to gradually increasing heights yielding graded increases in pressure. Six of these animals were followed for an additional hour of shock and four animals of this group were followed for a total of six hours of shock following endotoxin administration. The autoregulatory, venous-arteriolar, and carbon dioxide responses were tested in these animals after six hours of shock. Eight additional animals died prior to the end of the four hour shock period. Group II (N=4) consists of animals also receiving a dose of 2 mg/kg g. ggli_endotoxin, and treated similarly to those in Group I except that the experimental period was {7 interrupted é vascular par!» :ere'oral res;- iere studied. sure, cerebri pressure, res I! :zte were mea renous blood ud at 30 min thee, the re hourly, in tr before and af the venous-a1 four hours of Group I‘ cerebral res, '15 lug/kg) wa Sion pressur fluid pressu u r’“ ature We: gun ”0d gases control peri . L‘ .Lv “Oi A11 30 interrupted every hour to test the cerebral vascular response to inhalation of carbon dioxide. The cerebral vascular parameters were monitored as described for Group I. GrouijII (N=5) consists of experiments in which the cerebral responses to a lower dose of endotoxin (1 mg/kg) were studied. Cerebral blood flow, cerebral perfusion pres— sure, cerebral vascular resistance, cerebrospinal fluid pressure, respiratory rate, body temperature, and heart rate were measured as in Group I. Arterial and cerebral venous blood gases were obtained during the control period and at 30 minute intervals in five of the animals. In three, the response to carbon dioxide inhalation was tested hourly, in three the autoregulatory response was tested before and after four hours of shock, and in two animals the venous-arteriolar response was tested before and after four hours of shock. Group IV (N=6) consists of experiments in which the cerebral response to a greater dose of E. coli endotoxin (5 mg/kg) was studied. Cerebral blood flow, cerebral perfu- sion pressure, cerebral vascular resistance, cerebrospinal fluid pressure, respiratory rate, heart rate, and body tem- perature were followed as in Group I. Arterial and cerebral blood gases were measured in all six animals during the control period and at 30 minute intervals during the shock period. All animals were followed without interruption during the f dioxide befo- in all anima hours of she an additiona 13: hours of . Group \J iiliCh 20 CC t minute perio Cerebral 1310, Vascular res. tory rate, h( in the expem blOOd gases .he animals ‘l reSPOIlSe t0 . ‘o VEhOuS art-Er C031: r01 peri< AG q~par v StatiSt- ;Jdentl S I! t1 31 during the four hours of shock. The response to carbon dioxide before and after four hours of shock was completed in all animals except two, in which the response after four hours of shock was not completed due to death of the animal. An additional animal, not included in the study, died after 1% hours of shock. Group V (N=8) is the control series of animals, in which 20 cc of normal saline was administered over a 5 minute period and the animals followed for four hours. Cerebral blood flow, cerebral perfusion pressure, cerebral vascular resistance, cerebrospinal fluid pressure, respira- tory rate, heart rate and body temperature were followed as in the experimental animals. Arterial and cerebral venous blood gases (pH, PCOZ’ P02) were obtained during the control period and at 30 minute intervals in all animals. Two of the animals were followed for a total of six hours following after administration of saline. The cerebral vascular response to inhalation of carbon dioxide before and after four hours of the control period was tested in seven animals, and after six hours in two animals. Autoregulation and the venous arteriolar response was tested before and after the control period in seven animals. Control animals received heparin and pentobarbital as needed. Statistical analysis was completed using the standard Student's "t" test for paired replicates (78), and is described i variance we t values we yielding a 32 described in Appendix A of this thesis. Two-way analysis of variance was also used for between-group comparisons. F and t values were considered significant in these analyses when yielding a "P" value less than 0.05. Hemodvnamic K The avl sure (syste: blood flow, fluid press: are ShOWn i1 decreased 5: r‘C‘llklined be: Period. The of an initia Statisticai: :i-ecrease f0: Cerebm‘l blc Ell/min (‘505 endotoxin ac {36% below c CHAPTER IV RESULTS Hemodynamic Alterations The average responses of the cerebral perfusion pres- sure (systemic pressure minus venous pressure), cerebral blood flow, cerebral vascular resistance and cerebrospinal fluid pressure following administration of 2 mg/kg endotoxin are shown in Figure 2. Cerebral perfusion pressure decreased significantly during the first five minutes and remained below control during the entire 4 hour experimental period. 'The general pattern of most experiments was that of an initial rapid decrease followed by a partial (but statistically significant) recovery, and another smaller decrease followed by a gradual rise toward control levels. Cerebral blood flow rapidly decreased from 34.1 to 17.0 ml/min (-50%, P‘10.005) during the first five minutes after endotoxin administration. A small increase to 20.9 ml/min (36% below control) occurred at min 30, followed by a grad- ual decline over the remainder of the experiment to 16.9 (48% below control, P‘<0.005). Cerebral vascular resistance decreased significantly (-35%, P<:0.05) during the first five minutes, rose to levels not different than control 33 34 i A N9 MC» C..X0~0UCW OX\OE N ‘ onsmmoum UHDHM Hmcflmmonbouoo u mmmu oocmumflmon Hmadomm> Hmubonoo u m>o 30Hw popo Hangouoo n mmo ousmmoum coflmswuom Hmubouoo u moo .cflxoquco Haoo .m mx\mE N mo coflumuuchHEUM msoco>mupcw mcflsoaaow con Hoasomm> Hmubonoo may mo noncommou OHEdcmpoEon ommuo>¢ .N ousmflm 35 m ohsmwm DQOVQ. AEEV oEC. wmumcoos. O¢N OwN Om. Onv our om cm on \F 0.. 0.. n O _ 1 . _ q _ _ . $ . 1 .. nJo Xgl 31.55 H H H a a o. ammo .W A .H .H L41 .H l .IN . H- ..H .H F “1‘ F 375 5.88ch 9.3.: N» 255.53125 m>U m 8 E525 mmo ow on 3: ea. 00.. an on. i (”0.05), ; experiment, 133 min pos: Hours p05? above contrr: decreased t: and leveled lP<0.05) fo A repre is ShOWn in CerEbrOSPJina activity inc endotoxin ad :iOOd flow a 36 (P3>0.05), and continued rising over the remainder of the experiment, the increase becoming significant (P<=0.05) at 150 min post-endotoxin. Cerebral vascular resistance at 4 hours post-endotoxin was 6.00, a value that averaged 55% above control (P<:0.05). Cerebrospinal fluid pressure decreased transiently at min 5 (P‘<0.05), then began rising, and leveled off at a level significantly above control (P‘10.05) for the last three hours of the experiment. A representative tracing from one of these experiments is shown in Figure 3. Systemic pressure rapidly decreased, cerebrospinal fluid pressure decreased, and respiratory activity increased during the first five minutes following endotoxin administration. During this period, cerebral blood flow and cerebral vascular resistance (calculated) decreased. Systemic pressure partially recovered before decreasing again, and then rose to levels near control over the last two hours. Cerebrospinal pressure and respiratory activity were usually increased. Cerebral blood flow was ~depressed during the entire 4 hours, and cerebral vascular resistance had increased to 8.34 mm Hg/ml/min by the fourth hour. Average cerebral hemodynamic responses following admin- istration of 5 mg/kg endotoxin are shown in Figure 4. The pattern of the responses is similar to that of the lower dose of endotoxin. Cerebral perfusion pressure and cerebral 37 . ,.l. ._,_.._._,l_ll:111.111ia.ii.i_:.e...,§lOON 2.x05002m 0J\DEN 9 oocmumwmou amasomm> Humbouoo u m>o 30am cooHn Hmnnouoo u mmo onsmmoum Uflsam Hmcflmmoubouoo u mmom unannoum Hafiuouum owsoummm u emumm .cflxOUOUco Haoo xm mx\mfi m mo coaumnumficwspm msoco>muucfl mcflBOHHow unmafluomxo cw Eoum mafiomuu o>aumucomoumom .m ousmfim m oudmflm is 2:2. own on. ou. co 8 0n n— m o 1 I I I I I I I I I I A:1:\_E\OIEEV 2 N 24a. 96 and 86 . .t V «a an no N 3 a >U A:_E\.EV Mr.— —N— 92 5.3 «.2 05 flap 0..."— afin ununv >=>=u< 30.9238. 38 we; 1 LLL _ LL_L;, 256593 9.x?“ ¢ 39 K «Wu-L» CIXOIOUCU DX\OE0 \ ousmmonm madam Hmcflmmounouoo u mmmo oocmumflmon Hmasomm> Hmubouoo u m>u 30Hm pooHQ Hmunouoo u mmo oHSmmmHm cowmsmuom HmquHoo u mmu .cflxouOUco HHoo .m mx\mE m mo coaumuumwcflapm msoco>wuucfl mcfl3oaaom con smasomm> Hmunouoo onu mo noncommou anmcmposon ommuo>< .v ousmfim 40 e onsmflm DQOVQ. 2.5 2.: mmnmsas. 0¢N 03 00.. 00w 8.. 00 00 on LP 0.. 0.. n 0 _ _ a H . . . . § . m flJO H H H H K 1 Ao—LEEV H H L H H -9 ammo L JN Amucv c_x2oucm oxonn » l l c 32:53. as. $6 0 ON “ESE—5 “.8 0? on .2. .55 00.. nun—U 00—. aa—J... blood flow first ten : below cont: (P< 0.05) . to decrease this decree then gradua COfltI'Ol by P< 0.05) . (P< 0.05), 90 (P < 0.05 control frc Vidual res; illustrated Endotc ing 01. imme in 3 and 3 Atracing ( trated in 1 during the if“creased . 1:0an or t] cur: ‘ ”lg tile \‘ (15 41 blood flow decreased rapidly and significantly during the first ten minutes (56% and 47%, respectively) and remained below control during the entire 4 hour experimental period (P‘<0.05). Initially, cerebral vascular resistance tended to decrease, although due to the variation of the response this decrease was not significant (P>;0.05). Resistance then gradually rose, and reached levels significantly above control by the fourth hour of shock (37% above control, P‘<0.0S). Cerebrospinal fluid pressure initially decreased (P<<0.05), then increased to a level above control at min 90 (P‘<0.05) and decreased to levels not different from control from min 120-240 (P> 0.05). In general, the indi— vidual responses of these animals were similar to that illustrated in Figure 3 for the 2 mg/kg dose of endotoxin. Endotoxin administration resulted in death either dur- ing or immediately following the 4 hour experimental period in 8 and 3 animals in the 2 and 5 mg/kg doses, respectively. A tracing obtained during one of these experiments is illus- trated in Figure 5. Systemic pressure decreased dramatically during the first 5 minutes from 105 to 27 mm Hg, gradually increased to 63 mm Hg, and then declined during the last hours of the experiment. Cerebral blood flow rapidly fell during the first 5 minutes, and remained depressed during the experimental period. The transient increase in cerebral blood flow which occurred at 15 min post-endotoxin may be . due to cerebral dilation caused by increased levels of PCO 2 42 _ ,_ , __ a _ . I.--< Ii. .5“on _,_ L 1,. L_lL_ ._,_,1, _,1 ... w..i 1 . m . , , _. L: T... ......... __.1.1.NOON Z.X0~002w 04\DE m. ‘ mocmumamon unasommb Hmwbonoo 30am pooHn Hounouoo whommoum macaw Hmcammonhouoo ousmmoud amauounm OHEoumwm .cwx0000co waoo em mx\mE m u m>o H mmU mmom Bmwmm mo cowu Imnumficflspm msoco>muucfl 00 one Hmeficm may no snoop ca mcwumcflfiuop ucofiwuomxo cm Eonm mcwomuu o>wumucomoumom .m ousmflm m mudmflm 3.... 08:. 