Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

The effects of systematic hypercapnia on the hindlimb perfusion pressures of acute spinal cats Accili, Eric Anthony 1987

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

Item Metadata

Download

Media
831-UBC_1987_A6_7 A22.pdf [ 4.33MB ]
Metadata
JSON: 831-1.0096857.json
JSON-LD: 831-1.0096857-ld.json
RDF/XML (Pretty): 831-1.0096857-rdf.xml
RDF/JSON: 831-1.0096857-rdf.json
Turtle: 831-1.0096857-turtle.txt
N-Triples: 831-1.0096857-rdf-ntriples.txt
Original Record: 831-1.0096857-source.json
Full Text
831-1.0096857-fulltext.txt
Citation
831-1.0096857.ris

Full Text

T H E E F F E C T S O F SYSTEMIC H Y P E R C A P N I A ON T H E H1NDLIMB PERFUSION PRESSURES O F A C U T E SPINAL C A T S by ERIC A N T H O N Y ACCILI B.Sc, University of British Columbia, 1982 A THESIS SUBMITTED IN P A R T I A L F U L F I L L M E N T O F T H E REQUIREMENT S FOR T H E D E G R E E O F M A S T E R OF SCIENCE in T H E F A C U L T Y O F G R A D U A T E STUDIES (Department of Physiology) We accept this thesis as conforming to the required standard T H E UNIVERSITY OF BRITISH C OL U MB IA July, 1987 ©Eric Anthony Accili, 1987 In presenting this thesis in partial fulfilment of the requirements for an advanced degree at the University of British Columbia, I agree that the Library shall make it freely available for reference and study. I further agree that permission for extensive copying of this thesis for scholarly purposes may be granted by the head of my department or by his or her representatives. It is understood that copying or publication of this thesis for financial gain shall not be allowed without my written permission. Department of /^^'^/Ojy The University of British Columbia 1956 Main Mall Vancouver, Canada V6T 1Y3 Date DE-6(3/81) ii A B S T R A C T Normal levels of CO 2 are responsible for the maintenance of approximately 30% of sympathetic neurogenic vascular tone in intact cats. The central medullary chemoreceptors have been implicated as the major source of this CO2 dependent neurogenic vascular tone. However, it is possible that spinal cord CO2 sensitivity could also have mediated a portion of neurogenic vascular tone. Cats with acute and chronic spinal transections can maintain near normal levels of systemic arterial blood pressure, and show cardiovascular and sympathetic reflex changes in response to a variety of stimuli. Thus, it seemed likely that the acute spinal cat could exhibit the spinal component of CO2 mediated sympathetic neurogenic vascular tone. Therefore the effects of systemic CO2 increases on the perfusion pressures of vascularly isolated hindlimbs autoperfused at constant flow (as an indication of vascular resistance and sympathetic vascular tone) were studied in the acute cervical spinal cat. The contributions of the lumbar sympathetic system and the adrenal glands to perfusion pressure responses to CO2 were evaluated. Experiments were carried out in mongrel cats with acute cervical (C2) transections. Each cat had one leg denervated by cutting and stripping the lumbar sympathetic chain from L-pLy. In all cats each hindlimb was vascularly isolated and perfused with blood taken from the abdominal aorta. Bilateral adrenalectomy was performed on 8 animals. CO 2 administration for 5 minutes resulted in biphasic increases in the perfusion pressures of both legs which were designated peakl (Pi) and peak2 (P2). Increasing PCO2 from 16 to 38mm Hg, and from 16 to 62mm Hg resulted in significant PI and P2 responses of the innervated leg. This also resulted in a significant P2 response, and an observable but insignificant PI response, of the denervated leg. Adrenalectomy reduced P i and P2 responses of the innervated leg, and abolished the PI response and reduced the P2 response of the denervated leg. iii In non-adrenalectomized cats increasing PCO2 also resulted in a significant increase in systemic arterial pressure (SAP) with no changes in heart rate (HR). In adrenalectomized cats increasing PCO2 resulted in an observable but non-significant increase in SAP and a significant decrease in HR. These results suggested that: 1) The PI response was primarily a sympathetic neurogenic response to increased CO^. 2) The P2 response was primarily a hormonal response to CO2 in the denervated leg, and a combination of hormonal and sympathetic neurogenic responses to CO 2 in the innervated leg. 3) The adrenal glands were mostly involved in the P2 response to CO2, but possibly had a small role in the PI response. 4) Other non-adrenal vasoconstrictor hormones may have played a role in the P2 response to C0 2.. 5) Likely, CO2 initially activated the sympathetic system to directly increase neurogenic tone, perhaps by stimulating sympathetic afferent or efferent neurons, or hypothetical spinal chemosensitive regions. Progressively the adrenal and possibly other unidentified vasoconstrictor hormone systems became activated, either directly by CO2 or indirectly by CO2 mediated sympathetic activation. These hormone systems may have also played a role in CO2 mediated maintenance of vascular tone. IV T A B L E OF C O N T E N T S P A G E A B S T R A C T ii T A B L E OF C O N T E N T S iv LIST OF FIGURES vii LIST OF T A B L E S ix A C K N O W L E D G E M E N T S x I. INTRODUCTION 1 II. REVIEW O F T H E L I T E R A T U R E A. Resting SND and CV Parameters in Spinal Animals 3 B. C V Reflexes in Spinal Animals 10 C. Sympatho-adrenal Effects of C 0 2 12 D. Effects of CO2 on Vascular Resistance and SND 15 E . Hypercapnia and the A V P and Renin-Angiotensin Systems 21 III. PURPOSE OF T H E EXPERIMENTS 27 IV. M E T H O D S A. Rationale 28 B. Animals and Induction of Anaesthesia 30 C. Surgical Procedures 30 D. Hindlimb Perfusion Preparation 32 E . P C 0 2 and pH Maintenance 35 F. Blood Pressure Measurements 35 G. Determination of Spinal Reactivity 36 H . Experimental Protocol 36 I. Additional Procedures 37 V. D A T A A N A L Y S I S AND INTERPRETATION A. Perfusion Pressure and Vascular Resistance 38 B. Systemic Variables 38 C. Experimental Groups and Statistical Comparisons 38 VI. R E S U L T S A. Induction of Hypercapnic Acidosis 41 B. Perfusion Pressures 1. General 41 2. Non-adrenalectomized Spinal Cats 46 3. Adrenalectomized Spinal Cats 46 C. Vascular Resistance and Pump Flow Rates 46 D. Systemic Hemodynamics 1. General 59 2. Non-adrenalectomized Spinal Cats 59 3. Adrenalectomized Spinal Cats 60 E . Additional Procedures 60 VII. DISCUSSION A. Discussion of the Methods 1. The Experimental Preparation a. Effects of Vascular Isolation 69 b. Effects of Spinal Transection-Spinal Shock 69 2. The Experimental Procedure a. Hyperventilation 71 b. Constant Flow Method 73 c. Hyperoxia 75 B. Discussion of the Results 75 C. Concluding Comments 86 VIII. R E F E R E N C E LIST vii LIST O F F I G U R E S Figure Page 1 Schematic representation of the preparation used in these experiment 33 2 Char t recording of the responses of systemic arterial pressure and perfusion pressures of both legs to 7 minutes of hyperventi lat ion wi th 100% CO2, in a non-adrenalectomized animal 44 3 Char t recording of the responses of systemic arterial pressure, and perfusion pressures of both legs to hyperventilation wi th 10% inspired CO2 in a non-adrenalectomized animal 47 4 Perfusion pressure values for F TC02= 0% (B) and 5% ( P I , N and P2) in the innervated and denervated legs of non-adrenalectomized cats 49 5 Perfusion pressure values for F T C 0 2 = 0% (B) and 10% ( P i , N and P2) in the innervated and denervated legs of non-adrenalectomized cats 49 6 Perfusion pressure values for F T C 0 2 = 0% (B) and 5% ( P i , N and P2) in the innervated and denervated legs of adrenalectomized cats 51 7 Perfusion pressure values for F j C 0 2 = 0% (B) and 10% ( P I , N and P2) in the innervated and denervated legs of adrenalectomized cats 51 8 Vascular resistance values for F j C 0 2 = 0% (B) and 5% ( P I , N and P2) in the innervated and denervated legs of non-adrenalectomized cats 54 9 Vascular resistance values for F T C 0 2 = 0% (B) and 10% ( P i , N and P2) in the innervated and denervated legs of non-adrenalectomized cats 54 10 Vascular resistance values for F T C 0 2 = 0% (B) and 5% ( P I , N and P2) in the innervated and denervated legs of adrenalectomized cats 56 11 Vascular resistance values for F j C 0 2 = 0% (B) and 10% (P I , N and P2) in the innervated and denervated legs of adrenalectomized cats 56 Chart recording the perfusion pressures of both the innervated and denervated legs of one adrenalectomized cat in response to afferent femoral nerve stimulation. Two sets of parameters were used, one which produces a pressor effect, the other which produces a depressor effect in the intact cat ix LIST OF TABLES Table Page I Values of HR, SP, DP and M A P of non-adrenalectomized cats during normoventilation with 100% O2, and hyperventilation with 5 and 10% CO2 (minimum and end values) 42 II Values of HR, SP, DP and M A P of adrenalectomized cats during normoventilation with 100% 0 2, and hyperventilation with 5 and 10% CO2 (minimum and end values) 43 III Values of P gC02 and p H a in non-adrenalectomized cats during hyperventilation in 100% 0 2, 5 and 10% C 0 2 58 IV Values of P aC02 and p H a in adrenalectomized cats during hyper ventilation in 100% 0 2, 5 and 10% C 0 2 61 V Values of pump flow rates for the innervated and denervated legs of non-adrenalectomized and adrenalectomized cats 62 VI Values for flow rates, perfusion pressures of each leg before denervation, and