3“ on. cup on on on m. m o - I J u u u u u .i - - Ac_E\_E\u:EEV 8m in 3." a.» and an 2.. mm... 8..” ¢>U as 3. «.0. od— v..._ 0.2 93 .3. ouu 25:3 ”— m U >:>_.u< regimes 43 .17.? E We} 1 L veritL Lem will. .2. o 0mzm1u’0.05) and became significantly above control by the fourth hour of shock (25% above control, Ps<0.05). The only significant change which occurred in the cerebrospinal fluid pressure was a transient decrease (P<=0.05) which occurred at min 5. The control series Of animals demonstrated little ef- fect of the surgical maneuvers upon the course of the experiment over the four hour period. Figure 7 shows that the only changes in the cerebral hemodynamics occurring following administration of 20 cc of normal saline into the 45 V,INODF «Out» CIXOIOUCU UX\DEF ‘ ousmmoum pagan Hmcflmmouhonoo u ammo oocmumflmou HmHsomm> Hmubouoo u m>U 30am pooah Hmuhouoo u mmo ousmmonm coemswumm Hounoumo u mmu .cflxouoccm “H00 .m mx\ma H mo coaumuumflcflspm msoco>muucw mCH3OHH0m won “masomm> Hohhouoo 0:» mo noncommou owfiocmpoaon mmmuo>¢ .m ousmwm 46 0¢N 05 00* m ousmflm _ _ _ mOdVau AEEV 9:; mm H meow—2 00.. 0N_. 00 00 on tier 0* m 0 _ _ _ _ dies .; H 1;J..0 1H . - 3: SE. M M H/H Hi 1 9 ammo H H L .IN A H c: l 1582:.on Ea; H\m\n H [H m>o H 1 H 1 m IHLLIOH I” OH O." Q 0 I. 0N H ’41 35:5 .. mmo IL0¢ .100 o o o 4} 1 3: ES. H. 1 00¢ n—QU Bl: ExBoucm 9.3:: » 09. NH H H H H. H: H .Hr H N «was .OtCOU DEIOW .OELOZ ‘ ousmmoum pflus Hmcflmmouhouoo oocmumwmou Hoasomm> Hohhonoo 30am vocab Hmnnouoo ousmmoum cowmsmnom Hounouoo 47 on“ mmmU m>U th mmU .mucoEHHomxo Houusoo mcflusp won Hmasomm> Hohhouoo may 00 momcommon Daemchpoaos ommum>¢ .h madman 48 h madman 00.0VQ0 2:5 25... mm «meow—2 ocm o_.~ cm. 09. ON. om cm on n. 9 n o \4 H _ _ _ _ _ _ mm . _ . _.Lo 4 HH; a . 31.55 H) lat fir 0.. ammo JN . 352...}: .55 a m _ m m m mi. «>0 [0 I L8 E525. Lov 190m. - 316...... H. iii) H fl}! T H to». $1.; .8250 8:8 66.62 » preparati. pressure .I and a ver; increased systemic ; the conco: ance may {- ture, as (W 4 hours. Changed_ A con. Show” in F 5‘. “0" w«is 5 Point for t endotoxin < was reduCeg at Roma 0: Control sex endotOxiu t signifiCan t 2 m‘3/kg end 0o H Se 0 0f C eat' but th greater eff blood flow 49 preparation were a small increase in cerebral perfusion pressure and cerebral vascular resistance from min 90-180, and a very slight increase in resistance at min 15. The increased perfusion pressure was due to a small rise in systemic pressure during this time period, and as a result, the concomitant period of increased cerebral vascular resist— ance may be due to autoregulation of the cerebral vascula- ture, as cerebral blood flow did not change during the entire 4 hours. Cerebrospinal fluid pressure also remained un- changed. A composite of the percent change of the cerebral blood flow responses of each of the four groups mentioned above is shown in Figure 8. The percent change of cerebral blood flow was significant at least at the P‘<0.05 level at every point for each dose of endotoxin. At four hours post- endotoxin cerebral blood flow in the l, 2, and 5 mg/kg doses was reduced by 37, 48, and 45%, respectively. In contrast, at none of these points did the cerebral blood flow in the control series change significantly. At four hours post- endotoxin the percent change of cerebral blood flow was significant at the P<=0.001 levels for the animals receiving 2 mg/kg endotoxin, and at the P‘<0.0l levels for the 1 mg/kg dose. Of course, the n value for the two groups was differ— ent, but the general trend of the graphs shown indicates a greater effect of 2 mg/kg and 5 mg/kg endotoxin in reducing blood flow than the 1 mg/kg dose. 50 Exoaohucm ‘ .AmocHH pHHOmV conuopco Haoo am mx\mE m can m .H mafi>flooou massage as who Amcwa ponmmpv mucosfluomxo Houucoo maeugp 30am polo Hmubouoo mo omcmno ucoouom ommuo>¢ .m ousmflm 51 m ousmflm mmumcooz 2.525 0% 9m om. on. 8. om om on : 99 n o H H H H _ d _ _ : _ H .3100: HNHucVSBENH _ IIHI H 18- Gusszoent. LLILL \ BucvoxBEH \HJ ‘ .. ow; no? 3.3 oncogo low: 23th lo... ‘\\“W"U" |“|‘HIIUIHIIIIII#UUUU | I HmucrotcooT m. ... 1T u Tm. um... no Lo... Eonoucm H -‘fl" *7:- The these gro the perce' 2 IncJ/kg dc At four h] 0f endoto: 23, 55, a: resistance Value not‘ ' - 1 A5 lndlcaf 0f endoto) Cerebral ‘ leVels I 51’ EXperimem IESistanCE entire fir 52 The percent change of cerebral vascular resistance in these groups is shown in Figure 9. Again, it appears that the percent increase in resistance is greatest in the 2 mg/kg dose and least in the 1 mg/kg dose of endotoxin. At four hours post—endotoxin, the l, 2, and 5 mg/kg doses of endotoxin had increased cerebral vascular resistance by 25, 55, and 37%, respectively (P‘<0.05). Cerebral vascular resistance in the control experiments had increased 6.6%, a value not significantly different from control (P>»0.05). As indicated on the graph, the general response of all doses of endotoxin was that of an initial transitory decrease in cerebral vascular resistance, a leveling off at near control levels, and a gradual increase over the remainder of the experiment. Interestingly, however, in the lowest dose of endotoxin (1 mg/kg), the percent change of cerebral vascular resistance remained significantly below control for the entire first hour, whereas the 2 mg/kg dose returned to control levels by min 10, although the initial decrease for the two doses was nearly identical. Thus, it appears that the initial dilation period was extended in the lower dose of endotoxin; indeed the subsequent increase in resistance was also delayed and became significant only during the final 30 minutes of the fourth hour. Although individual points of the 2 and 5 mg/kg doses were not significantly different using the Student's t test, two-way analysis of variance was performed on the data comprising these two 53 2.. XOIOUCW‘ .Hmocwa pHHOmV :HxOHOUGo mx\mE m can N .H mcw>flmoou mHmEHcm CH paw AocHH poammpv mucoEHHomxo Houucoo mcwusp monopmfimou Hmaoomm> Hmnnouoo mo omcmno pcoouom omwuo>< .m ousmwm 54 m ousmwm 3.9 "C 2622 3.5 2.: ovm 05 00.. 09 Our 00 00 0n 9 0H m 0 H H H I ;H _ . _ «w _ H H _ 0n: 3... mococo HemoLon. a"... 6.2.8 H. Busofoc: .mucVOxonm an... 9.35 -8. Los+ Eonoecm H 55 curves. The "F" value demonstrated that these curves were significantly different from each other at the 0.01 level of confidence. Thus, indeed, the 2 mg/kg dose appeared to be most effective in increasing cerebral vascular resistance. The l and 5 mg/kg dose curves were not significantly differ- ent utilizing this method of analysis. Results of the experiments which were given a dose of 2 mg/kg endotoxin and were followed for a total of 6 hours are illustrated in Figure 10. This graph demonstrates that the increase in resistance described previously as occurring from hours 2 to 4 began leveling off from hours 4 to 6. In turn, cerebral blood flow decreased only slightly from hours 4 to 6, and no further increase in perfusion pressure was noted during these hours. Alterations in Cerebral Vascular Reactivity Responses of the cerebral vascular bed to progressive decreases in cerebral perfusion pressure (autoregulation), increases in cerebral venous pressure (venous-arteriolar response), and increases in arterial P levels (response CO to C02 challenge) are illustrated in Figires 11-14. Figure 11 exhibits the plots of the autoregulatory curves before and after 4 hours of shock due to 2 mg/kg endotoxin. In the upper graph A, the mean cerebral vascular resistance prior to endotoxin administration (solid line) decreased as the perfusion pressure decreased over the range from control Figure 10. 56 Average hemodynamic responses of animals receiving 2 mg/kg E, coli endotoxin and followed for six hours. CPP = cerebral perfusion pressure CBF = cerebral blood flow CVR = cerebral vascular resistance CPP | (mqu) CBF (MI/min: ( CVR 150 F CPP 100 '— ’.—.’.-.-.—.-._. ("mu”) V 50 — 50 - CBF 40 r- l (m/nin) 30 __ 20 b Oh~ /\‘.—.—.‘0—.—.’ 6 r CVR y,_,__.__.-~ 'min) 2 _ L l 1 l l l J 57 ) 2 "lg/kg Endotoxin (n = 4) TIME (hours) Figure 10 58 CVR (mmHg 4% Figure 11. Average responses of cerebral vascular resistance (A) and cerebral blood flow (B) to graded levels of hemorrhage before and after four hours of shock due to administration of 2 mg/kg E. coli endotoxin. CPP = cerebral perfusion pressure CBF = cerebral blood flow CVR = cerebral vascular resistance CBF (ml /hhn 59 6.0 -- A ,o 00" n ' 6 — o” 0—0 Before Endotoxin O." °---o Mm 4 m Endotoxin 5.0 .— '0' CVR i" ( mm Hg/ __ 0' m/min) "o” 4.0 '- O"' / at) La 1 l l .L i. 20 40 60 80 100 1 20 140 CPP (mm Hg) 40 - B /. o lCBF' 3£>b o—""’ ("H ”/’I’ /min) 20 __ . ,.o III-I-O III-III- O‘.‘.. l 0 g l 1 l l 1 20 40 60 80 100 1 20 140 CPP (mm Hg) Figure 11 030 mm signific in all 'r endotoxi decrease. pressure lower gr.- hlood flc Pressure. from 31.8 dmpped r 19%. Aft was again! decteased being 0111' l endOtoxin MOIe horiz train (801 unimpaired FiQUr reSPOHSe t 4 hOurS 0 f This graph Changes, i I J 300 then (1 60 (130 mm Hg) to 40 mm Hg. The decrease in resistance was significant (P‘<0.05) at 80, 60, and 40 mm Hg and occurred in all but one animal at 100 mm Hg. Following four hours of endotoxin shock (dashed line) cerebral vascular resistance decreased in every animal with each decrease in perfusion pressure over the range from 100 mm Hg to 40 mm Hg. The lower graph, B, shows the changes occurring in the cerebral blood flow during the same decreases in cerebral perfusion pressure. Before endotoxin, cerebral blood flow decreased from 31.8 to 28.4 ml/min as cerebral perfusion pressure was dropped from 100 to 80 mm Hg, the average decrease being 19%. After 4 hours of shock, as cerebral perfusion pressure was again drOpped from 100 to 80 mm Hg, cerebral blood flow decreased from 19.3 to 16.0 ml/min, the average decrease being only 9%. In addition, the slope of the line after endotoxin (dashed line) is not as great, i.e., the line is more horizontal to the pressure axis, than that before endo- toxin (solid line). Thus, the autoregulatory response is unimpaired following shock. Figure 12 illustrates the cerebral "venous-arteriolar" response to increases in venous pressure before and after 4 hours of shock due to administration of 2 mg/kg endotoxin. This graph indicates the cerebral vascular resistance changes, in absolute value and in percent change, as the cerebral venous pressure was increased to 6 and 11 mm Hg and then decreased back to control levels. Prior to Figure 12. Average response of cerebral vascular resistance to increases in cerebral venous pressure before and after four hours of shock due to administration of 2 mg/kg 61 E, coli endotoxin. CVR P V cerebral vascular resistance cerebral venous pressure CVR (mmHg my CVR 5‘ OF CONTROL Pv (Mm 62 0—0 Before Endotoxin 7-0 l" o----o After 4 hrs Endotoxin .,¢‘ ‘~ ( CVR _}_..»—"}f‘ n - s m H%%n 5.0 t ) {/E/Ri 3.0 — “m: t SE 'P<0.05 120 F" CVR “W 110— CONTROL 1OOL. PV (mmHQ) : 1 Figure 12 endotox; each ti: respect; reduced. resistar. . sure (7. near cor. the S€Cc signific Thus, t3; Pressure toxin sh. The incIrease due to a: The absoj written % cerebral large dec of 42.9% the Cereb arterial (p. 0.001 during Sh indicatin L0 dilate 63 endotoxin administration, cerebral resistance increased each time the venous pressure was increased (5.2 and 16.8%, respectively) and then decreased as the venous pressure was reduced. After 4 hours of shock, the mean cerebral vascular resistance again increased at each increase in venous pres- sure (7.1 and 13.8%, respectively), and then decreased to near control as the venous pressure was reduced. Although the second increase in resistance prior to shock was not significant, only one of these animals showed a decrease. Thus, the cerebral vascular response to increases in venous pressure was maintained following the four hours of endo- toxin shock. The reactivity of the cerebral vascular bed to an increase in arterial PCO before and after 4 hours of shock 2 due to administration of 2 mg/kg is shown in Figure 13. The absolute levels of the cerebral vascular resistance are written at the top of each bar. Prior to endotoxin, the cerebral vascular bed responded to the CO2 challenge with a large decrease in cerebral vascular resistance to a level of 42.9% of control (P‘<0.001). After 4 hours of shock, the cerebral vascular resistance during the increase in arterial Pcoz decreased to a level of only 60.2% of control (P‘<0.001). In addition, the percent change in resistance during shock was significantly less than that prior to shock, indicating a lessened ability of the cerebral vascular bed to dilate in response to the CO challenge after endotoxin 2 Figure 13. 64 Average percent change of cerebral vascular resistance due to inhalation of 7.5% carbon dioxide prior to and four hours after admin- istration of 2 mg/kg E. coli endotoxin. CVR = cerebral vascular resistance C1 = initial control state, room air ventilation C2 = control state after CO2 challenge, room air ventilation CVR 7.0? courno 65 BEFORE AFTER 4 HOURS 12° F ENDOTOXIN ENDOTOXIN 100 80 CVR 9. OF CONTROL 6° 40 20 ".6 01 002 02 PaCOz 40.7 59.7 - 27.3 50.8 — at. A P0002 477d 86%? *p<0.05, compared with control, C1 +p<0.05, compared with C02 before Endo Figure 13 66 administration. This occurred in spite of increased vascu- lar resistance (6.16 compared to 3.83) and a greater percent change in the PCO (86% increase post endotoxin compared to 2 only 47% increase prior to endotoxin). Although the abso- lute level of arterial P during C0 C02 2 after 4 hours of shock, this decrease is not significant, inhalation is lower and in fact, the animal which demonstrated the greatest loss of CO2 responsiveness also showed a greater arterial PCO 2 during CO inhalation after 4 hours of shock than before 2 shock. Figure 14 illustrates the hourly responses of the cere- bral vascular bed to the CO2 Challenge before and during endotoxin shock. These animals are not the same as those illustrated in Figure 13, but also received 2 mg/kg endo- toxin, and were interrupted every hour to test this response. Prior to endotoxin, these animals demonstrated a marked decrease in cerebral vascular resistance in response to the increased arterial PCO , to an average of 37% of control. 2 At only one hour post-endotoxin, this dilatory response was markedly diminished, as resistance only decreased to a value C02 Average values of percent of control at 2, 3, and 4 hours averaging 81% of control, in response to the increased P post-endotoxin were 80, 79, and 75%, respectively. In every animal, the cerebral vascular resistance decreased with each CO2 challenge, but the decrease was always considerably smaller than that observed in the control state. 67 coaumHHuco> Ham Boon .oumum Houusoo u U mosmumwmmu umasomo> Hmunmuoo u m>o ouommoum coamomuom amunoumo n mmu .cwxouOUco HHoo um mx\mE N no coaumuumHCfifipm Houmm mam>uousw Maudos no can on HOHHQ opwxowo conumo wm.n mo aowumaoncw on map mocmumflmou Hmasomm> Hmunonoo mo omcwno unmouom ommuo>¢ .va madman 68 ea mosses oucm 223 ~00 2 3.368 nodva+ .268 2 notanoo modvae ~e¢.¢>o me¢.¢>u Q8. . .5 Q8 . to Eb . .Zo o~80 oNoou ouooo N n c 2.x0hoozm I Fmom Z . XOFOOZM rhmom 2. XOFOOZUIPmOm ONP mKDOI 0 mum—JO: .v $130... n km.” a ¢>O mnfi u ¢>O hwfi n ¢>O 0&0 a mac 9N5 u mmo FKN— :30 0800 ouooo omooo z _ XOPOO 2w -kmom 2.x2002m akmom z _xo._.oozu manor N ~30: — mach—mm Jomhzoo no «r m>o Jam—#28 no «v «>0 69 Two additional animals from another group were studied at 6 hours post-endotoxin, and their data is also presented here for comparison. As indicated, even at 6 hours, the CO2 response was still depressed. The cerebral vascular resistance at this time only decreased at a level of 69% of control. Cerebral vascular resistance and cerebral perfu- sion pressure at each hour prior to CO inhalation is also 2 given to point out that the resistance pattern of this group of animals was similar to that previously reported for the other experimental groups. The control series of experiments clearly maintained the autoregulatory and venous-arteriolar responses four hours after initiation of the control period. In addition, these animals, in contrast to those which received endotoxin, did not demonstrate a decrease in the cerebral vascular response to increased arterial P Initially, cerebral CO2 bascular resistance decreased to 33% of control in response to CO2 inhalation, and four hours later, cerebral vascular resistance decreased to 35% of control during CO2 inhala- tion. In two control experiments which were followed for a total of 6 hours, cerebral vascular resistance decreased to 32% of control in response to the CO challenge both before 2 and after 4 hours of shock. 70 Cerebral Metabolic Changes Systemic arterial and cerebral venous blood gas changes during endotoxin shock are shown in Figures 15-18. As shown in Figure 15, immediately following the administration of 2 mg/kg endotoxin, the respiratory rate increased dramatical- ly and remained significantly above control during the entire four hours. Arterial P02 remained increased over nearly all of the experimental period, while cerebral venous P02 decreased by 60 min and remained depressed. Arterial PCO decreased below normal, and except for a slight decrease 2 at 30 min, the cerebral venous PCO remained unchanged 2 following endotoxin administration. Blood pH decreased con- siderably in the cerebral venous blood except at the fourth hour, and except for an initial transient decrease during the first hour, the arterial pH was unchanged following endotoxin administration. Thus, the major changes which occurred in the cerebral venous blood during endotoxin shock due to 2 mg/kg endotoxin were a decrease in P and pH. 0 As seen in Figures 16 and 17, the blood gis and respira- tory alterations which occurred following administration of the 5 mg/kg and 1 mg/kg doses of endotoxin were comparable to those described above. The respiratory responses of these doses were variable, but often significantly increased. The venous blood gases of both doses indicate a decreased P02 and a significantly unchanged but tending to increase PCO over the four hours. The cerebral venous pH decreased 2 71 Figure 15. Average responses of respiratory rate, and systemic arterial and venous blood gases following administration of 2 mg/kg E, coli endotoxin. 72 ‘2 mg/ kg Endotoxin (n= 8) —- systemic arterial blood --- cerebral venous blood 110 - 100 - ' . . I. I, . 90 *- l l ‘1‘ ‘1; 4.}- P02 80 .— 1: (nmeg) 1“;- 40 e ‘n-ni... 30 _ ..i:-..-'§‘---'.!:-u_§$----Cig----'I:--.-'I. L - T .1 . .- P I. I‘ -{' i .r J- ~-.I.--" I. ---I 002 40 (nman) 20 7.40 ”l' 7.30 (units) ~ 6‘ “. u... .‘..-I:..- I 7.20 - V 53......rn-lh-IQ-QQ: -e. 20 P C Respiratory Rate L (breaths/min) 0"Ll 11 1 1 1 1 1 1 I 0 3° 60 90 120 150 180 210 240 Mean tSE Time (min) 'p<0.05 Figure 15 73 Figure 16. Average responses of respiratory rate, and systemic arterial and venous blood gases following administration of 5 mg/kg E. coli endotoxin. 74 iqu/kq Endotoxin (n=6) — systemic arterial blood -- cerebral venous blood 110 l- .. J. 100 M T T 1' T P 90 02 80 (mm Hg, 5 50 4o 30 P902 40 (mm Hg) 20 140 130 pH (units) 120 710 20 Respiratory Rate (breathe/min) 0" LLJ 1 I l 1 1 L l 1 0 3O 60 90 120 150 180 210 240 Means 1 SE Time (min) 'p<0.05 Figure 16 75 Figure 17. Average responses of respiratory rate, and systemic arterial and venous blood gases following administration of 1 mg/kg E. coli endotoxin. 76 ‘ 1mg/kg Endotoxin (n=5) — systemic arterial blood -- cerebral venous blood 110 F 100 l’ I I To IL I T 90 — l l l l l l ‘1 P02 80; (mm H9) 501; 40 — I...- 30 ’- ..I.~. t T .. nun-Iolonnnn ICC”.." 3. L. _ T 20 _ .10 e "e “I 60 - _ I 1'__ I T pc02 I-- i—=‘=-‘I i I 1' 1 l 1 40 - (mm Hg) t I\I\ T , i I 20 — I. I. + " 7’. l. ‘1' 74° F r I -— T I I I 1 T I p h— LQ. ’ a... (mum 75° ~~~{-—-" 1' {“7”} T I I I l 1 1‘ 1- 7.20 — 30 - Respiratory Rate 2° ’ (breathe/min) “Dr L l J L l l l l l l I 0 30 60 90 120 150 180 210 240 MeansisE Time (min) 6 'p<0.05 Figure 17 77 further and more quickly following administration of the 5 mg/kg dose than either the 1 mg/kg or 2 mg/kg doses of endotoxin. The animal's body temperature and the P , P , CO2 02 pH values shown above for the 2 mg/kg dose were used to and calculate the arterial and cerebral venous oxygen satura— 3 I mEq/l plasma) from the Siggaard-Anderson alignment nomogram. tion, total CO2 (mMol/l plasma) and bicarbonate (HCO The changes occurring in these parameters during endotoxin shock are shown in Figure 18. As illustrated, the arterial- venous differences of all three parameters increased during endotoxin shock. The increase in the arterial-venous dif- ference of the oxygen saturation was due to decreased oxygen saturation of the cerebral venous blood. Although hemo- globin content was not measured, one can tell intuitively from this graph that the arterial-venous difference for oxygen content increased dramatically, also. In turn, the increases in arterial—venous differences of total CO2 and HCO3- were due to disproportionate decreases in arterial levels of these values rather than to increases in the venous values; indeed, the venous levels of total CO2 and HCO3- actually tended to decrease. None of these alterations in systemic arterial or cere- bral venous blood gases occurred in the saline control experiments, as seen in Figure 19. Respiratory rate also 78 .conuOUGo fiaoo am mx\wE N «0 cowumuumflcflaom mcwsoaaom ucmucoo imoum can moo Hmuou .coflummsumm cmmwxo msocm> Hounmuoo can Hafiuouuo anoummm mo mmmcommou mmmno>< .ma magmas 79 ma masons Es. use 8.9% O¢N O.N 00. On. ON. 00 00 On 0 1i 1 1 1 4 1 1 1 .Illl‘llll... lllll OIII'IOIIIIIOIIIIIOIIIII’I 0m .0. little tilliettllne iiiii Oliiiieii . oiliil 11.07:: ON one... 26:2, .2983 uuuuu . On 0003 .23.... 0.6293 \|\|.lll J on iiiibl ii) Iiielltie IIII.e\ // 0.. I / // . 00 «a. II A 201. nu . 0m T‘ 1+ T 0 i” [ii i JillltlL .00. 8.2. ExoBvcm 3.3qu acacia amp. m8: 068: .. .22 E N8 .28 .x. .omwo 80 Figure 19. Average responses of respiratory rate, and systemic arterial and venous blood gases during control experiments. 81 i Saline Control (n=8) 110 _ _systemic arterial blood --- cerebral venous blood 100 - 90 - I T I I (mm Hg) 80:;- 50r 4o .- !e-nueioluiliii——: ; _ ‘1 i __ .1 ————:.-----i 3OL I I _____ - - l- I T .1 -::_ I. T l 1 I. l P 1 1 I I 1 T I T 4‘J (mmnm _1 l 1 I l 20* '140- L .. _ .. pH 1 v : 0 TA. (units) 13° F I I I_ .. I T I l I l l J. L l 3 —— I: -1 “120— 20~ Respiratory Rate __ 1- 1' T 1- T J (breaths/min) 10 ‘- 1 l .L I I I .1. 0*"L1 11. 1 1 1 1 ..1 1 1 O 30 60 90 120 150 180 210 240 Meansi SE Time (min) 'p<0.05 Figure 19 82 remained statistically unchanged throughout the four hour observation period of the control experiments. ‘21... 1.. ....- :v CHAPTER V DISCUSSION The intravenous administration of purified E. coli endotoxin into dogs produced significant changes in each of the three areas studied: 1) cerebral hemodynamics, 2) cerebral vascular reactivity, and 3) arterial and cere- bral venous blood gases. Cerebral blood flow was consider- ably decreased during the entire four to six hour experi- mental period. Cerebral vascular resistance transiently decreased, returned to control levels, and gradually in- creased to levels above control by three hours post- endotoxin. The reactivity of the cerebral vascular bed to increased levels of PCO was diminished, although the venous- 2 arteriolar response and cerebral autoregulation were main- tained. Finally, blood gas alterations in the cerebral vascular bed include increased arterial-venous differences of PO , PCO , and pH and calculated differences of percent 2 2 02 saturation, total C02, also. The three doses of endotoxin used in this study (1, and bicarbonate content increased 2, and 5 mg/kg) produced responses similar in magnitude and direction, although the lowest dose tended to produce the least hemodynamic alterations. 