vascular resistances of each leg before denervation, and 0 and 3 hours after denervation, for two non-adrenalectomized cats 65 VII Values for changes in perfusion pressure in response to carotid clamping, and inspiring 5% CO2 (a peak 1 change), in one adrenalectomized cat 66 X A C K N O W L E D G E M E N T S I take this opportunity to extend my thanks and appreciation to: Dr. Franco Lioy for allowing me to work in his laboratory, and for his assistance with, and supervision of, this project and thesis. Drs. J.R. Ledsome, N. Wilson, and P. Vaughn for their pertinent criticism and advice concerning this project and thesis. The B.C. Heart Foundation, the Faculty of Medicine and the Department of Physiology for their financial support. Pat Leung for preparation of this thesis, and the use of the computer and his brain. Marie Sweeney Greene for the use of the computer and printer, and for her effective methods of help and encouragement. Mary Forsythe and the other administrative staff for their help. The boys in the workshop, Joe Tay and especially John Sanker for their technical assistance with this project. Drs. Carol-Ann Courneya and Andrew Rankin for their advice and constant encouragement during the course of my work in this department. The graduate students and post-docs of physiology, for their friendship and advice. I hope our association, both professionally and personally, will continue in the future. The graduate student supervisor Dr. Ray Pedersen, for his constant support of myself and all graduate students, most of which probably was not in his job description. Sheela Puttaswamaiah for her technical assistance and for listening to my jokes. I hope we remain friends and colleagues, and I wish her the best of luck in her future. To all members of this department for helping make my stay in physiology fun and worthwhile. xi This thesis is dedicated to my family and Tina. 1 I. I N T R O D U C T I O N The effects of carbon dioxide on the cardiovascular system are complex because of its direct effects on the heart and blood vessels, and its various effects on the autonomic nervous sjrstem and endocrine systems. Carbon dioxide has a direct vasodilatory effect on most vascular beds, and has direct negative inotropic and chronotropic effects on the heart. Carbon dioxide stimulates both central medullary chemoreceptors and peripheral chemoreceptors, leading to peripheral vasoconstriction and conflicting cardiac effects. In addition, carbon dioxide stimulates the adrenal medulla, the renin angiotensin system and the A V P system. Finally, other neural, endocrine and neurohormonal systems, which can affect the cardiovascular system, may be sensitive to carbon dioxide. Thus, the hemodynamic consequences of changing blood carbon dioxide tensions is a balance of these various systems whose actions are central and peripheral, and may affect each vascular region differently. Therefore it is difficult to attribute the actions of carbon dioxide to specific structures and systems. Changes in carbon dioxide tensions also cause concomitant changes in the pH of the blood, and in most cases these two factors are not separated. Previous studies have demonstrated C 0 2 can generate a portion of sympathetic neurogenic vascular tone. Spinal animals have been able to maintain systemic arterial pressure (SAP), and respond to a variety of stimuli. Therefore, it was decided to determine whether the acute spinal cat could demonstrate a spinal portion of C 0 2 mediated vascular tone.' This thesis has attempted to: a) Comprehensively review the literature regarding the state of blood pressure and sympathetic nerve discharge (SND), and the reactivity of the cardiovascular and sympathetic systems and SND after spinal transection. 2 b) Review the effects of CO2 on the sympathetic nervous system, and on various hormone systems with respect to their possible contributions to vascular tone. c) Provide and discuss evidence that supports the notion that the spinal cord can provide CC"2 mediated vascular tone in the spinal animal and perhaps in the intact animal. Also, to discuss the possibility that hormone systems may be involved with CO2 mediated vascular tone. 3 IL REVIEW OF THE LITERATURE A. Resting SND and CV Parameters in Spinal Animals Acute spinal transection of the cervical spinal cord, has been shown to dramatically reduce systemic arterial pressure (SAP) and sympathetic nerve discharge (SND) (Bernard, 1863; Alexander, 1946). However, near normal levels of resting SAP and vasomotor reflexes have been obtained in the chronic and acute spinal situation (Brooks 1933). Whether the recovery of SAP could have been due to a possible concomitant recovery of SND, was not known. Alexander (1945) found reduced and less synchronous firing in the inferior cardiac nerve of dogs subjected to thoracic cord section (T5) hours earlier. In acute spinal cats (Cl , C4), pre-ganglionic neurons recorded from the cervical sympathetic trunk exhibited low firing rates. (Mannard and Polosa, 1974; Zhang et al., 1982). Also in acute cervical spinal cats some pre-ganglionic (at T1-T4 and L1-L4) and post-ganglionic nerve fibres exhibited some background firing (Beacham and Perl, 1964a,b; Fernandez de Molina and Perl, 1965). Post-ganglionic vasoconstrictor neurons of the muscle and skin of the hindlimbs of chronic spinal cats (T l , T7-L1) also showed reduced activity (Horeyseck and Janig, 1974; Kummel, 1983). This rate of firing was lower than in intact animals, and the recovery of resting activity was suggested to take from seven days to four weeks. In seven patients with chronic spinal lesions ranging from C8-T8, spontaneous sympathetic activity recorded from the muscle fascicles of the leg was reduced, and pulse synchronous activity was not seen (Stjernberg and Wallin, 1983). Therefore after spinal transection it was clear that sympathetic activity was reduced, not usually synchronized to respiratory or cardiac cycles and although it has been shown to recover, basal SND was lower than in the intact animal. Whether or not the sympathetic nervous system could have played a role in the maintenance of SAP or regional vascular resistance after spinal transection has been 4 investigated using sympathetic blockade, with equivocal results. In unanaesthetized, atropinized dogs subjected to cervical transection, 8-12 days earlier, ganglionic blockade failed to reduce SAP (Kaneko et al., 1964). These authors suggested that in chronic spinal dogs, resting sympathetic activity was insufficient to support resting SAP. Interestingly, it seemed that the drop in SAP after spinal transection was primarily due to a decrease in stroke volume (SV) and (CO) while total peripheral resistance (TPR) did not seem to be affected only 2 hours after transection. This may have indicated the development of non-neural (i.e. humoral) vascular support. However, in acute cervical spinal cats, administration of hexamethonium did cause a fall in SAP, suggesting that sympathetic activity is responsible for maintenance of some resting SAP (Johnson et al., 1967). Species differences could have accounted for the different results obtained. Also, vagotomized animals may not have had any sympathetically mediated cardiac support that could have developed after spinal transection because the HR may have been already very high. Conversely, non-vagotomized preparations could compensate for a reduction of SAP produced by ganglionic blockade, by decreasing vagal activity and therefore overestimating the contribution of the symapthetic nervous sytem. However, in vagotomized cats subjected to acute thoracic (T5) transection hindlimbs perfused at constant flow exhibited higher perfusion pressures (PP) than those seen with an intact spinal cord, after alpha-blockade (with phentolamine) (Lioy et al., 1978). It was suggested that some neurogenic tone was therefore present in the hindlimb vessels. However, because of the direct vasodilatory properties of phentolamine the hindlimb vascular tone present after alpha-blockade may have been underestimated. Therefore the PP difference between the intact alpha-adrenergically blocked animals and the spinal animals may have been overestimated. Further observations of studies in unanaesthetized decerebrate cats subjected to cervical (Cl) transection and whose hindlimbs were perfused at constant flow (Rohlicek and Polosa, 1986), suggested that hexamethonium caused reductions in hindlimb PP. Re-analyses of studies in vagotomized rats (Lioy and Trzebski, 1983), suggested that 5 the effect of cervical section (C6) 30 mins earlier, caused a lesser reduction in SAP, than ganglionic blockade in intact rats. Again these results could have indicated the presence of sympathetic and/or increased humoral support for SAP and vascular resistance which may develop during the period after spinal transection. Therefore whether or not any sympathetic support for SAP or vascular resistance (hindlimbs) was present after spinal transection, based on studies with ganglionic blockade, has remained unclear due to species differences, different experimental methods (i.e. vagotomized versus non-vagotomized preparations), and the possible development of humoral support for resting SAP. Attempts to study basal SND from several vascular regions, in conjunction with measurements of various hemodynamic parameters have proven to be useful in elucidating the contributions of basal SND to cardiovascular support. Background firing of pre-ganglionic