83 84 The technique used in this study to isolate and direct- ly measure the cerebral outflow was that described by Rapela and Green (44,94). The measurement of the cerebral venous outflow is considerably more practical and less traumatic than measuring arterial inflow. The measurement of arterial r inflow requires more extensive dissection and metering of t blood flow in at least four arteries simultaneously because of the numerous arterial anastomoses between intra and extra- 5 cranial circulations (44). With the current technique, a portion of the cerebral venous outflow is collected directly without contamination from extra-cerebral vascular beds. The surgery involved is a relatively simple dissection which eliminates communications of the cranial and extracranial veins and sinuses of the skull. This separation is supported by latex injection studies which demonstrate that no sig- nificant amount of the blood escapes from this portion of the brain via communications with collateral venous channels (44). It is important to remember that all experimental preparations were required to meet the following criteria 1 before considered acceptable for use: 1) a rapid increase in cerebral venous pressure to at least 40 mm Hg during transient occlusion of the outflow tubing, and 2) good cerebral reactivity as demonstrated by autoregulation and good vasodilation in response to CO2 inhalation. The first criterion insures a tight, relatively communication-free system for collection of the cerebral venous outflow, and 85 autoregulation and cerebral vasodilation during systemic hypercapnia are considered outstanding characteristics of a reactive cerebral vascular bed (9,35,70,94). Administration of endotoxin caused a rapid and severe depression of cerebral blood flow over the entire observa- FE tion period of 4 to 6 hours. By the fourth hour of shock, the l, 2, and 5 mg/kg doses of endotoxin had reduced cere- bral blood flow by 37, 48, and 45%, respectively. These i" results are in agreement with those previously reported in similar experiments (30,31) performed in artificially respired dogs. In these animals, a 2 mg/kg dose of endo- toxin initially decreased cerebral blood flow to 63% of control, and by the fourth hour of shock, cerebral blood flow had decreased to 40% of control. In contrast, Buyniski gt_gl. (15) similarly measured cerebral blood flow and reported that cerebral vascular conductance was increased following administration of §, typhosa endotoxin, and that cerebral blood flow decreased only slightly. However, blood flow was measured for only one hour after endotoxin adminis— . tration. Other investigators, utilizing different techniques to measure cerebral blood flow, have reported evidence for a decrease in cerebral blood flow during endotoxin shock. Gallo and Schenk (39) reported a decrease in carotid and vertebral blood flows following administration of 5 mg/kg endotoxin to dogs. Weiner (123) administered 10 mg/kg 86 E. coli endotoxin (LD88) into rhesus monkeys and reported that cerebral blood flow had decreased from 35.2 to 7.0 ml/min at 120 minutes post-endotoxin. This decrease in blood flow is greater than that observed in the present study. Blood flow in these experiments was only followed w for two hours, and was not measured directly, but was 131 _ :F.’N".*t 'xd accomplished by cephalic counting of injected Cs activ— ity. Thus, radioactivity of the animal's entire head was '” counted introducing the skin and muscle vascular responses as additional considerations. Miller (83) reported the cardiovascular effects of endotoxin in unanesthetized monkeys using the distribution of radioactive microspheres to measure regional fractions of cardiac output and blood flow. This group found that the fraction of the cardiac output to the brain and brain blood flow was decreased 80 minutes after endotoxin administration. Thus, the evidence presented in the present study for decreased cerebral blood flow following endotoxin adminis- tration is supported by most of the studies cited above, ' although most of these studies employed indirect methods of measuring cerebral blood flow and were not studied for longer than two hours. It is of interest to compare these results with cere— bral blood flow alterations during shock induced by hemor- rhagic hypotension. Rittman and Smith (101) measured blood flow in the carotid and vertebral arteries and reported 87 that a 24% reduction of the blood volume was associated with a 40% reduction in "cerebral" blood flow. However, Green and Rapela (44) measured cerebral outflow directly and found that as systemic pressure was lowered stepwise, cerebral vascular resistance declined and cerebral blood flow remained F? at near control levels for nearly two hours of hypovolemic i hypotension. The difference thus observed in the response of the cerebral vasculature in hemorrhagic hypotension and r shock and in endotoxin shock may consequently indicate a difference in the significance of the cerebral vasculature and central nervous system in the pathogenesis of the two types of shock. As with the other vascular beds of the body, the blood flow to the cerebral vascular bed is dependent upon changes in perfusion pressure (arterial pressure minus venous pres- sure) and the resistance to blood flow through the cerebral vasculature. In the present study, the initial decrease in cerebral blood flow occurred concomitantly with decreased systemic arterial pressure, and thus decreased cerebral per- a fusion pressure. Therefore, the early decrease in cerebral blood flow was due to a decreased pressure gradient across the bed rather than an increase in resistance. In fact, the immediate decrease in blood flow was associated with a decrease in cerebral vascular resistance. Subsequently, the decreased blood flow was due partially to an increasing resistance and partially to a decreased perfusion pressure, 88 and in the last hours studied, cerebral perfusion pressure was nearing (but not equaling) control values and as a re- sult a greater amount of the decrease in cerebral blood flow was due to the increased vascular resistance. Thus, although the decreased cerebral perfusion pressure played a role in decreasing cerebral blood flow throughout the entire experi- mental period, it was primarily responsible for the decrease seen in the early hours of shock. The pattern of cerebral vascular resistance alterations during endotoxin shock are very interesting. Following endotoxin administration, resistance transiently decreased, rose to control levels, and gradually increased to levels significantly above control by minutes 150-240, depending on the dose given. By the fourth hour of endotoxin shock, cerebrovascular resistance had increased by 25, 55, and 37% in the l, 2, and 5 mg/kg doses, respectively. Thus, there appear to be two phases in the resistance response of the cerebral vascular bed to endotoxin: 1) an initial decrease, and 2) a long term increase. The initial decrease in cerebral vascular resistance occurred concomitantly with the decreased cerebral perfusion pressure. Thus, it resulted from an active increase in vessel calibre, as the decreased arterial pressure would lower transmural pressure and produce a passive decrease in vessel calibre. This early decrease in resistance is prob- ably an exhibition of the cerebral autoregulatory 89 phenomenon, as most authors have reported that endotoxin has no direct effect upon cerebral vasculature, as shown by local injections (15,118,123). In turn, this phenomenon may be theoretically due to: 1) increased concentration of vasodilator metabolites, 2) a myogenic response of the cere- bral vascular smooth muscle, or 3) a neurogenic response mediated through decreased pressure in the carotid sinus. The involvement of a neurogenic mechanism appears un- likely. Although it has been suggested that decreased pres- sure in the carotid sinus may cause cerebral dilation (100), most authors report that neither the carotid sinus or aortic pressure receptor reflexes play a necessary role in cerebral vascular autoregulation (95). Thus, the probable causes of the initial decreased vascular resistance are increased con- centration of vasodilator metabolites and/or an intrinsic myogenic response. Theoretically, increased concentrations of vasodilator metabolites would occur as a result of the decreased cerebral blood flow (subsequent to decreased per— fusion pressure). Possible metabolic chemicals which have been postulated as participating in local blood flow regu— lation in many vascular beds include oxygen, carbon dioxide, potassium, hydrogen, adenosine, adenine nucleotides, and intermediary metabolites (49,72,104) and most of these chemicals have been shown to be vasodilator in the cerebral vascular bed (72,116). The myogenic response is based on tne theory that the vascular smooth muscle is inherently res man cau vas ver ism the tall Spo anc the six cer inv not 3S 4111: Sta How. Geri 01m lino-h 90 responsive to stretch or changes in tension (6,37). In this manner, a decrease in transmural pressure within the vessel causes a decreased wall tension and hence relaxation of vascular smooth muscle, resulting in vasodilation (and vice versa). Thus either cu: both of these two general mechan— isms, metabolic and myogenic, may be important in producing the initial decreased cerebral vascular resistance concomi— tant with the decreased cerebral perfusion pressure. These studies show that the long—term resistance re- sponse of the cerebral vascular bed is an increase. Resist- ance increased with all three doses of endotoxin by at least the fourth hour of shock, and was maintained through the sixth hour post-endotoxin in the extended experiments. Resistance increased in the face of a rising arterial (and thus transmural) pressure, indicating an active decrease in cerebral vessel calibre, although passive factors could be involved, as will be discussed later. This response was not simply a loss of the cerebral autoregulatory response, as the cerebral vascular bed was at least as capable of autoregulation after 4 hours of shock as during the control state prior to endotoxin administration (Figure 11). However, a theoretical possibility of an altered but intact cerebral autoregulatory response around a lowered cerebral blood flow cannot be excluded. The mechanism involved in the resistance rise is un- known. This effect, however, is probably due to some 91 biological response of the animal to the endotoxin rather than a direct effect of the endotoxin upon the vasculature. It is known that endotoxin administration causes the release of vasoactive substances (27,31,83,123). In this regard, the release of histamine (61,119), kinins (123), and cate- cholamines (50,109) have been reported to occur during endo— toxin shock. In general, however, the cerebral vascular bed has been shown to be poorly responsive to most of these chemicals, including catecholamines (97), although this is highly controversial. Vasoconstrictor prostaglandins are another possibility, as they have been suggested as being involved in the local regulation of cerebral blood flow under certain conditions (126) and have been shown to be re- leased during endotoxin shock (3). However, Emerson gt_§l. (32,33) recently demonstrated little or no vasoconstrictor effects of PGFZaor PGEl in the cerebral vasculature. Serotonin has also been reported to significantly decrease carotid artery blood flow (46) and produce vasoconstriction of pial vessels when applied topically to the surface of the brain (98). However, Davis et_al. (22) have shown that the amounts of serotonin released during endotoxin shock are small and occur only within the first 5 minutes. The time course in the present experiments for the development of the increased cerebral vascular resistance was 3 to 4 hours, and so serotonin does not appear to be a likely candidate for causing the increased resistance. Thus, although 92 certainly a possibility, it is difficult to define a speci- fic vasoactive agent responsible for the long term increase in resistance observed during endotoxin shock, and, in addi- tion, if such a substance is responsible, it does not decrease or abolish the cerebral autoregulatory or venous— arteriolar responses. Changes in plasma osmolality have been shown to affect vascular resistance and blood flow (41,105). Also marked hypoglycemia has been reported to occur late in shock (36, 65), and theoretically, lowered plasma osmolality due to decreases in plasma glucose levels might increase cerebral vascular resistance. However, measurements of osmolality changes in dogs (121) demonstrate a slight hyperosmolality during the third and fourth hours of endotoxin shock, and cannot explain the long-term increase in cerebral vascular resistance. The cerebral vascular bed has been shown to be highly responsive to changes in arterial PCO levels. Thus, in 2 theory, lowered P levels might cause the cerebral vaso— CO2 constriction. Indeed, arterial PCO was decreased follow- 2 ing administration of l and 2 mg/kg, and in particular, during the period of increased resistance. However, 3 and 4 hours following administration of the 5 mg/kg dose, cerebrovascular resistance was increased in spite of the normocapnia occurring over the same period. In addition, even in the l and 2 mg/kg doses (Figures 15 and 17), the 93 arterial PCO levels were constant or tending to increase 2 over the period in which the resistance was increasing. Thus, although the arterial levels of P may play a role CO2 in increasing the cerebral vascular resistance, it is not a necessary one. Although the absolute values of PCO were not decreas- 2 ing in concert with the period of increasing vascular resistance, an alteration of the vascular responsiveness to any given level of PCO could have theoretically caused an 2 increase in the cerebral vascular resistance. Indeed, the cerebral vascular response to a CO challenge was clearly 2 depressed from 1 to 6 hours after administration of endo— toxin. The initial response lowered resistance to 37% of control. After 1, 2, 3, and 4 hours of shock cerebral vascular resistance only decreased to 80, 79, and 75% of control, respectively, during increased arterial PCO levels. 2 PCO arterial levels were measured during the pre-endotoxin 2 and 4-hour CO2 challenge, and although the absolute value of the PCO2 during CO2 of shock (Figure 13), this difference was not significant. inhalation was lower after 4 hours In fact, the experiment which demonstrated the greatest decrease in the CO2 re3ponse after endotoxin also demon- strated a higher arterial PCO during CO inhalation after 2 2 4 hours of shock. Thus, the diminished cerebral reactivity to C02 does not appear to be due simply to a difference in the arterial PCO levels at these times. Buyniski et_§l, 2 94 (15) also reported a reduction in the cerebral vascular response to C02, although the response was only studied for one hour post-endotoxin, and occurred when cerebral re— sistance was well below control. Clinical reports have also indicated reduced reactivity of cerebral vessels to hypercarbia in patients with occlusive cerebrovascular disease (35,81,82). However, these authors suggested that this observation is based on the theory that vasodilation is limited due to existing partial or maximal vasodilation in the ischemic areas. This explanation is not applicable to the loss of response to hypercarbia during endotoxin shock, as in the current study, the dilatory response to the CO2 challenge is depressed in spite of normal or in- creased resistance. Although these studies demonstrate a depressed cerebro- vascular response to CO during endotoxin shock, the criti- 2 cal question, of the importance of this effect in determin- ing the long term increase in resistance, is uncertain and subject to several considerations. For example, the diminished responsiveness to CO was well-established by the 2 first and second hours of shock, times at which the resist- ance was not increased above control. In addition, certain levels of hypotension have been shown to abolish or decrease the cerebral vascular responsiveness to CO (52). This may 2 play a role in the present experiments, as systemic pressure, and thus cerebral perfusion, was decreased during the entire 95 experimental period, eSpecially during the early hours. However, again, the explanation generally given for this effect is prior dilation, and in the 3rd and 4th hours post- endotoxin the CO2 response was diminished despite an ele- vated vascular resistance. The possibility that another mechanism, other than prior dilation, is involved in the CO2 response alterations during systemic hypotension cannot be excluded. At any rate, hypotension does not explain the continued depression of the response during the period of increasing systemic pressure. A passive decrease in cerebral vessel calibre could theoretically occur if alterations in the capillary membrane caused an increase in the capillary permeability to protein. In turn, increased protein concentrations in the intersti- tial fluid would promote fluid filtration and an increase in tissue pressure, resulting in a decreased transmural pressure gradient across the cerebral vascular wall. Increased tissue pressure in the brain would probably be reflected by an increased cerebrospinal fluid pressure. Indeed, cerebrospinal fluid pressure increased and cerebral vascular resistance increased in the 2 mg/kg dose. It seems, though, that an increase in cerebral tissue pressure would also impair the cerebral autoregulatory response, when, in fact, this did not occur. In addition, resistance increased intimaS mg/kg dose in spite of a normal cerebro- spinal fluid pressure. Also, cerebral blood flow has been 96 shown to be little affected by large increases(40 mm Hg) in cerebrospinal fluid pressure (4,112). However, one cannot exclude the possibility that an increase in brain tissue pressure did occur, and was not accurately represented by the cerebrospinal fluid pressure. Perhaps measurements of intracranial pressure, such as those described (79) using sagittal sinus wedge pressures or intracranial balloons would be of value in testing this hypothesis. The development of microemboli in the cerebral vascular bed must also be considered as a possible means of increas- ing cerebral vascular resistance during shock. Indeed, these emboli have been shown to develop in other vascular beds due to endotoxin shock (21,51). Anticoagulants have also been suggested as reducing the lethality of shock (51, 123), and may thus be an indirect indication of the basis of the results obtained in this study. To test this hypothe- sis, further studies might be completed in this preparation, using the increased levels of anticoagulants reported in these studies. In summary, the mechanism of the increased cerebral vascular resistance observed during endotoxin shock is un— certain, and may involve any one or a combination of the considerations listed in the preceding paragraphs. Despite the mechanism involved in this response, the elevated resistance combines with the lowered perfusion pressure to produce cerebral ischemia. 97 Systemic arterial blood gas changes during endotoxin shock included an increased PO , decreased PCO , and an 2 2 early decrease in pH, except in the 1 mg/kg dose, in which no alteration in pH was noted. The increases in arterial PO and decreases in PCO were usually associated with an 2 2 increased respiratory rate (Figure 15). Apparently, the hyperventilation and hypocapnia did not compose an adequate compensation initially, indicated by the early acidosis in the 2 mg/kg and 5 mg/kg doses. Previous investigators have noted the development of a metabolic acidosis during endo- toxin shock (28,29). However, reports in unanesthetized dogs (93) indicate that an adequate respiratory compensation results in normal or elevated arterial pH during endotoxin shock. Cerebral venous blood gases have not been previously reported. These included decreased PO , decreased pH, and 2 no significant change (but a tendency to increase) in the PC02' The sagittal venous system, as it accounts for most of the venous drainage of the cerebral hemispheres in the dog (108),rapid1y reflects cerebral tissue respiratory or metabolic acidosis, and in turn, the increased arterial- venous differences accompanying reductions in blood flow. In this manner, brain tissue acid-base balance, reflected in cerebral venous (sagittal sinus) pH and bicarbonate concentrations, demonstrated increased acidosis following administration of 2 mg/kg endotoxin, as venous pH decreased significantly and bicarbonate concentrations tended to 98 decrease. In addition, calculated arterial-venous differ- ences of oxygen saturation increased greatly (Figure 18) following endotoxin administration. If hemoglobin concen- tration is assumed not to decrease (in fact it probably increased due to increased hematocrit (18) observed by other authors), one can also infer from this graph that the arterial-venous difference of oxygen content increased dramatically, also. The alterations of cerebral hemodynamics, cerebral vascular reactivity, and arterial and cerebral venous blood gases described in this study are possible indications of the degree of cerebral ischemia occurring during endotoxin shock in the dog. Shock has been described as "inadequate tissue perfusion", and the central nervous system has been shown to be most sensitive to such deprivation (14,85). The function of the central nervous system during shock, therefore, is highly dependent upon the state of the vascu- lar nutritional supply, and vascular changes in hemodynamics or reactivity may subsequently play a role in the function and response of the nervous system following administration of endotoxin, and in turn, the pathogenesis of the shock state. CHAPTER VI SUMMARY AND CONCLUSIONS The purpose of this study was to determine the cerebral hemodynamics, vascular reactivity, and cerebral metabolic alterations following intravenous endotoxin administration in the dog. Cerebral venous outflow was measured directly from the cannulated confluence of the sagittal, straight, and transverse (lateral) sinuses with the transverse sinuses occluded (94), thus eliminating the primary communications of the cerebral venous outflow with the extracranial circu- lation. Purified E, coli endotoxin was administered in three doses, 1, 2, and 5 mg/kg, to determine if the effects were dose dependent over these doses. Cerebral blood flow and cerebral perfusion pressure decreased immediately following administration of all three doses of endotoxin, and remained significantly below control during the entire experimental period. By the fourth hour of shock, cerebral blood flow was decreased 37, 48, and 45% in the l, 2, and 5 mg/kg doses, respectively. Cerebral vascular resistance transiently decreased, returned to con- trol levels, and progressively rose to levels significantly 99 100 above control by minutes 150 to 240, depending on the dose administered. By four hours post-endotoxin, cerebral vascu- lar