neurons of acute spinal cats and post ganglionic vasconstrictor neurons of chronic spinal cats was not obviously correlated to cardiac or respiratory cycles (Beacham and Perl, 1964a; Horeyseck and Janig, 1974). However, studies in acute spinal cats showed that some tonic activity of pre-ganglionic and post-ganglionic neurons may reflect a vasomotor component (Fernandez de Molina and Perl, 1965). SAP and blood flow in some arteries exhibited spontaneous fluctuations, which could be blocked by a short-acting ganglionic blocker (tetraethylammonium, or TEA) . Although not mentioned by the authors, T E A did seem to reduce SAP and blood flows below the minimum levels prior to blockade, further suggesting a vasomotor component. The heart rate (HR) was stable during these fluctuations, implying that sympathetic outflow to the heart and adrenal medulla was not involved and therefore the resting activity was specifically distributed. Finally, there was sometimes a temporal relation between sympathetic post-ganglionic discharge and SAP fluctuations. Because the drop in SAP with T E A was unclear and also because uncontrolled afferent inputs such as anoxia may have contributed to the fluctuations in SAP the contributions of basal SND to the maintenance of SAP in these experiments may have been very small. On the other hand, these results support the 6-notion that specific or discrete basal SND may have had a vasomotor component that is not always reflected in the resting SAP. Meckler and Weaver (1985) provided further evidence that spinal transection variably reduced basal SND in different nerves i.e. basal SND was discretely organized. In cats, with cranial nerves 9 and 10 sectioned, and subjected to cervical transection (Cl) 30min. to 2hr. earlier, splenic nerve discharge was unchanged while renal and cardiac nerve discharge was reduced by 50%. Intravenous infusions of phenylephrine were required to maintain SAP and the authors therefore concluded that the basal SND present was not adequate to support SAP completely. Interestingly, they did suggest that varying levels of SND may have contributed to cardiovascular support, presumably in a variety of ways. For example, splenic nerve discharge may have mobilized blood cell stores and renal nerve activity may have helped to control sodium and water balance. These studies by Meckler and Weaver, and Fernandez de Molina and Perl, have suggested that although reduced, basal SND may have been only regionally reduced and that the remaining basal SND did provide cardiovascular support in the form of specific contributions to various vascular beds . These contributions may have been masked by the drop in resting SAP (e.g. regional vascular resistance changes) or may have affected hemodynamic parameters indirectly (e.g. through renal sodium and water balance). Although in the acute situation basal SND did not visibly support restng SAP there was evidence that this support may have developed over time. Ardell and Barman (1982) showed that in atropinized cats subjected to cervical transection (C6) two days earlier, ganglionic blockade with hexamethonium initially failed to alter SAP, and post-ganglionic renal or cardiac SND. However, ganglionic blockade caused a statistically significant fall in SAP 9-37 days after transection. They concluded from this study that the sympathetic component for SAP support in the cervical spinal cat develops sometime between 2 and 9 days after transection . On further inspection, SND and SAP both seemed to be greater in the spinal animal after 2 days than in the intact animal with ganglionic blockade. This 7 seemingly contradictory and unexplained result may not have been significant, but may have indicated a change in the ability of the ganglionic blocker to effectively have caused sympathetic blockade in the spinal animal. Humoral support could have maintained an elevated SAP and perhaps also have affected basal SND. In summary, basal SND is reduced in acute spinal animals, increases over time and is differentially distributed. Studies of basal SND, or SAP with sympathetic blockade were not sufficient to determine the contributions of basal SND to resting SAP. Studies of basal SND and specific hemodynamic variables have suggested that specific basal SND may have contributed to cardiovascular support in the acute spinal animal but this support may not have been reflected in resting SAP. Similarly, basal SND may have been destined for a function not necessarily cardiovascular in nature. The sympathetic contribution to resting SAP and other hemodynamic variables has also been suggested to be time dependent. Maintenance of vascular tone could have also been a reflection of increased humoral sensitivity of vascular territories as suggested by comparison of spinal and intact animals with ganglionic blockade. Alternatively this comparison could have reflected a difference in the ability of a ganglionic blocker to effectively have caused sympathetic blockade. However, it is likely that each vascular region would have received unique neurogenic and humoral support which was altered by spinal transection and which varied over time depending on the recovery of the basal SND output to that vascular region. 8 B. Cardiovascular Reflexes in Spinal Animals After spinal transection, animals show changes in SAP in response to a variety of stimuli. In dogs subjected to low cervical transection 300 days earlier, Sherrington (1906) found that stimulation of sciatic afferents could cause large increases in SAP, and possibly release of adrenalin. Stimulation of the sciatic nerve also caused increases in HR and SAP in chronic spinal cats which were attenuated but not eliminated by adrenalectomy (Brooks, 1933). Visceral stimulation caused increases in SAP and vasoconstriction in acute spinal cats (Downman and McSwiney, 1946). More "natural" stimuli are also effective in eliciting changes in SAP after spinal transection. Hemorrhage of 20-25% of blood volume caused vasoconstriction, increases in HR and increased adrenalin release (Brooks, 1935). Bladder distension has resulted in vasoconstriction in acute spinal cats (Mukherjee, 1957) and vasoconstriction and HR changes in humans depending on the location of the transection (Wurster, 1979). Stretching the thoracic aorta resulted in increases in HR, vasoconstriction and catecholamine release in acute cervical (Cl) cats (Lioy et al., 1973). These results have indicated that the spinal animal can make cardiovascular adjustments to stimuli and that the adrenal glands may be of particular importance for these responses in spinal animals as compared to intact animals. After spinal transection, many stimuli have resulted in more specific changes in sympathetic nerve activity. Sciatic nerve stimulation has altered efferent sympathetic nerve activity as demonstrated by increased external carotid activity in chronic spinal cats (Ardell, 1982), but caused no changes in inferior cardiac nerve activity in acute spinal dogs (Alexander, 1945). This difference could be attributed to species differences, or acute versus chronic spinal transections but could also have indicated a specific versus mass activation sympathetic response. Stimulation of the sciatic nerve in the acute spinal cat evoked discharges primarily in the sympathetic pre-ganglionic neurons (SPNs) of the lumbar region, but rarely in the thoracic region indicating some specificity in the response 9 (Beacham and Perl, 1964a,b). In this study stimulation of dorsal roots or spinal afferent nerves evoked reflex discharges in the preganglionic rami of the same and adjacent spinal segments. Afferent nerve stimulation also resulted in extensive but non-uniform activity in the post-ganglonic nerves of stellate and upper lumbar ganglia in a similar preparation (Fernandez de Molina and Perl, 1965). These authors noted transient increases in SAP, and decreases in femoral, inferior mesenteric, and brachial artery blood flow. They did not see changes in HR and therefore suggested that efferent cardiac and adrenal medullary sympathetic nerves were not activated. In these experiments, other stimuli such as noxious and mechanical stimuli, resulted in responses which were greater than those elicited with electrical stimulation. Electrical stimulation may have indiscriminately activated all afferent fibres, including groups of afferent inhibitory fibres whereas "natural" stimuli could have activated more specific structures (Beacham and Perl, 1964b). Specific cardio-cardiac reflexes have been elicited in acute spinal cats. Electrical, mechanical and chemical activation of cardiac afferent nerve fibres resulted in increases in thoracic SPN activity and tachycardia (Malliani et al., 1975). Also, infusions of saline produced tachycardia, which was decreased by section of T1-T4 outputs (Malliani et al., 1973), and decreases in most SPNs activity (Fernandez de Molina and Perl, 1965). Increases in pressure with adrenalin have produced decreases in cardiac nerve activity, no change in renal nerve activity and decreases in the activity of a number of SPNs (Meckler and Weaver, 1985; Ardell et al., 1982; Alexander, 1945; Fernandez de Molina and Perl, 1965). Blood withdrawal resulted in an increase in the activity of splenic nerve, and cardiac nerves (Meckler and Weaver, 1985), and primarily increases in SPNs activity (Fernandez deMolina and Perl, 1965). A small portion of SPNs reacted in the opposite direction in response to increases and decreases in pressure