resistance was 25, 55, and 37% above control in the l, 2, and 5 mg/kg doses, respectively. Thus, the cerebral vascular resistance response to endotoxin was biphasic: 1) an initial decrease, and 2) a long-term increase. The mechanism causing the increase in resistence is unknown, but various possibilities discussed were: 1) circulating vasoactive agents, 2) alterations in vascular responsiveness to carbon dioxide, 3) passive changes via increases in tissue pressure, and 4) development of microemboli in the cerebral vasculature. In summary, cerebral blood flows decreased following endotoxin administration, due initially to the decreased perfusion pressure and subsequently, primarily to increased cerebral vascular resistance. Vascular reactivity studies indicated that the auto- regulatory and "venous-arteriolar" responses were well maintained after four hours of endotoxin shock. The cere- bral vascular responsiveness to carbon dioxide, however, was depressed during shock. Endotoxin administration generally caused an increased respiratory activity, resulting in increased PO and de- 2 creased PCO in the systemic arterial blood. Initially, 2 the hyperventilation did not compose an adequate compensa— tion, indicated by the early acidosis in the 2 mg/kg and 5 mg/kg doses. Cerebral venous blood gas changes include 101 decreased PO and pH, and increased arterial-venous differ— 2 ences of percent oxygen saturation, total CO2 and HCOB- concentrations. Thus, brain tissue and acid-base balance, reflected in cerebral venous (sagittal sinus) pH and bi- carbonate concentrations, became more acidotic during endotoxin shock. BIBLIOGRAPHY 10. BIBLIOGRAPHY Alican, F. M., M. L. Dalton, Jr., and J. D. Hardy. Experimental endotoxin shock. Am. J. Surg. 103:702, 1962. Altemeier, W. A., J. C. Todd, and W. W. Inge. Gram- negative septicemia: A growing threat. Ann. Surg. 166:530, 1967. Anderson, F. L., W. Jubiz, T. J. Tsagaris, and H. Kuida. Endotoxin-induced prostaglandin E and F release in dogs. Am. J. Physiol. 228:410, 1975. Avhandling, A. Effects of increased cerebrospinal fluid pressure on the blood flow and on the energy metabolism of the brain. Acta Physiol. Scand. Suppl. 339:1, 1970. Ayres, S. M., H. Mueller, 3. Giannelli, Jr., P. Fleming, and W. J. Grace. The lung in shock. Am. J. Cardiol. 26:588, 1970. Bayliss, W. M. On the local reactions of the arterial wall to changes in internal pressure. J. Physiol. (London) 28:220, 1902. Becn-Jansen, P., C. R. Brinkman III, G. H. Johnson, and N. S. Assali. Circulatory shock in pregnant sheep. Am. J. Obstet. Gynecol. 113:37, 1972. Berman, I. R. The lung lesion in shock. In: The Funda— mental mechanisms of Shock, Edited by L. B. HihShaw and B. G. Cox. New York, Plenum Press, 1972. Betz, E. Cerebral blood flow: Its measurement and regulation. Physiol. Rev. 52:595, 1972. Blattberg, B. and M. N. Levy. Extrahepatic factors in the early hypotensive response to bacterial endotoxin. Can. J. Physiol. Pharmacol. 48:583, 1970. 102 11. 12. 13. 14. 15. 16. 17. 18. 19. 20. 21. 22. 103 Blattberg, B. and M. N. Levy. Sites of extrahepatic pooling in early response to endotoxin. Am. J. Physiol. 220:1267: 1971. Bohs, C. T., M. E. Turbow, and S. N. Kolmen. Improve- ment of coronary blood flow in endotoxin shock without improvement of myocardial contractility. Physiologist 16:268, 1973. Brand, E. D. and A. M. Lefer. Myocardial depressant factor in plasma from cats in irreversible post- oligemic shock. Proc. Soc. Exptl. Biol. Med. 122:220, 1966. Brown, R. S., P. A. Mohr, and W. C. Shoemaker. Effect of cerebral hypotension on the neural regulation of the cardiovascular system. Surg. Gynecol. Obstet. 131:436, 1970. Buyniski, J. P., M. Smith, and M. E. Bierwagen. Effect of Gram—negative endotoxin on cerebral circulation. Europ. Neurol. 6:253, 1971/1972. Campbell, C. D. and L. C. Carey. Shock-Differential diagnosis and immediate treatment. Postgrad. Med. 55:85, 1974. Charrin, A. and E. Gley. Recherches expérimentales sur l'action des produits sécretes par la Bacille pyo- cyanique. Arcn. de Physiol. S. 2:724, 1890. Chien, 8., C. Chang, R. J. Dellenback, S. Usami, and M. I. Gregersen. Hemodynamic changes in endotoxin shock. Am. J. Physiol. 210:1401, 1966. Chien, S., R. J. Dellenback, S. Usami, K. Treitel, C. Chang, and M. I. Gregersen. Blood volume and its distribution in endotoxin shock. Am. J. Physiol. 210:1411, 1966. Chien, S., D. G. Sinclair, R. J. Dellenback, C. Chang, B. Peric, S. Usami, and M. I. Gregersen. Effect of endotoxin upon capillary permeability to macromole- cules. Am. J. Physiol. 207:518, 1964. Corrigan, J. J. and B. M. Bell. Endotoxin—induced . intravascular coagulation: Prevention with Polymyxin B Sulfate. J. Lab. Clin. Med. 77:802, 1971. Davis, R. B., W. R. Meeker, and W. L. Bailey. Serotonin release by bacterial endotoxin. Proc. Soc. Exptl. 23. 24. 25. 27. 28. 29. 30. 31. 32. 33. 104 Difco, Technical information. Lipopolysaccharides- Bacterial. 1973. Dobbins, D. E. Centrally mediated cardiovascular effects of systemically administered E. coli endo toxin. Ph.D. Dissertation, Michigan State University, 1975. Donald, W. D., C. H. Winkler, and K. Hare. Investiga- tions on the mechanism of the leukopenic response to Shigella toxin. J. Clin. Invest. 37:1100, 1958. Elkins, R. C., J. R. McCurdy, P. P. Brown, and L. J. Greenfield. Effects of coronary perfusion pressure on myocardial performance during endotoxin shock. Surg. Gynecol. Obstet. 137:991, 1973. Emerson, T. E., Jr. Participation of endogenous vaso- active 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. Emerson, T. E., Jr. and C. C. Gill. Effect of slow intravenous infusion of endotoxin in dogs. J. Appl. Physiol. 22:874, 1967. Emerson, T. B., Jr. and K. C. Kelly. Effect of living E. coli cells on hemodynamics and mortality in the dog. J. Appl. Physiol. 23:609, 1967 Emerson, T. E., Jr. and J. L. Parker. Cerebral hemo- dynamics during endotoxin shock in the dog. Physiol- ogist 15:125, 1972. Emerson, T. E., Jr. and J. L. Parker. Cerebral hemo- dynamics and endotoxin shock in the dog. Circ. Shock. In Press, 1975. Emerson, T. B., Jr., D. Radawski, J. C. Havran, and R. M. Daugherty, Jr. Comparison of the effects of alcohol vs. water base prostaglandin E on cerebral hemodynamics in the dog. Clin. Res. 22:594A, 1974. Emerson, T. E., Jr., D. Radawski, M. Veenendaal, and R. M. Daugherty, Jr. Cerebral hemodynamic effects of prostaglandin (PG) F2“ in the dog. Physiologist 17:217, 1974. 34. 35. 36. 37. 38. 39. 4o. 41. 42. 43. 44. 105 Emerson, T. E., Jr., T. P. Wagner, and C. C. Gill. In situ pressure—flow-resistance relationship during endotoxin shock in dogs. Proc. Soc. Exptl. Biol. Med. 122:366, 1966. Fieschi, C., A. Agnoli, and E. Galbo. Effects of carbon dioxide on cerebral hemodynamics in normal subjects and in cerebrovascular disease studied by carotid injection of radioalbumin. Circ. Res. 13: 436, 1963. Filkins, J. P. and R. P. Cornell. Depression of hepatic gluconeogenesis and the hypoglycemia of endotoxin shock. Am. J. Physiol. 227:778, 1974. Folkow, B. Intravascular pressure as a factor regulat- ing the tone of the small vessels. Acta Physiol. Scand. 17:289, 1949. Frohlich, E. D., J. B. Scott, and E. S. Dooley. Hemo- dynamic alterations due to Salmonella typhosa endo- toxin with special reference to the coronary vascular bed. J. Clin. Invest. 41:147, 1962. Gallo, E. and W. G. Schenk. Endotoxin shock: Altera— tions in cardiac output and regional blood flow with modifications resulting from whole blood transfusion with and without ganglionic blockade. J. Trauma 7:667, 1967. Gallone, L. Problems of clinical research in shock. In: Shock: Biochemical, Pharmacological, and Clini— cal Aspects, Edited by A. Bertelli and N. Back. New York, Plenum Press, 1970. Gazitua, S., J. B. Scott, B. Swindall, and F. J. Haddy. Resistance responses to local changes in plasma osmolality in three vascular beds. Am. J. Physiol. 220:384, 1971. Gilbert, R. P. Mechanisms of hemodynamic effects of endotoxin. Physiol. Rev. 40:245, 1960. Gillenwater, J. Y., E. S. Dooley, and E. D. Frohlich. Effects of endotoxin on renal function and hemo~ dynamics. Am. J. Physiol. 205:293, 1963. Green, H. D. and C. E. Rapela. Cerebral vascular re- sponses to localized and systemic hypotension induced by hemorrhage and shock. In: Microcirculation as Related to Shock, New York, Academic press, 1968. 45. 46. 47. 48. 49. 50. 51. 52. 53. 54. 55. 106 Grega, G. J. and F. J. Haddy. Transvascular fluid fluxes during prolonged canine hemorrhage or endotoxin hypotension. In: Neurohumoral and Metabolic Aspects of Injugy, Edited by H. A. Arisztid, G. B. Kovach, H. B. Stoner and J. J. Spitzer. New York, Plenum Press, 1972. Grimson, B. S., S. C. Robinson, E. T. Danford, G. T. Tindall and J. C. Greenfield, Jr. Effect of serotonin on internal and external carotid artery blood flow in the baboon. Am. J. Physiol. 216:50, 1969. Guntheroth, W. G. and A. J. Felsenfeld. Splenic con- striction with endotoxin shock in the dog. J. Appl. Physiol. 30:517, 1971. Haddy, F. J. and J. B. Scott. Cardiovascular pharmacol- oby. Ann. Rev. Pharmacol. 6:49, 1966. Haddy, F. J. and J. B. Scott. Metabolically linked vasoactive chemicals in local regulation of blood flow. Physiol. Rev. 48:688, 1968. Hall, R. C. and R. L. Hodge. Vasoactive hormones in endotoxin shock: A comparative study in cats and dogs. J. Physiol. (London) 213:69: 1971- Hardaway, R. M., E. A. Husni, E. F. Geever, H. E. Noyes, and J. W. Burns. Endotoxin shock. A manifestation of intravascular coagulation. Ann. Surg. 154:791, 1961. Harper, A. M. and H. I. Glass. Effect of alterations in the arterial carbon dioxide tension on the blood flow through the cerebral cortex at normal and low arterial blood pressures. J. Neurol. Neurosurg. Psychiat. 28:449, 1965. Hinshaw, L. B. Mechanisms and therapy of endotoxin shock (Shock Tour Symposium). J. Okla. State Med. Assoc. 59:407, 1966. Hinshaw, L. B., L. T. Archer, M. R. Black, R. C. Elkins, P. P. Brown, and L. J. Greenfield. Myocardial func- tion in shock. Am. J. Physiol. 226:357, 1974. Hinshaw, L. B., L. T. Archer, L. J. Greenfield, and C. A. Guenter. Effects of endotoxin on myocardial hemodynamics, performance and metabolism. Am. J. Physiol. 221:504, 1971. 56. 57. 58. 59. 60. 61. 62. 63. 64. 65. 66. 107 Hinshaw, L. B., L. T. Archer, L. J. Greenfield, J. A. Miller, and C. A. Guenter. Effect of endotoxin on myocardial performance. J. Trauma 12:1056, 1972. Hinshaw, L. B., L. T. Archer, J. J. Spitzer, M. R. Black, M. D. Peyton, and L. J. Greenfield. Effects of coronary hypotension and endotoxin on myocardial performance. Am. J. Physiol. 227:1051, 1974. Hinshaw, L. B., C. M. Brake, T. E. Emerson, Jr., M. M. Jordan, and F. D. Masucci. Participation of sympatho- adrenal system in endotoxin shock. Am. J. Physiol., 207(4):925, 1964. Hinshaw, L. B., T. E. Emerson, Jr., and D. A. Reins. Cardiovascular responses of the primate in endotoxin shock. Am. J. Physiol. 210:335, 1966. Hinshaw, L. B., L. J. Greenfield, S. E. Owne, L. T. Archer, and C. A. Guenter. Cardiac response to circulating factors in endotoxin shock. Am. J. Physiol. 222:1047, 1972. Hinshaw, L. B., M. M. Jordan, and J. A. Vick. Histamine release and endotoxin shock in the primate. J. Clin. Invest. 40:1631, 1961. Hinshaw, L. B., H. Kuida, R. P. Gilbert, and M. G. Visscher. Influence of perfusate characteristics on pulmonary vascular response to endotoxin. Am. J. Physiol. 191:293, 1957. Hinshaw, L. B. and D. L. Nelson. Venous response of the intestine to endotoxin. Am. J. Physiol. 