which implied some differentiation in the response of the SPNs (Fernandez de Molina and Perl, 1965) Also, vasodilator fibres have been shown to exist in the cat hindlimb (Janig, 1985; Lioy and White, 1973). Both electrical stimulation of the ventral funiculi of the spinal cord and 10 bladder inflation have resulted in decreases in SAP and increases in the blood flow of the hindlimbs of acute cervical spinal cats (Dorokhova et al., 1974). In chronic spinal cats, vasoconstrictor fibres to the skin and those to the muscle of the hindlimb responded differently to mechanical and noxious stimuli, again implying a differentiated response (Horeyseck and Janig, 1974). These results have implied that the sympathetic vasomotor system is able to respond to various stimuli in a discrete and integrated manner in spinal animals, and that the sympathetic outflow could have a dual function (i.e. both constrictory and dilatory functions). These responses seem to be due to the spinal neuronal circuit portion of the responses seen in the intact animal. There have been differences between the responses of spinal versus intact animals which have indicated that supraspinal input is superimposed on these spinal circuits. Electrical stimulation of spinal sensory nerves evoked discharges in renal and inferior cardiac nerves, which could be separated into a late and sometimes early response (Coote and Sato, 1978). After acute thoracic spinal transection only the early response was obtained, and it was not always present. Pressor and depressor reflexes obtained by contracting skeletal muscle have been demonstrated in the intact cat (Mitchell, 1985). After acute spinal transection, only a reduced pressor reflex could be obtained (Iwawoto et al., 1985). In chronic spinal patients stimulation below the level of the lesion has resulted in cutaneous vasoconstriction which lasts much longer than in intact subjects (Stjernberg and Wallin, 1983). Both skin and muscle fasicles were activated simultaneously in response to these stimuli, unlike in the intact subjects suggesting a less differentiated response in the spinal subject. The long lasting vasoconstriction has also been seen in response to visceral stimulation, in the chronic spinal cat (Kummel, 1983). Again, simultaneous activation of both muscle and skin fasicles of the legs was seen in the spinal animal unlike the intact animal. 11 These results have indicated that responses of spinal animals, although somewhat complex and integrated, were less differentiated than responses in the intact animal. Therefore the spinal cord may act as an integrator of peripheral and supraspinal inputs (Janig, 1985; Lind, 1983). Furthermore, this integration of information by the spinal cord would be a dynamic process, and would not be activated only when perturbations of the system occur (Malliani et al., 1975). 12 C. Sympatho-adrenal Effects of COg Carbon dioxide has been shown to stimulate the sympato-adrenal system. In anaesthetized dogs, 15 minutes of diffusion respiration, which resulted in P a C U 2 = 108 mm Hg and pH= 7.01, decreased diastolic pressure (DP) and increased systolic pressure (SP), noradrenalin (NA), and adrenalin (A) (Miller, 1960). It was suggested that these increases in plasma catecholamines were due to central stimulation, and also peripheral effects of CO2 such as adrenal medullary stimulation and N A liberation from nerve terminals. In acute cervical spinal cats inhaling 12% CO2 for 10 minutes caused release of catecholamines, detected by a nictitating membrane bioassay (Tenney, 1956). Adrenalectomy decreased by 60%, the release of catecholamines induced by inhaling 30% CO2 for 10 minutes. When the cord was destroyed, release of catecholamines was reduced at the lower levels but not at higher levels of inhaled CO2. The pithed and adrenalectomized cats exhibited no release of catecholamines, at any level of inhaled CO^-In acute cervical spinal dogs, a P aCU2= 127 mm Hg (and pH= 6.93) resulted in increased levels of plasma catecholamines, primarily adrenalin (Cantu et al., 1966). Perfusion of the the isolated dog adrenal gland with blood or Locke solution with PC02 = 30 mm Hg and pH= 7.44 resulted in a release of 70 ng catecholamines/gland/min (Nahas et al., 1967). With P C 0 2 = 82 mm Hg and pH= 7.01, catecholamine release doubled. With P C 0 2 = 172 and pH= 6.84, catecholamine release increased by 7 times. CO2 would have stimulated the sympathetic nervous system and therefore at low levels of CO2 most catecholamine release would likely have been N A spillover from nerve terminals. At higher P C 0 2 levels, the adrenal medulla would also have been stimulated releasing both A and NA. Hypercapnia also resulted in catecholamine release in unanaesthetized animals. In humans, breathing 7-14% CO2 for 10-20 minutes caused increases in catecholamines, SP, DP, pulse pressure , HR and ventilation (Sechzer et al., 1960). However, CO2 inspiration also caused sometimes severe agitation which could have enhanced catecholamine release. 13 In unanaesthetized sheep with P aC02= 80 mm Hg and pH= 7.15, there were increases in plasma adrenalin (pA), pNA, cardiac output (CO), SAP, stroke volume (SV), HR, and a decrease in TPR, after 15 mins (O'Brodovich et al., 1982). In this study the agitation of CO2 inspiration was likely augmented by having the heads covered with plastic bags for administration of hypercarbic gases. Another study of unanaesthetized sheep found that breathing 7% CO2 for 15 minutes resulted in no significant increase in catecholamine release (Matalon et al., 1983). Breathing 10% CO2 for 15 minutes did result in a significant increase in catecholamines. However, there was a wide scatter of both baseline and hypercapnic catecholamine levels once again likely due to agitation from both the experimental procedure and the inspiration of C02- In unanaesthetized dogs a P aC02 = 53 mm Hg and pH= 7.19 resulted in significant increases in pNA, HR, and CO, while SAP increased or showed no change, after 20 minutes. pA showed an increasing trend which was not statistically significant (Rose et al., 1983). This may have been due to their attempts to control for stress and agitation, and also because of their comparatively lower P aC02 values which may have not significantly stimulated the adrenals. In all unanaesthetized animal experiments, increased P aC02 also increased ventilation and this could have had cardiovascular consequences particularly in the intact animal. Whether it is the acidosis or the hypercapnia which is responsible for the catecholamine release has not been determined. Both factors were probably involved acting at different locations. In the isolated dog adrenal gland, the increased catecholamine release was similar for both hypercapnic and lactic acid acidosis. More recently in anaesthetized dogs, lactic acid acidosis which resulted in P a C 0 2 = 40 and pH = 6.939 and increased in pNA, pA and TPR as compared to non-acidotic controls (Tajimi et al., 1983). Finally catecholamine release mediated by neural structures would have taken longer under conditions of metabolic acidosis since H + does not cross the blood brain barrier as easily as CO2. 14 Therefore, hypercapnia releases catecholamines by stimulating sympathetic nerve activity and by directly stimulating the adrenal gland. The adrenal release is probably of particular importance in the spinal animal. Possibly, CO2 is more responsible for neurogenic catecholamine release while H~^ ~ and CO2 are both involved in the peripheral release of catecholamines because of their different abilities to cross the blood brain barrier. 15 D. Effects of COg on Vascular Resistance and SND Increases in P &C02 have a dual effect on the vasculature. Vasodilation is the direct effect of CO2 on blood vessels (Somylo and Somylo, 1970), but vasoconstriction can occur as a result of central sympathetic stimulation. Hypercapnia did not induce changes in TPR in concious dogs (Horowitz et al., 1968) and rats (Lagneaux and Remacle, 1981), probably as a result of these opposing actions of CO^- Hypercapnia has also resulted in the increase in SAP and vascular resistance of the skeletal muscle of intact cats but a variety of resistance changes in other organs (Weissman et al., 1976). A variety of responses of organs to increases in P a C 0 2 have been seen in sheep (Matalon et al., 1983) and in intact rats (Lioy, 1986). Sympathetic responses to hypercapnia were regionally differentiated , and local and humoral effects may also be similarly patterned (Weissman et al., 1976). In the dog, small physiological changes in the blood perfusing the cephalic circulation caused increases in myocardial contractility (Hainsworth et al., 1984), abdominal vascular resistance (Ford et al., 1985), and hindlimb vascular resistance (Soladoye et al., 1985). Normal levels of CO2 have been shown to be responsible for the maintenance of a significant portion (about 30%) of neurogenic vascular tone in the hindlimb of the intact cat (Lioy et al., 1978), and changes in P aC02 can alter the blood flow of skeletal muscle in the rat (Lioy, 1986). The background activity of sympathetic preganglionic neurons (SPNs) in the intact cat seemed to be dependent on arterial CO2, and was responsible for the neurogenic sympathetic tone of cardiovascular and other effector organs (Preiss and Polosa, 1977). This was likely the result of the C 0 2 action on the ventral medullary (Lioy et al., 1981) and peripheral chemoreceptors(PCh) (Hanna et