203:870, 1962. Hinshaw, L. B. and S. E. Owens. Correlation of pooling and resistance changes in the canine forelimb in septic shock. J. Appl. Physiol., 30:331, 1971. Hinshaw, L. B., M. D. Peyton, L. T. Archer, M. R. Black, J. J. Coalson, and L. J. Greenfield. Prevention of death in endotoxin shock by glucose administration. Surg. Gynecol. Obstet. 139:851, 1974. Hinshaw, L. B., L. L. Shanbour, L. J. Greenfield, and J. J. Coalson. Mechanism of decreased venous return in subhuman primate administered endotoxin. Arch. Surg. 100:600. 1970. 67. 68. 69. 70. 71. 72. 73. 74. 75. 76. 77. 78. 108 Hinshaw, L. B., J. A. Vick, M. M. Jordan, and L. E. Wittmers. Vascular changes associated with develop- ment of irreversible endotoxin shock. Am. J. Physiol. 202:103, 1962. Horwitz, D. L., T. V. N. Ballantine, and C. M. Herman. Acute effects of septic shock on plasma and red cell volumes in baboons. J. Appl. Physiol. 33:320, 1972. Kety, S. S. and C. F. Schmidt. The nitrous oxide method for the quantitative determination of cerebral blood flow in man: theory, procedure, and normal values. J. Clin. Invest. 27:476, 1948. Kety, S. S. and C. F. Schmidt. The effect of altered arterial tensions of carbon dioxide and oxygen on cerebral blood flow and cerebral oxygen consumption of normal young men. J. Clin. Invest. 27:484, 1948. Kitzmiller, J. L., W. E. Lucas, and D. F. Yelenosky. The role of complement in feline endotoxin shock. Am. J. Obstet. Gynecol. 112:414, 1972. Kontos, H. Mechanisms of regulation of the cerebral microcirculation. Curr. Con. Cerebrovasc. Dis. (Stroke) 10:7, 1975. Kovach, A. G. B. Recent progress and present problems in the field of shock (conference proceedings) Fed. Proc. 20:125, 1961. Kuida, H., R. P. Gilbert, and L. B. Hinshaw. Species differences in effect of Gram—negative endotoxin on circulation. Am. J. Physiol. 200:1197, 1961. Lefer, A. M. Role of a myocardial depressant factor in the pathogenesis of circulatory shock. Fed. Proc. 29:1836, 1970. Lefer, A. M. Role of a myocardial depressant factor in shock states. Modern Con. Cardiovasc. Dis. 42: 59, 1973 Lefer, A. M. and J. J. Martin. Origin of myocardial depressant factor in shock. Am. J. Physiol. 218: 1423, 1970. Lewis, A. E. Biostatistics. New York, Reinhold Pub- lishing Corp., 1966. 79. 80. 81. 82. 83. 84. 85. 86. 87. 88. 109 Lowell, H. M. and B. M. Bloor. The effect of increased intracranial pressure on cerebrovascular hemodynamics. J. Neurosurg. 34:760, 1971. MacLean, L. D., W. G. Mulligan, A. P. H. McLean, and J. H. Duff. Patterns of septic shock in man--A de- tailed study of 56 patients. Ann. Surg. 166:543, 1967. Mamo, H., J. Seyloz, A. Rey, and R. Houdart. A patho— physiological approach to altered CO effect on cere- bral blood flow in post-acute ischemia in man. Europ. Neurol. 6:306, 1972. McHenery, L. C., H. I. Goldberg, M. E. Jaffe, E. J. Kenton, III, J. W. West, and E. S. Cooper. Regional cerebral blood flow response to carbon dioxide in- halation in cerebrovascular disease. Arch. Neurol. 27:403, 1972. Miller, R. L., R. P. Forsyth, B. I. Hoffbrand, and K. L. Melmon. Cardiovascular effects of hemorrhage during endotoxemia in unanesthetized monkeys. Am. J. Physiol. 224:1087, 1973. Moore, F. D. Recent progress and present problems in the field of shock. Fed. Proc. Suppl. 20:9, 1961. Moss, G., C. Staunton, and A. A. Stein. Cerebral etiology of the "shock lung" syndrome. J. Trauma 12:885, 1972. Neuhof, H., E. Glaser, D. Hey, and H. G. Lasch. Patho- physiologic mechanisms in endotoxin shock and its therapeutic approaches. In: Shock: Biochemical, Pharmacological and Clinical Aspects, Edited by A. Bertelli and N. Back. New York, Plenum Press, 1970. Palmerio, C., B. Zefferstrom, J. Shammash, E. Euchbaum, E. Frank, and J. Fine. Denervation of the abdominal viscera for the treatment of traumatic shock. New Eng. J. Med. 269:709, 1963. Parker, J. L. and T. E. Emerson, Jr. Vascular reactiv- ity and metabolic alterations in shock. Physiologist 16:417, 1973. 89. 90. 91. 92. 93. 94. 95. 96. 97. 98. 99. 100. 110 Parker, P. E., D. E. Dobbins, W. J. Weidner, F. J. Haddy and G. J. Grega. Effects of hemorrhagic, endotoxin, and catecholamine shocks on canine gracilis muscle vasculature. Proc. Soc. Exptl. Biol. Med. 138:971, 1971. Penner, A. and A. I. Bernheim. Studies in the patho- genesis of experimental dysentery intoxication. J. Exptl. Med. 111:145, 1960. Penner, A. and S. H. Klein. The pathogenesis of experi- mental dysenteric intoxication. J. Exptl. Med. 96: 59, 1952. Petersdorf, R. G. Septic shock. In: Principles of Internal Medicine, 7th ed., Vol. 1, p. 734, 1974. Priano, L. L., R. D. Wilson, and D. L. Traber. Cardio— respiratory alterations in unanesthetized dogs due to Gram-negative bacterial endotoxin. Am. J. Physiol. 220:705, 1971. Rapela, C. E. and H. D. Green. Autoregulation of canine cerebral blood flow. Circ. Res. 14815:I-205, 1964. Rapela, C. E., H. D. Green, and A. B. Denison. Baro— receptors and autoregulation of cerebral blood flow in the dog. Circ. Res. 21:559, 1967. Rao, P. S. and D. Cavanagh. Endotoxin shock in the sub- human primate. Arch. Surg. 102:486, 1971. Raper, A. J., H. A. Kontos, E. P. Wei, and J. L. Patterson, Jr. Unresponsiveness of pial precapillary vessels to catecholamines and sympathetic nerve stimu- lation. Circ. Res. 31:257, 1972. Raynor, R. B., J. G. McMurtry, and J. L. Pool. Cerebro- vascular effects of topically applied serotonin in the cat. Neurol. 11:190, 1961. Reeves, J. T., F. S. Daoud, and M. Estridge. Pulmonary hypertension caused by minute amounts of endotoxin in calves. J. Appl. Physiol. 33:739, 1972. Rein, H. Vasomotorische regulationem. Ergeb. Physiol. Biol. Chem. Exptl. Pharmakol. 32:28, 1931. 101. 102. 103. 104. 105. 106. 107. 108. 109. 110. 111. 111 Rittman, W. W. and L. L. Smith. Canine cerebral blood flow following hemorrhage. Effect of dextran, dibenzyline and blood replacement. Am. Surgeon 32: 695, 1966. Robb, H. J., R. R. Margulis, and C. M. Jabs. Role of pulmonary microembolism in the hemodynamics of endotoxin shock. Surg. Gynecol. Obstet. 135:777, 1972. Rosenberg, J. C., R. C. Lillehei, J. Longerbeam, and B. Zimmermann. Studies on hemorrhagic and endotoxin shock in relation to vasomotor changes and endogenous circulating epinephrine, norepinephrine and sero- tonin. Ann. Surg. 154:611, 1961. Rubio, R. and R. M. Berne. The release of adenosine by the normal myocardium in dogs and its relationship to the regulation of coronary resistance. Circ. Res. 25:407, 1969. Scott, J. B., M. Rudko, D. Radawski, and F. J. Haddy. Role of osmolarity, K+, H+, Mg++, and O2 in local blood flow regulation. Am. J. Physiol. 218:338, 1970. Selkurt, E. E. Status of investigative aspects of hemorrhagic shock. Fed. Proc. 29:1832, 1970. Selmeyer, J. P., D. G. Reynolds, and K. G. Swan. Renal blood flow during endotoxin shock in the sub- human primate. Surg. Gynecol. Obstet. 137:2, 1973. Shannon, D. C., N. Shore, and H. Kazemi. Acid-base balance in hemorrhagic cerebrospinal fluid. Neurol. 22:585, 1972. Shires, G. T., C. J. Carrico, and P. C. Canizaro. Septic shock. In: Shock. Philadelphia, W. B. Saunders, Co., 1973. Shoemaker, W. C. Sequential hemodynamic patterns in various patterns of shock. Surg. Gynecol. Obstet. 32:411, 1970. Shoemaker, W. C. Physiologic mechanisms in clinical shock. In: Fundamental Mechanisms of Shock. Edited by L. B. Hinshaw and B. G. Cox. New York, Plenum Press, 1972. 112. 113. 114. 115. 116. 117. 118. 119. 120. 121. 112 Shulman, K. and G. R. Verdier. Cerebral vascular resistance changes in response to cerebrospinal fluid pressure. Am. J. Physiol. 213:1084, 1967. Siegel, J. H. The recognition and therapy of patho- physiologic patterns in human shock. In: Shock: Biochemical, Pharmacological, and Clinical Aspects. Editéd by A} Bertelli and N. Back. New York} Plenum Press, 1970. Simmons, R. L., R. W. Anderson, T. B. Ducker, H. K. Sleeman, J. A. Collins and K. P. Boothman. The role of the central nervous system in septic shock: II. Hemodynamics, respiratory, and metabolic effects of intracisternal or intraventricular endotoxin. Ann. Surg. 167:158, 1968. Simmons, R. L., T. B. Ducker, A. M. Martin, Jr., R. W. Anderson, and H. E. Noyes. The role of the central nervous system in septic shock: I. Pathologic changes following inteaventricular and intracisternal endotoxin in the dog. Ann. Surg. 167:145, 1968. Toda, N. Responsiveness to potassium and calcium ions of isolated cerebral arteries. Am. J. Physiol. 229:1206, 1974. Trippodo, N. C., J. H. Jorgensen, L. L. Priano, and D. L. Traber. Cerebrospinal fluid levels of endo- toxin during endotoxemia. Proc. Soc. Exptl. Biol. Med. 143:932, 1973. Vaughn, D. L., C. A. Gunter, and J. L. Stookey. Endotoxin shock in primates. Surg. Gynecol. Obstet. 126:1309, 1968. Vick, J. A., B. Mehlman, and H. Heiffer. Early hist- amine release and death due to endotoxin. Proc. Soc. Exptl. Biol. Med. 137:902, 1971. 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. Weidner, W. J., G. J. Grega, and F. J. Haddy. Changes in forelimb weight and vascular resistances during endotoxin shock. Am. J. Physiol. 221:1229: 1971- 113 122. Weil, M. H., L. D. MacLean, M. B. Visscher, and W. W. Spink. Studies on the circulatory changes in the dog produced by endotoxin from Gram—negative micro— organisms. J. Clin. Invest. 35:1191, 1956. 123. Weiner, D. Effects of endotoxin on cerebral blood flow in the monkey. Am. J. Physiol. 218:160, 1970. 124. Wilson, R. F. ... Or biochemical blockade? Emergency Med. Oct:37, 1973. 125. Wilson, R. P., E. J. Sarver, and J. Rizzo. Hemody- namic changes, treatment, and prognosis in clinical shock. Arch. Surg. 102:21, 1971. 126. Yamamoto, Y. L., W. Feindel, L. S. Wolfe, H. Katoh, and C. P. Hodge. Experimental vasoconstriction of cerebral arteries by prostaglandins. J. Neurosurg. 37:385, 1972. 127. Zwadyk, P., Jr. and J. S. Snyder. Effect of endo- toxin on hepatic glycogen metabolism in vitro. Proc. Soc. Exptl. Biol. Med. 142:299, 1973. APPENDICES APPENDIX A STATISTICAL METHODS 114 APPENDIX A STATISTICAL METHODS The Student's t test for paired observations (75) was used for statistical analysis in these experiments. The experimental condition was preceded by a suitable control period, and therefore, each animal served as its own control. Since the experimental design thus permitted pairing of the data, the application of the null hypothesis implied that the mean of the differences of the paired measurements (universe mean) should be zero. The alternative to the null hypothesis, therefore, would be that the mean difference is not zero, i.e., the populations are different. This is stated as follows: = 0, HA : pd # 0 The probability of obtaining such a difference was calcu- 1ated, utilizing the Student's t test and tables, and confidence levels (usually 95%) were used to accept or reject the null hypothesis. To use the Student's t test, the control values were designated X and the experimental values designated X'. The sum of the differences, ID, was calculated, where D = (X - X'). The mean of this sample of differences, 5, and 115 116 its standard deviation (S) were used to calculate t as follows: D = 23D S _~/nZD2 - (23D)2 n _ n(n-l) :6 t = S/fi The probability of obtaining this mean difference was then found in the Student's t-table of values. If the signifi- cance level was not within the confidence limits (P 0.05) the null hypothesis was rejected and the alternative accepted. A computer program (in FOCAL language) was written incorporating these formulas, and calculations were then run on a PDP 8/E (Digital Equipment Corporation) high speed magnetic core memory unit. APPENDIX B RAW DATA 117 CEREBRAL PERFUSION PRESSURE 2 Ins/kg endotoxin Time (min) 15 30 6O 90 120 150 180 210 ZAO 10 D08# 118 O‘NOMOOOOOOOO 0 o o o I o o o o o o NNFWNMJ’F:\O\O\OU\ O‘B-fiFCOFnoaomfi Ov-OVDOOOOLHOOO . . . . C . C . C . . . MN (\0 (\u- d'l-QJO d‘O O‘BWPQF “00"" um FF" '- OOd’OOOOOt-ROOO O O I O O O. O O O C O MN no N BO RMMOV-fi O\L\\Ov-OOON®O\v-\OD- FNMJ’MWFQO‘OPN FF" 2 Ins/k5 endotoxin CEREBRAL BLOOD FLOW Time (min) 60 120 150 180 210 240 30 90 15 10 D08# 119 “\0 CNN '- (\mawwrm \Sno N ammmmmm: P—"p FFFFFF N\Mwou\mu3c~bu-U\oun 4M0 NKO d'B-Inwo “\d" v-v-v-v— v-v-v- “\Omdev—NOBON Md'Ov-KO Jul-“COO “\J' v—v-v-v- Q-v-v-NI-F- O d'N N O\O®U\U\~+d’0\ :m—mogmd-xo—m: FF—F FF—NF— JMKOd’CD O\\O\DO “4'0 O\O\O 4'0 {~00 “\d’nm éO‘Md’N O‘BéffigO‘a FF ...