al., 1981), and its direct action on SPNs and/or antecedent neurons. In the cat the perfusion pressure of vascularly isolated, innervated hindlimbs, perfused at constant flow and maintained at constant PCO2, PO*2 and pH, was directly related to the PgCC^ present in the rest of the body (Lioy et al., 1978). These experiments, carried out also after PCh denervation, showed that the contribution of peripheral chemoreceptors was not the major 16 source of CO2 sensitivity. Subsequently, it was shown that cold block of the chemosensitive areas of the ventral surface of the medulla of PCh denervated cats attenuated (but did not eliminate) this response (Hanna et al., 1979). In addition, superfusion of the ventral medulla of cats with an acidic and hypercapnic solution caused increases in systemic arterial pressure, phrenic nerve activity, heart rate, inferior cardiac nerve activity, and vertebral nerve activity as well as a decrease in hindlimb flow (Szulzyck and Trzebski, 1976; Lioy et al., 1981). Responses of a group of CO2 sensitive SPNs in the cervical sympathetic trunk of cats demonstrated that approximately one half of these SPNs received input from both central and peripheral chemoreceptors while the remainder received input from only the central chemoreceptors (Hanna et al., 1981). Therefore, central chemosensitive structures are important for maintaining vascular tone while the peripheral chemoreceptors appear to play a lesser role in the intact cat. Hypercapnia resulted in increased splenic nerve activity, but no changes in cardiac and renal nerve activity (Meckler and Weaver, 1985). In the cat hindlimb, 7% inspired CO2 resulted in increases in the activity of muscle vasoconstrictor nerves but decreases in cutaneous vasoconstrictor nerves (Janig, 1985). This could have been because the levels of hypercapnia may have had a depressant effect on some nerves. However, this could once again illustrate a differentiated sympathetic response to hypercapnia. Although central and peripheral chemoreceptors were responsible for much of the C 0 9 mediated neurogenic tone, there may have been other areas that were sensitive to CO2 in cats. In dogs the specific area of stimulation could not be identified and could have included the upper spinal cord as well as the central chemosensitive areas (Rankin, 1983). Cold blockade of the ventral medulla, attenuated but did not eliminate the response of hindlimb resistance to systemic hypercapnia in PCh denervated cats (Hanna et al., 1979). The source of this residual sensitivity was not determined. Results of experiments involving systemic hypercapnia in spinal animals have been conflicting. In spinal cats (T5) there was no change in the perfusion pressure of vascularly isolated 17 hindlimbs associated with changes in inspired CO2 (Lioy et al., 1978). However, the cats were anaesthetized with a barbituate anaesthetic and the experiment was performed only 30 minutes after spinal transection, which may have depressed any reflex responses. Hypoventilation caused increases in the femoral artery resistance in acute cervical cats (Gowdey and Patel, 1964). In another study of spinal (mid-cervical) cats inspiration of 10% CO2 caused increases in blood pressure which were abolished bj' hexamethonium (Johnson et al., 1965). The authors suggested that this hypertension resulted from a sensitivity to C 0 2 of the spinal cord and/or its sympathetic nerve connections. In addition, significant increases in mean arterial pressure by inhalation of 5% CO2 have been observed in spinal rats (C4) (Lioy and Trzebski, 1984). In chronic spinal patients, hypercapnia resulted in increased SAP due to systemic vasoconstriction (Downing et al., 1963). Finally, hypercapnia has resulted in the increase of hindlimb vascular resistance of the acute cervical spinal cat (Rohlicek and Polosa, 1986). 18 The results of SND responses to hypercapnia have also been conflicting. In acute high spinal cats, systemic hypercapnia did not affect discharge of the inferior cardiac nerve (Alexander, 1945). Increases in P aCU2 to approximately 50 mm Hg caused no changes in renal, cardiac or splenic nerves in acute spinal cats (Meckler and Weaver, 1985). This could have, again, been due to the high levels of CO2 which may have depressed the already compromised nerves. However, in chronic spinal cats, hypercapnia produced increases in the activity of cutaneous and muscle vasoconstrictor neurons (Gregor and Janig, 1977). Also in chronic spinal cats, 4-6% inspired CO2 pulse synchronized renal nerve activity (Ardell, 1982). In the acute CI or C4 spinal cat a large fraction of spontaneously active units of the cervical sympathetic trunk were excited by systemic hypercapnia and a significant number of silent units were also recruited into activity (Zhang et al., 1982). These results have suggested that CO2 and/or H + is causing excitation of sympathetic preganglionic neurons, spinal interneurons or hypothetical spinal "chemoreceptors", antecedent and excitatory to the SPN, with consequent increases in vascular resistance and arterial pressure. In spinal cats (Tl), superfusion of the spinal cord at T l with an acidic and hypercapnic solution caused increases in the activity of the inferior cardiac and vertebral nerves, and in systemic arterial blood pressure (Szulzyck and Trzebski, 1976). Therefore, the spinal cord may have been able to generate some vascular tone in response C02- Dorsal rhizotomy has not eliminated basal SND (Meckler and Weaver, 1985; Mannard and Polosa, 1973), indicating afferent influences are not necessary to generate this SND. Therefore some of the basal SND present in the spinal animal, could have been generated by the spinal cord in response to physiological levels of co 2. Depolarizing, hyperpolarizing, depressant and excitatory effects of hypercapnia have been shown in a variety of neuron types. In perfused abdominal ganglia of Aplysia, 50% CO2 caused depolarization of the giant neurons (Chalazonitis, 1963). Longer exposures to CO2 caused narcosis, reducing all phases of spike activity. In neurons that were actively firing, CO2 induced a permanent depolarization which produced narcotic 19 effects. A similar study of perfused abdominal ganglia of Aplysia showed that 5% CO2, which resulted in a pH change from 8.0 to 6.5, produced depolarization and increased discharge of visceromotor neurons (Brown and Berman, 1969). Pacemaker neurons were little affected, while one half of the giant neurons were depolarized and the other half hyperpolarized. Perfusion with HC1 and H2SO4, which resulted in the same pH change as hypercapnic acidosis, also produced similar results as did the local application of these acids. With the pH kept constant, 5-50% CO2 produced no effect and the authors concluded that CO2 induced depolarization was mediated via extracellular H + . These results in invertebrates have demonstrated that hypercapnic acidosis can increase the excitability of some neurons. In mammalian studies the direct effect of CO2 on neurons have been mainly depressant primarily as a result of increased membrane conductance (permanently depolarizing the neurons) and an increase in action potential threshold. Cortical cells of cats and monkeys exposed to 2-20% CO2 mainly exhibited depressed activity with transient excitatory effects at the beginning and end of CO2 inspiration (Krnjevic et al., 1965). The early excitation was not associated with membrane depolarization and was therefore considered to be unlike the excitatory response seen by Chalazonitis in the neurons of Aplysia. The activity of the lumbo-sacral dorsal and ventral roots were little affected by inspiration CO2. Spinal motoneurones in cats inspiring 5% and 10% CO2, exhibited mainly decreases in excitability (Papajewski et al., 1969). In anaesthetized, sino-aortic denervated rats, inspiration of 10% CO2 for 45 minutes caused a decrease in 2-deoxyglucose uptake in gray matter, but increases in uptake in the V L M , dorsal hypothalamic area, and the thoraco-lumbar cord (Ciriello et al., 1985). It was suggested that the increases were possibly due to the direct effect of CO2 on these structures, or to the effects of CO2 excited structures on other structures i.e. indirect effects. It is also possible that this effect was caused by hypercapnic induced blood flow changes. Certainly, in ventral medullary slices in vitro, decreases in pH from 7.45 to 7.2 resulted in increases in activity of these neurons (Fukuda et al., 1980). Therefore, CO2 and H could have excited neuronal structures other than the classical chemoreceptors, but there has not been much evidence to support this notion in mammals. The possibility exists also, that CO2 has a depressant effect on inhibitory neurons resulting in excitation of effector neurons. Hypercapnia may have effects at the synaptic terminal. Decreasing pH has sometimes reduced the vasoconstrictive effects of sympathetic stimulation and catecholamine infusion. The pressor effects of i.v. injections of A and N A were reduced during hypoventilation induced hypoxia and acidosis in acute cervical spinal cats (Gowdey and Patel, 1964). Hypercapnic and or metabolic acidosis equivalently reduced the pressor response to noradrenaline and to other alpha-adrenergic agonists. This suggested that it was the decrease in pH and not CO2 per se, which altered vascular reactivity. The vasoconstrictor effects of adrenergic agonists were also reduced in the pithed and reserpinized rats, which implied that the mechanism was at least partly a peripheral one. The vasoconstrictor responses to sympathetic nerve stimulation were also reduced but not to the same extent as the initially equivalent responses to N A . Vasoconstrictor responses to sympathetic nerve stimulation were also reduced in the rabbit ear artery, by acidaemia but the responses were reduced to a lesser extent than were those to N A (Dusting and Rand, 1974). Therefore these authors suggested that acidaemia had an action on noradrenergic transmission which compensated for the reduction of the vasoconstrictor response to NA. One method of compensation would be to increase the availability of NA to the receptors, which could be accomplished by reducing N A uptake. These authors found that low pH reduced the inactivation of N A through a mechanism that was blocked by cocaine, in addition to reducing the post-junctional vasoconstrictor effect The use of [ ° H ] N A suggested that decreased pH enhanced the release of N A from the nerve endings, which could have also accomplished compensation for reduced post-junctional vasoconstriction. However, sympathetic stimulation may have also released vasoconstrictive co-transmitters such as neuropolypetide (NPY), whose 21 E. Hypercapnia and the AVP and Renin Angiotensin Systems Hypercapnia has been shown to affect not only the sympatho-adrenal (SA) system but also the arginine vasopressin (AVP) and renin angiotensin (RA) systems. Recent evidence has suggested a more important function for these systems in the control of SAP and regional blood flow, particularly with respect to A V P . Therefore the role of these systems in the control of hindlimb vascular resistance in response to hypercapnia must be considered. ln the absence of one or two pressor systems, the remaining system(s) attempt to maintain SAP. In concious rats individual inhibition of each system using antagonists caused no marked changes in SAP (Paller and Linas, 1984). Inhibition of A V P and RA resulted in no change in SAP and increased plasma levels of N E . Alpha adrenergic and A V P blockade also resulted in no significant change in SAP. A combination of alpha adrenergic and RA blockade resulted in a drop in SAP and no significant increase in plasma vasopressin (pAVP). Therefore it was further suggested that A V P release by an alpha adrenergic mechanism was neccessary for maintenance of SAP in the absence of the RA and alpha adrenergic systems in this model. In unanaesthetized rats subjected to adrenalectomy five days earlier, M A P was found to be lower than in the control animals while resting levels of A V P and renin were higher than controls (Elijovich et al., 1983). M A P was further reduced by either captopril or A V P antagonists while these had no effect on M A P in the control animals. Although they suggested hypotension as a possible stimulus for the increased level of A V P and renin, other factors may also have been involved such as reduced osmolality and reduced glucocorticoid levels. In concious unanaesthetized anephric rats it was found that the pAVP levels did not exhibit a clear relationship to SAP. However, A V P antagonism produced decreases in SAP which were most marked during alpha 1 adrenergic blockade, and also present during alpha2 adrenergic and beta adrenergic blockade. A V P is thought to be the strongest vasoconstrictor known (Altura and Altura, 22 1976) but in order to produce pressor effects it must be present in unphysiological plasma concentrations in non-pathological situations. This is because A V P may sensitize the baroreceptors thereby negating its vasoconstrictor effect (Cowley et al., 1974). Angiotensin II is a more potent pressor, but less potent vasoconstrictor than A V P (Osborn et al., 1987). In New Zealand white rabbits it has been shown that A V P enhances baroreflex inhibition of lumbar sympathetic activity by two mechanisms (Schmid et al., 1985). Firstly, it shifts the stimulus response curve to the left so that a lower carotid sinus pressure results in a given level of sympathetic inhibition. Secondly, A V P has a central facilitatory effect on the baroreflex. Angiotensin II centrally attenuates baroreflex inhibition of its vasoconstrictor effect. In conscious dogs, infusions of A V P (200, 1000, 5000 fmol min"^ kg""*-) produced larger increases in aortic pressure and TPR in cardiac denervated animals as compared to controls (Wang et al., 1987). The effects of angiotensin II and phenylephrine were comparable in both the cardiac innervated and denervated groups. The authors suggested that A V P elicits more effective reflex mechanisms possibly by potentiation of a vasodepressor reflex arising from cardiac receptors. Experiments where the sympathetic system was compromised supports these ideas. In conscious rats ganglionic blockade resulted in a 60-fold increase in the pressor sensitivity of A V P (Robinson et al., 1986). A study of two patients with Shy-Drager's syndrome suggested that infusion of physiological levels of A V P increased SAP more than did N E or Angiotensin II, with little change in HR indicating a lack of baroreflex activity (Mohring et al., 1980). Patients with progressive autonomic failure showed that infusions of A V P (resulting in pAVP levels from 0.8 to 30.0 pmol/1) produced a pressor response at a p A V P level of 5.5 pmol/1 (Williams et al., 1986). In normal controls, p A V P levels of 3000-4000 pmol/1 produced no pressor responses. In conscious dogs, muscarinic and ganglionic blockade potentiated the effects of infusions of phenylephrine (PE), and physiological levels of A V P (Robinson, 1987). The potentiation was greater for A V P and the authors suggested the change in baroreflex sensitivity during A V P administration was due to a greater level of parasympathetic tone to the heart in response to a given pressure 23 stimulus. In humans with mid cervical spinal cord transection, A V P infusions (0.05-2.0 pmol mm"* kg" )^ resulted in a decrease in HR and an increase in SAP (Frankel et al., 1986). The parasympathetic baroreflex pathway was intact and therefore the increase in SAP was suggested to have been due to a failure of the baroreflex sympathetic pathway in response to A V P or an increase in the sensitivity of the pressor vascular effects to A V P . Finally, A V P may have a direct effect on ganglionic transmission. In the superfused rabbit superior cervical ganglion, A V P inhibited ganglionic transmission (Wali, 1984). In the acute cervical spinal rabbit, infusion of A V P reduced renal nerve activity, but did not affect pre-ganglionic lumbar nerve activity (Imaizumi and Thames, 1986). These authors also suggested an inhibitory effect of A V P on ganglionic transmission. However several other explanations were possible such as the activation of a spinal neural inhibitory reflex by the increase in SAP produced by A V P (e.g. as discussed in the experiments by Fernandez de Molina and Perl, 1965). The increases in vascular resistance due to A V P have varied depending on the vascular region studied. In concious dogs A V P infusions resulting in increases in pAVP of 1 lpg/ml, caused decreases in cardiac output, skeletal muscle and skin blood flows using the microsphere technique (Liard et al., 1982). In this study, dehydration increased pAVP by 6pg/ml and caused decreases in cardiac output and skeletal muscle blood flow indicating the sensitivity of muscle to the constrictor effect of A V P . Using the washout method in healthy humans, infusions of A V P to obtain pAVP levels from 1.6 to 8.7pg/ml decreased skeletal muscle and skin blood flow, decreased PRA but did not significantly alter SAP or HR (Hammer and Skagen, 1986). It has been suggested that the amount of vasoconstrictor effect of A V P in different regions varies depending on the endogenous vasomotor tone due to the RA and sympathetic systems (Tabrizchi et al., 1986). In anaesthetized rats, A V P antagonism caused decreases in SAP and TPR which were greater in animals subjected to prior alpha adrenergic or RA blockade. This is similar to results in concious animals although the effects of A V P blockade may be enhanced in the anaesthetized, surgically stressed animals due to elevated levels of AVP. Without 24 concomitant blockade of other systems, A V P antagonism increased blood flow to the stomach and skin. In the absence of the RA system, A V P exerts its influence primarily in the vascular beds of the skin and muscle. Most importantly, in the absence of the alpha adrenergic system A V P exerts its greatest vasoconstrictor role in the vascular beds of skeletal muscle. It is therefore apparent that the RA and A V P systems are important for maintenance of vasomotor tone and SAP. A V P seems to have an augmented effect in pathological situations particularly where the sympathetic nervous system is compromised. In addition A V P seems to be important for regulation of skeletal muscle blood flow in all conditions. It has been been demonstrated that the RA and A V P systems respond to hypercapnic acidosis. In conscious sheep, hypercapnic acidosis (pH=7.1-7.2, PCC>2= 60-70mm Hg) caused significant increases in PRA after 30 minutes (O'Brodovich et al., 1982). Anaesthetized dogs subjected to hypercapnic acidosis (pH = 7.1, P C 0 2 = 75mm Hg) also showed significant increases in PRA after 30 minutes (Wang et al., 1984). These increases were thought to be a result of both neural and non-neural factors. In anaesthetized dogs, hypercapnic acidosis (pH = 6.87, P C 0 2 = 107.7) resulted in significant increases in PRA which were reduced but not eliminated by renal nerve section (Fujii, et al., 1985). A similar study in anaesthetized dogs subjected to less severe hypercapnic acidosis (pH= 7.02-7.24, PC02= 47-70mm Hg) resulted in increases in PRA which were reduced by renal denervation and beta adrenergic blockade (Andersen, et al., 1980). Further studies in dogs subjected to hypercapnic acidosis (pH= 7.2) resulted in significant increases in PRA after 20 minutes both before and after chronic carotid chemodenervation (Raff et al., 1984). Thus, the increase in renin seems to be dependent on a beta adrenergic component initiated by the action of CO 2 on a site other than the carotid chemoreceptors in this situation. In anaesthetized dogs, inhalation of 12% CO2 for 10 minutes resulted in increased renin release from denervated and innervated kidneys (Kurz et al, 1978). 