-P «>0meon “mN N #N '- 3M0 «VI-mo 5d; FMI-o- FP—P meowmi‘wNJNN éRMMLNU: N Nooooo O\ v-c-v-c-Nc-v- oomwémméNmNJ o o o o o o o o o o FNMd’lfixOBwO‘OF-N PF" 34.1 3.5 SEM Mean 120 Mac PM“) 411““ng O 0‘ O o: mmnmnawmammm m6 “\v- \0 “\v-ko JUV-AMO\ 9a (\ONKOIAONMKOM w-d' ogmv—OOMNIANBMON 41'?“ 0....0....00 \OMWQM-¢l\\OMU\JU\ “\O 01“ N v-\DNNO\\DOCD O‘KD 00‘ng JO (\— d‘KON \0 4' KO J'LHO‘MJLNMMIAJ'M “\O ammo om— mmoumguJR—Jm mo mm—Nw 0% mom m: w:mwm:omm¢:¢ mo of :n.o mm.¢ mw.n mm.¢ m—.4 mm. m:.m w—.m :o.n NN.N we. _M.¢ mo.m N—.w ON. L\ ON .0 ommm" 000 .0. O‘J‘PO‘MOCOPCOBM CO M g nmmwn-— O O mmmmm:m¢¢nnm :0 afiNOpouao mx\wa N MazdamHmmm mdaaom<> A4mmmmmo mnoo 0:.0 am... 9... m3 min 8... 8A Ed 3d —o.m mw.¢ RA 3... HR 93 .5... 8R ON.N M.N Rd N. mw.¢ N—. 8.5 $3 Rd N. 4. 8 on 33 3: bio mo.m mm.~ om.~ mn.n om.¢ mm.~ :m.n mm.& m”. m_.n mm.. m_.m 3 :—mmmn¢émmm¢ :0 OCOv-Ov-v-OBIAQRO‘ a? F 0 GM!- «Dd-memo 0 3k '- N "\d'lfim (\CO 0 H) CEREBROSPINAL FLUID PRESSURE 2 tug/kg endotoxin Time (min) 15 30 6O 90 120 150 180 210 ZAO IO D°8# 121 CO OI-nI-nOI-nI-nOOI-n C C C C C C C C C C C Gun ”NI-nInNWNNF' N.- M .- mo Olnlnoooomo 0 0 0“; 0':\o0nI.—0F: so an a- _ In oomomoomo I L\ IQInKHNCOInR—O ooooooomm 6m:mmo:mmmnm '- N on ouxooooooo 0 . . . . v-l\ £100wa :nn OO omoomooom 0 0 . 0 0 0 0 0 0 0 0 NN O\InOInIan\InO .- 0- 1-1- 0- I “)9 oooomoooo COO 84.":BNO‘I-nd'v': OO ooouwxoomu'xo . . 0 . . . 0 0 0 OO oomomomoo . 0| 0 . 0 . . . . . \ON l\Ot\v-.-d'c- «3mm? 00 mommmommo . . 0 0 0 0 0 0 0 InI-n I-nI-nI-nOInI-QInInO FNln-d’InwLNCDO‘OC-N pp— 122 0 ewswmmmeemsa 9N '- m m m m— m— m— mn an mm ON :— JN m m m mm mm. @— NN :N o— —N w. .J— pm —N m— mm 9 mm m.— m. m.— m— m. m.— w m +1 0— o— 0 0m 00 on AGfiav mafia 333nm mia m ma q¢mmmmMU —.—N N.¢n o.m— m.m— mum" o.mm om n.n n.m m._m ..am m.mm o.mm :.:_ m.m_ n.m_ m. _ m.m_ :.m~ o.n~ ..m_ m.m~ m.m_ om on Aaaav cede ufiNOpouao wx\wa m Bonk Gooqm Admmmmmo d’lN PQNG F4.” 0 BjMMMm . 0 0 0 0 MMMd'v- N no MN F FM 0 o o a: OOWOMOO 0 0 0 0 Pv-Nv-d’N '- \OPMJ’OB LR m.— N.w— :.¢N o.—N m.m. m.m— 0.4— n.5— O— o “‘4'“! “PM no 88 $8355 RR m 0 0 o In c-l\\OO\Ov- CO!- OOOOOQN coo w.m u.mn 00 em. m.m~ o.m¢ m._n m.wn 2m cu m FNMJ‘UWW m 0: zmm and to —NN\.1-U'\\o oz CEREBROSPINAL FLUID PRESSURE 5 mg/kg endo toxin Time (min) 15 30 60 90 120 150 180 210 ZhO 10 O DOB # omomoo ... \OOMO 4'0 FPF 0ln0000 mom—mo -..-F. lnl-nO Lnl-nO \o—onm— FI-I— . anOI-noo \Ov—O'fiJO FF" tmmnooo “\00000 [\N N me— oomooo 000000 000000 «SI-nd‘anv: 00|n000 \BQ Income wanna-um) 128 RESPIRATORY RATE 5 nag/kg endotoxin Time (min) 15 30 60 90 120 150 180 210 240 10 0 Dog # BWO‘MO‘Q '- M MOW-04'"- FF" d‘MO‘NBd‘ r-r- .— '- coma-mom '- "NM-31““) SEM Mean Dog # O 1 95 2 96 3 90 k 95 5 83 6 82 Mean 90 SEM 2.6 D08 # o 1 #6 2 h? 3 #9 h 40 5 39 6 #7 Mean #5 SEM 1.7 30 99 105 92 109 85 97 3.6 60 101 97 Ill 9h 85 98 3-5 3.3 129 ARTERIAL P02 5 mg/kg endotoxin Time (min) 90 120 150 180 98 101 10? 102 109 107 110 93 109 108 106 106 111 113 115 1 82 79 7 7 76 83 6 73 98 99 97 9# 6.2 5.8 8.3 6.1 VENOUS PC 2 5 mg/kg endotoxin 90 120 150 180 #0 36 55 33 31 55 33 36 51 32 3h 52 50 50 #9 53 210 240 104 107 115 114 103 95 114. 109 53 53 60 53 92 89 11.3 11.5 210 ago 3h 35 43 39 38 41 3h 37 20 20 35 47 3h 37 3.1 3.7 130 ARTERIAL P C02 5 mg/kg endotoxin Time (min) 90 120 150 180 210 30 20 20 21+ 22 21+ 21 24 32 23 26 25 27 26 28 25 33 36 3# 44 1 8 20 51 #7 65 67 77 29 28 32 35 36 6.6 #.# 7.2 6.7 9-1 5 mg/kg endotoxin Time (min) 90 120 150 180 210 43 #4 #2 Ah 43 3o 41 41 A3 #3 u3 44 43 49 as 48 52 57 56 7o 36 39 33 ha 39 62 so 70 7s 82 #7 47 #8 51 Sh 3.6 3-2 5.3 5-2 7.3 131 ARTERIAL pH 5 mg/kg endotoxin Time (min) Dog # 0 30 6O 90 120 150 180 210 240 1 7-3h 7.29 7.30 7.28 7.3h 7.37 7.33 7.35 7.37 2 7.37 7.23 7.26 7.28 7.34 7.33 7.23 7.32 7.34 3 7.31 7.28 7.32 7.35 7.37 7.35 7.35 7.33 7.33 4 7.35 7.3# 7.33 7.36 7.38 7.40 7.31 7.39 7.35 5 7.36 7.21 7.20 7.15 7.15 7.15 7.18 7.10 7.10 6 7.26 7.20 7.15 7.10 7.08 7.05 7.03 6.99 6.95 Mean 7.33 7.26 7.26 7.25 7.27 7.28 7.24 7.25 7-2h SEM 0.08 0.03 0.08 0.06 0.08 0.05 0.05 0.08 0.07 VENOUS pH 5 mg/kg endotoxin Time (min) Dog # 0 3o 60 90 120 150 180 210 240 1 7.27 7.21 7.23 7.22 7.23 7.2# 7.23 7.2h 7.2# 2 7.32 7.18 7.17 7.19 7.22 7.23 7.19 7.21 7.26 3 7.26 7.21 7.25 7.29 7.28 7.28 7.25 7.29 7.26 h 7.27 7.26 7.24 7.30 7.27 7.32 7.24 7.28 7.26 5 7.29 7.13 7.10 7.09 7.08 7.07 7.07 6.95 6.95 6 7.22 7.15 7.12 7.06 7.04 7.00 6.99 6.93 6.91 Mean 7.27 7.19 7.19 7.19 7.19 7.19 7.16 7.15 7.15 SEN 0.03 0.00 0.00 0.05 0.05 0.08 0.05 0.06 0.08 CEREBRAL BLOOD FLOW 1 mg/kg endotoxin Time (min) 15 30 60 90 120 150 180 210 ZAO 10 0 Dog.# ¢0N\OOO MKOO PFC-F Nd’Ntn .0... #0300- N P—P— PQONO ...-o...- v-Nl‘nd'ln 132 CEREBRAL PERFUSION PRESSURE 1 mg/kg endotoxin Time (min) 15 30 60 90 120 150 180 210 240 10 0 Dog # 64 134 105 112 120 U\I- PF '11-..- 72 135 85 81 127 81 100 103 78 120 77 113 97 80 11h 6h 90 90 76 9O #9 8O 71 107 11.8 108 11.2 98 8.5 97 8.8 97 11.8 70 15.1 69 12.0 76 21.3 139 5.1 Mean SEM 133 F-KO MN 0‘4" 9 e “.“1 \OInln |n0 .5 .3 Q (N .0 [\c— C O O mono OLn M501- O—N 0‘4’ 0 mmtmmn mo oo—mxo moo \OO‘O‘F 0pm 0:.0 $4.0 w¢.¢ —N.m wo.n mm.m : .m om— 0500 4’15- . C O O In InLNOOM CO"- (I) d‘ e e ‘3 lnko 3&0“ mo [\ONnJ’kO InL\O\'-O\ O O O O O .8 e e o N \mem CON omoo OM- mm.o .Nm.: wm.n oo.m mm”. i ON— m.m w.m m.o_ n.o_ o.m_ o.“— m.m 0.: m.u o... o.m o.o 06 on Aqfiav .aaa qfiuouovno mx\wa _ mmpmmmmm anqh Adszmommmmuo m:.o #0.: .u.m N¢.¢ m~.n m¢.m :n.: 0a —:.o mn.o um.m mo.¢ mm.m ——.n :m.¢ mg.m um.m :N.¢ Fm.m m¢.¢ mm.¢ mm.m ow on Aafllv 0369 nfiuovocno mx\wa _ muz Admmmme um.o :n.m mm.~ mN.m .m.m om.n mo.N m— wm.o w:.m m .N m .m mm.m 9.: N.N 0— 134 5 mm m.— m p mm 94 OJN m 3 ON m _ NN ON m. 05 0 mm om w P om +3 m ow— +N —N :m m. o— :m. o— of w mm on ON ON +3 — — ON— w mm :m ON ON 3 Z om 5338 min — m. :m on ON w p :N N — om A53 2:9 54m Egdemmmm +2 mm mm ON ON mm m— m mm mm mm mm 0— am 582 1:0 v-Nl‘nd'ln Dog # 0 1 89 2 81 3 93 A 93 5 75 Mean 86 SEM 3.5 Dog # o 1 35 2 3A 3 .37 h 26 5 37 Mean 36 SEM 2.0 30 1A6 90 90 100 11.6 1.3 60 26 23 28 27 25 1.4 135 ARTERIAL P02 1 mg/kg endotoxin Time (min) 90 120 150 180 101 96 93 91 97 97 92 87 117 117 114 117 89 89 94 96 73 75 81 83 95 95 95 95 7.] 6.8 504 509 VENOUS P02 1 mg/kg endotoxin 90 “126%. 180 26 28 26 28 27 21 27 26 29 26 29 32 25 24 26 27 1.3 1.2 1.0 1.9 210 91 77 117 93 95 6.4 210 23 25 23 26 2.6 240 15 26 20 24 6.0 136 ARTERIAL P 1 mg/kg endotoxin 002 Time (min) Dog # 0 30 60 90 120 150 180 210 240 1 33.5 28.2 22.0 21.0 22.0 27.0 32.0 30.5 27.5 2 33.2 12.0 19.0 18.5 17.5 20.0 22.5 19.5 16.0 3 31.0 26.5 13.0 13.5 16.5 19.5 21.5 19.5 20.0 4 27.5 27.0 23.0 17.0 20.0 18.5 17.5 20.0 19.5 5 44.5 39.5 31.0 34.0 33.0 32.5 32.0 43.5 44.5 Mean 33.9 26.6 21.6 20.8 21.8 23.4 25.1 26.6 25.5 SEM 2.8 4.4 2.9 3.5 3.0 2.7 2.9 4.7 5.1 VENOUS Pcoa 1 mg/kg endotoxin Time (min) Dog # 0 3O 60 90 120 150 180 210 240 1 47.9 46.1 45.4 45.4 46.4 48.5 51.0 54.5 47.5 2 38.8 47.0 44.0 44.0 46.4 53.0 60.0 57.0 56.5 3 40.0 39.0 36.7 42.0 37.0 41.0 41.5 40.5 41.0 4 41.5 40.5 40.5 38.0 36.0 38.0 36.5 38.5 37.5 5 52.0 56. 54.5 57.0 56.5 58.0 60.0 63. 69.0 Mean 44.0 45.7 44.2 45.3 44-5 47.7 49.8 50.g 50.3 SEM 2.5 3.0 3.0 3.2 3.7 3.7 4.8 4. 5.7 137 ARTERIAL pH 1 mg/kg endotoxin Time (min) Dog # 0 30 60 90 120 150 180 210 240 1 7.37 7.36 7.36 7.37 7.37 7.36 7.36 7.35 7. 38 2 7.38 7.35 7.35 7.35 7.34 7.30 7.27 7.30 7. 27 3 7.29 7.32 7.35 7.33 7.36 7.38 7.37 7.34 7. 31 4 7.41 7.43 7.43 7.47 7.43 7.47 7 49 7.46 7. 48 5 7.39 7.34 7.44 7.38 7.35 7.33 7 34 7.38 7 37 Mean 7037 7038 7039 7038 7037 7037 7037 7037 7036 SEN 0.02 0.02 0.02 0.02 0.02 0.03 0.04 0.03 0.04 VENOUS pH 1 mg/kg endotoxin Time (min) Dog # 0 30 60 90 120 150 180 210 240 1 7.30 7.28 7.28 7.25 7.26 7.24 7.23 7.25 7.26 2 7.31 7.22 7.31 7.31 7.25 7.21 7.18 7.18 7.20 3 7.27 7.27 7.26 7.26 7.28 7.26 7.28 7.26 7.21 4 7.35 7.32 7.36 7.34 7.34 7.38 7.35 7.34 7 36 5 7.31 7.29 7.31 7.33 7.26 7.24 7.25 7.26 7.27 Mean 7.31 7.28 7.30 7.30 7.28 7.27 .26 7.26 7.26 SEM 0.01 0.02 0.02 0.02 0.02 0.03 0.03 0.03 0.03 138 Q. Q. '— Lxgxonn Ooo ooh \O v-v-v— MNmR¢m n oomomooo 01>- ozm 00000000 CD- 8 . S7 O 0 mun“).- 4'0 meimonmw oomooooo 00‘ e C-v-v-v—v-v-v-v- 0 N .- mmbmmmmm ZOHmemHm Admmmmmo N.m —.m o.wm~ o.nn— o.—m— o.Mm_ o.mmp o.mm— o.mN— o.mm— o.mo— o.wo— o.mmp m.MN— m.o¢— m.wN_ m.¢:_ o.N¢— o.mm_ o.m:_ om om Andav mafia mpaoaflaonua Houpcoo O O C I O "In anln Vn I-F-v-F—v—v-e—I- '- O I 0000‘ CNN “MNNg 0N000000 041" O— :.m o.—n_ o.wm_ o.—Mp O.MN— o.wop o.mmp o.mm— m.n:_ o.mn— m §5 :53 1!) v-an-J'anD-(X) z1:1: CEREBRAL BLOOD FLOW control experiments Time (min) 15 30 60 90 120 150 180 210 240 23.6 10 0 Dog # 139 meno nomm rmmntx In N figN MN Nan '- J'J’Bl-n-S'NOO :1" 8888888 [\— MO 30 "\N L\.U\O\v-.l\I-I\N\ NNMNNNM 0 L\ N 23.0 25.5 30.4 23.0 27.6 24.1 33.2 41' e U\ N 26.8 34-3 27.6 26.0 32.6 22.9 #‘N “\d'Owd'In "\KOChw NNNNNNNM 4'0 {M0 N 00 88Rn8888 \Olnd'N‘N\00\o e e 00 e co. \omoxxo \OLNN NNNMNNNM '- N Ind-mm 1500 28. 1. SEM Mean 140 Rd 3.: 8: $2 om mucoaduomuo Hohuaoo mm.o :N.o Nm.: mu.: n¢.m mo.m No.m mm.¢ :1...“ Wm N O O mm.m om.n mo.¢ mm.¢ m:.m :m.m pm.m m¢.m om on adfiav mafia mN.o _w.: mm.¢ #5.: MW”... —¢.n m—.m wm.m mm.m m— moz Admmmmmo m ‘th 5?. 2 ho wmnd-mmtxoo CEREBROSPINAL FLUID PRESSURE control experiments Time (min) 15 30 60 90 120 150 180 210 240 10 0 Dog # 141 InInOv-InOOI-n n-éd'éNv-tm FF 000In00In0 [\MNOO M41113“) 00000000 moxrnmmo O\O\ '— '- OInOInOOInI-n (XDKOMCD 4'0 511‘- F 00InIn0001n C C C C C C C C n-O d"- Inme P F'- 0In0000In0 [\BMCO-d’wno '- 0InLnIn0000 0:00 oxtmn—o oLmnooooo e e e e e e e O (\Noormnoomcx- '- v- '- omoomooo \ONmnmwzm 00In0 0000 éicnmnoxon 00In0000In 0N000000 \ouwwxaro—rn pop v-N In4’In\D [\CO RESPIRATORY RATE control eXperiments Time (min) 15 30 60 90 120 150 180 210 240 10 Dos# 142 v-Koww -¢O\N\C\ ...-I'- N (\Nw nouns 10 \OO\0U\.-N\(\ v-N P v-\O\OU\ 27.0 6. -1601 10.8 -114'06 10.9 - 3, 10.3 Mean SEM PERCENT CHANGE CEREBRAL VASCULAR RESISTANCE control eXperiments Time (min) 15 30 6O 90 120 150 180 210 ZUO 10 Dog # 152 (\Nd'Odufl'M‘n O O O O O O O O \n—uNNMOn-h-a .0) TH”! 00-03013an 0 O O O O O O O W‘OBW‘GONM H'N IHN O\O\®O\(\OO\(D 0 O O O O O O (1):) Omit“: (\c-c H I NH 3' "\N (\m MW“ VDCD WQOfiOM O‘ONBNBM u 7'” ‘” H4? o~L\Ho~.-«~ “(£50.40an . 1' ". MN H OHIWVWOQ -=)'O\Q' 0:01am NOM—IMHOd'O O O mmqu—I??o «manna: (\3 N MMMTOHO H CNN“: “\ONU‘ \OL\ \03 0G) run O’\("\ d'm NN m: (fic-a 0!“ 011-0 \00\ HO Mean SEM 153 can N.d omH 00.0 00.0 00.0 00.0 00. 00.0 00.0 00.0 00.0 00. 00.0 00.0 00.0 00.0 00. n- 00.0 00.0 00.0 00. .. 00.0 00.0 00.0 0.. u- n: 00.0 00.0 0.. 00.0 00.0 00.0 00.0 00.0 00.0 00.0 00.0 00.0 00.0 002000.000 000000<> 00000000 00 00 00 00 00 00 00 00 00 00 00 00 00 00. 00. u- 00 .0 00 00 nu 00 00 .0 00 u- n- 00 00 00 00 00 00 .0 00 00 00 00 00 00 00000000 00.000000 00000000 0. 0 0.0 0.0 0.0 0.0 0. 0 0.0. 0.0. 0.0. 0.0 0. 0. 0.0. 0... 0.0. 0.0. .. 0.0. 0.0. 0.0. 0.0. u- 0.0 0. 0 0.0 0.0 .. --- 0. 0 0.0. 0.0. 0.0 0.0 0. 0. 0... 0.0. 0.0. 0.0. 0. 0. ..0. 0.0. 00. 00 00 00 0. AGHEV 0809 Azaxov00C0 ux\ws 0 av 050v 59000 :. 02.00:.2000 0920-090000 309m GOOAm A