25 Adrenalectomy eliminated the increase in renin release from the denervated kidney in this study, which suggested a role for the adrenal gland in mediating the response to hypercapnia. These authors also found increases in renin after 30 minutes of inhalation with 4%, 8% as well as 12% CO^. This suggested even low levels of hypercapnia could release renin, but they presented no evidence that the renin release in response to these lower levels of CO2 (4% and 8% CO2) occurred before 30 minutes. These authors presented no evidence that renin release in response to 4% or 8% inhaled CO2 depended on the adrenals. The neural and adrenal components demonstrated in the above experiments, may or may not affect non-neural factors which can influence renin release such as altered renal hemodynamics, electrolyte handling or direct humoral effects. Finally, both the increases in renin and increases in angiotensin II, seen in anaesthetized dogs and blocked by propranolol, have been responsible for the late pressor effects seen with hypercapnia which implied a role for the RA system in hypercapnia induced cardiovascular changes (Staszewska-Barczak, 1976). Therefore renin is probably released during hypercapnic acidosis, and may play a role in the hemodynamic changes which occur in this condition. However, is not clear whether small increases in P a C 0 2 for relatively short periods of time can influence the RA system and affect hemodynamics, particularly when the sympathetic nervous system is compromised. Systemic hypercapnia has also been shown to release AVP. In anaesthetized dogs, inhaling 6% CO2 for 20 minutes caused pAVP to increase only at PO2 levels of 80mm Hg or less (Raff et al., 1983). Also in anaesthetized dogs, at PC02= 75mm Hg and pH= 7.1, both pAVP and CSF-AVP increased after 30 minutes (Wang et al., 1984). In anaesthetized, spontaneously breathing cats, 5% inspired CO2 significantly raised pAVP after only 5 minutes but only if ventilation increased concommitantly by 15% or greater. So although A V P is released during hypercapnia, the minimum levels of CO2 and time required is equivocal. The relative contributions of pH and PCO2 to the release of A V P is 26 not clear. Wang et al. (1984) have demonstrated that metabolic acidosis produced by infusions of hydrochloric acid results in a greater increase in pAVP than in hypercapnia, although the pH levels in both situations were the same. The levels obtained with acidosis were similar to those obtained with combined hypoxia and hypercapnia, indicating that the effect may be an indirect one via pH changes. Metabolic acidosis did not cause any changes in C S F - A V P , which may indicate A V P release by hypercapnia and hypoxia is a result of a more complex interaction. In anaesthetized dogs which received infusions of sodium bicarbonate to maintain pH constant, PCO2 = 69mm Hg resulted in an increase in pAVP only after 90 minutes (Coren et al., 1987). Several factors have been suggested to cause this release including chemoreceptor mediation and hemodynamic effects such as general hypotension and neurohypophyseal blood flow changes. Since more than one mechanism seemed to be involved, pH, PCO2 and PO2 may have a variety of effects on A V P release. The effects of hypercapnia released A V P on hemodynamic variables has recently been looked at in concious rats (Walker, 1987). At P C 0 2 = 53mm Hg, pH= 7.35, HR decreased, SV increased and SAP remained unaltered for up to 60 minutes. V-^ vasopressinergic blockade at 10 minutes produced no effects on these variables. At P C 0 2 = 108mm Hg, pH= 7.2, HR and CO decreased while TPR and SAP increased. V1 vasopressinergic blockade at 10 minutes caused decreases in both TPR and BP, with an increase in CO due to an increase in SV at 20 minutes. Therefore during severe acute periods of hj^percapnia, A V P could have been released in quantities sufficient to cause significant changes in SAP,TPR and SV. This does not preclude the possibility that moderate hypercapnia could have released A V P in amounts sufficient to cause significant changes in regional vascular resistance which will not be manifested in TPR or SAP variables. 27 III. PURPOSE OF THE EXPERIMENTS Whether there are other areas of CO 2 sensitivity which mediate cardiovascular changes, in addition to peripheral and ventral medullary chemoreceptors, has remained an open question. These experiments were designed to answer the question of whether systemic hypercapnia and normal levels of CO 2 caused increases in vascular resistance of vascularly isolated hindlimbs in acute cervical spinal cat and to assess the contributions of the lumbar sympathetic system, and of the adrenals in CO2 mediated hindlimb vascular tone. This preparation eliminates any cardiac effects and illustrates the effects of CO2 specifically on hindlimb vascular tone. More generally, these experiments would illustrate whether normal levels of CO 2 could be responsible for a portion of the neurogenic vascular tone which remains after acute spinal transection. However, because these experiments utilize systemic hj^percapnia, the site of action of CO 2 was not localized. The neurogenic actions were attributed to spinal structures e.g. spinal excitatory or inhibitory interneurons, hypothetical chemoreceptors, SPNs, sympathetic ganglia, or sympathetic afferents. Also, because the responses is an effector organ response, there is a certain amount of processing that likely probably goes on along the pathway, such as in the sympathetic ganglia (Bulgyin, 1983). 28 IV. METHODS A. Rationale In these experiments, the hindlimbs of spinal cats were vascular ly isolated, and perfused at constant flow. The changes in hindl imb perfusion pressure (PP) therefore reflected associated changes in hindlimb vascular resistance (VR) . Attenuat ion of P P and V R responses to hypercapnia with hindlimb sympathetic denervation, ascertained whether the changes were neurogenic, and of sympathetic origin. H i g h cervical spinal transection carried out in these experiments eliminated sympathetic efferent activity originating or having a relay point above the transection, such as from classical receptors, (e.g. medullary and peripheral chemoreceptors, and arterial baroreceptors). Thus, although this experimental preparation did not localize the site of action of C0 2 to any specific structure, i t did attribute this action to structures, or systems, other than the classical supraspinal systems (e.g. to spinal chemosensitive regions). The transection was made in the cervical region to comple te^ disconnect the pre-ganglionic neurons from the medullary vasomotor centres. A transection at cervical segment 3 could be accomplished quickly, rel iably, and easily. The spinal cord could be transected completely wi th little bleeding using only a blunt spatula , leaving most of the cord intact. Rather than pharmacological blockade, unilateral lumbar sympathectomy was performed so that one leg of one animal could be selectively denervated. Thus, the effects of systemic hypercapnia on the P P and V R of both the denervated and the innervated leg was looked at simultaneously i n the same animal . This method also eliminated the potential unwanted side effects of circulating pharmacological blockers, for example central and renal effects. 29 Hyperventilation in 100% 0 2 was utilized during each experimental run to produce hypocapnia. In previous experiments, the animals continued hyperventilating between experimental runs, with 5% C0 2 to maintain appropriate blood gas levels. However, positive pressure hyperventilation mechanically reduces venous return thereby reducing cardiac output and SAP. This effect could have been especially severe in the cardiovascularly depressed spinal cats and the animals were therefore nor mo ventilated in 100% 0 2 between experimental runs. Mechanical ventilation and reduction in SAP could have caused release of humoral vasoactive agents over this prolonged period which may have changed the reactivity of the hindlimb vessels over time. 30 B. Animals and Induction of Anaesthesia Experiments were performed on twenty adult, male, mongrel cats weighing between 3.5-6.0 kg. The animals were given half of their food allotment two days prior to the experimental day and fasted one day prior to the experimental day. Animals showing signs of illness were not used, or deferred for later use. Anaesthesia was induced with halothane (Fluothane, 4% in oxygen) via a small animal anaesthesia apparatus (Foregger, Model No. SA-10), and was maintained with a mixture of 2% alpha-chloralose (0.006g/kg) and 30% urethane (0.3g/kg) injected i.p. (total volume less than 20mls). Alpha-chloralose (2g) was dissolved in a 3% sodium tetraborate saline solution (lOOmls), at room temperature in order to reduce by-product formation. By-product formation could have occurred when dissolving the alpha-chloralose in saline by heating, particularly at temperatures greater than 6 0 ° C. By-product formation could have reduced the effective dose of alpha-chloralose and produced some unwanted side effects associated with this type of anaesthesia. C. Surgical Procedures During each procedure, tissue bleeding was carefully controlled with an electric cautery, and sutures. Incisions were covered with saline-soaked gauze,and appropriate infusions of dextran were given, to replace the small amounts of blood lost during surgery. Rectal temperature was measured and the animals were maintained at 3 7 ° C with heating lamps and a heating pad attached to a servo-control unit (Tele-Thermometer, Model 73, Yellow Springs Co. Inc.). After tracheotomy, and tracheal tube insertion, the right carotid artery, and the right jugular vein were exposed and cannulated with tubing (P.E. 200, Clay-Adams) filled with heparinized saline (lOOOunits/ml saline). The right carotid artery was used for determination of blood gases and blood pressure, and the right jugular vein was used for drug and fluid infusion. 31 The right femoral nerve was isolated, sectioned and covered with warm mineral oil. Femoral nerve isolation was performed for later afferent stimulation to determine spinal reactivity after spinal transection. The animals were then placed in a stereotaxic device for spinal transection. Respiration was maintained with a Harvard respirator, and pancuronium bromide (Pavulon, 0.05mg/kg) was administered for neuromuscular blockade . Midline incisions were made from the interparietal bone of the skull to just above the shoulders. The fascia and muscle were cut and retracted until the second and third vertebrae were exposed. A laminectomy of the third cervical segment was performed to •open a window to the spinal cord. The dura was cut and the cord was injected with mepivacaine hydrochloride (carbocaine, 1%) to reduce spinal stimulation during transection. After SAP had stabilized (20-40s), the cord was transected with a blunt spatula after which gelfoam was placed in the transected area for hemostasis. Saline soaked gauze was placed over the transection and the incision was closed with surgical staples. The animals were removed from the stereotaxic frame and prepared for abdominal surgery. To maintain blood sugar, blood pressure, appropriate blood CO2 and blood pH levels, and neuromuscular blockade, a continuous i.v. infusion of 10% dextrose and 6% dextran in saline (12ml/hr) was started containing pancuronium bromide (0.2mg/kg/hr), and sodium bicarbonate (lmEq/lg^p/hr). E C F was calculated as 20% of body weight. A mid-line laparotomy was performed, from just below the sternum to the lower abdomen. The bladder was exposed, and cannulated with tubing (P.E. 210), to allow proper urine flow and reduce changes in bladder distension which could have reflexively altered hindlimb vessel tone. The large and small bowel were moved aside and placed in a plastic bag with saline soaked gauze to keep them warm and moist, after which the abdominal aorta was exposed from the renal artery to the aortic bifurcation. Every branch of the abdominal aorta, from (but not including) the renal arteries to the aortic bifurcation, was ligated and cut. The mid sacral artery was ligated and cut, while the deep femoral arteries 32 were ligated only. Next, the left sympathetic trunk was cut at the level of the renal vein and stripped down to the aortic bifurcation (L1-L7). Bilateral adrenalectomy was performed in eight cats after which they were administered hydrocortisone sodium succinate (HSS), (Solu-Cortef, 0.5g) to compensate for reduced glucocorticoid levels. HSS was also added to the infusion solution for these animals (0.06g/hr). D. Hindlimb Perfusion Preparation (figure 1) Each experiment required two pumps (Cole-Parmer, Masterflex Model 7520-20) with single heads (Model 7014.21), fitted with silicon tubing (Cole-Parmer, Type 6411.14., inside diameter .064"). The system had a volume of 9mls for each perfusion line and was primed with heparin and dextran. Heparin (3ml, 1000 units/ml) was administered to the cats and the perfusion system was connected. The abdominal aorta was centrally cannulated with tubing (P.E. 210) to obtain blood from the upper body. Each external iliac arterj' was peripherally cannulated with tubing (P.E. 190) to pump blood to the hindlimbs. The blood returned to the upper body via normal venous channels. The cannulas were secured, the abdomen was covered with saline-soaked gauze, and the animals were allowed to recover for a period of 30-60 minutes, until the perfusion pressures were stable. The pumps delivered constant flow against pressures of up to 250mm Hg. The pressure drop across the iliac cannulas was less than 1mm Hg with flows up to 30 mls/min. Hindlimb vascular isolation from the upper body was considered to be satisfactory if the perfusion pressures dropped to less than 15mm Hg (well below SAP) when both pumps were stopped for approximately thirty seconds. Likewise, collateral circulation between hindlimbs was considered to be insignificant if the perfusion pressure of each leg dropped to below 15mm Hg when only its pump was stopped for thirty seconds. 33 Figure 1 Schematic representation of the preparation used in these experiments. A A , Aortic Artery; C A , Carotid Artery; EIA, External Iliac Artery; J V , Jugular Vein; T, Trachea; SAP, Systemic Arterial Pressure; PP, Perfusion Pressure. 35 E. PCOg, POg, and pH Maintenance End tidal carbon dioxide levels were continuously monitored by a carbon dioxide analyzer assembly (Beckman Instruments, Model LB-1), via a tracheal side needle and heated sampling tube, and displayed on a chart recorder (Gould, Model ES 1000). For blood gas and pH determination, samples of arterial blood (0.5-0.7ml) were collected in 1ml syringes, containing heparin-saline (0.1ml), and analyzed using a blood gas analyzer (Corning, Model 165/2). The blood was replaced by an equal volume of 6% dextran in saline, or saline. The animals were ventilated with room air enriched with 100% oxygen using a demand valve, thereby maintaining PO2 levels •above 150 mm Hg. Ventilation rate, and tidal volume were adjusted ( 20-25 strokes per min, and 40-60 cc/breath at an inspiration/expiration ratio of 40/60) and appropriate infusions of sodium bicarbonate solution (lmEq/ml/lgQp) were given, to maintain pH and carbon dioxide at physiological levels (PC0 2= 25-30 mm Hg, pH= 7.35-7.45), (Fink and Schoolman, 1963; Herbert and Mitchel, 1971). F. Blood Pressure Measurement Systemic pulsatile and mean pressures were measured via cannulae placed into the right carotid artery and advanced to the aorta. Hindlimb perfusion pressures were measured on each perfusion line less than 10cm from the entry of the cannula into the external iliac artery. All pressures were measured by strain gauges (Statham, P23Db), which were flushed with heparin-saline and connected to a transducer/converter, an oscilloscope (Tektronix, Model 5443) and a chart recorder (Gould, Model ES 1000). The system was calibrated with a mercury manometer at the beginning of each experiment, and for true zeros the transducers were adjusted to the level of the heart. 36 G. Determination of Spinal Reactivity The central stump of the femoral nerve was stimulated with electrodes attached to a stimulator (Grass, Model S8) to ascertain whether the spinal cord was unreactive due to anoxic damage or insufficient recovery time after spinal transection. In spinally intact animals, two sets of stimulus parameters were used, the first of which (2V, 5ms, 5-7Hz) caused reflex hypotension, and the second of which (15V, .5ms, 20-25Hz) caused reflex hypertension. Three to four hours after spinal transection, both sets of parameters sometimes caused small increases in the PP of the innervated leg. These stimulations were performed in the first several animals but were discontinued in subsequent cats because of inconsistent PP responses and when it became apparent that changes in PP occurred in response to carbon dioxide, beginning immediately after the experimental preparation was complete. H. Experimental Protocol Each experimental run consisted of the following procedures. The respiratory pump rate was set at 40 strokes per minute for all animals and the tidal volume was adjusted on the first run (50-75 cc/breath) to obtain P & C 0 2 levels of 16-20mm Hg and was not further altered. Therefore the tidal volume varied between animals but the pump rate and inspiratory/expiratory ratio was constant for all animals. The animals were hyperventilated in room air enriched with 100% 0 2 for 1-1.5 minutes until the systemic and perfusion pressures were stable, at which point blood gas samples were collected. Five or ten per cent CO2 in O2 was administered via a demand valve for five minutes after which blood gas samples were collected. Two to four runs were randomly carried out at each FjC02 in each animal, separated by at least fifteen minutes to allow for pressure stabilization and correction of acid-base status when necessary. In some animals, the initial 1-1.5 minute period of hyperventilation with 100% O2 was extended by 5 minutes and CC"2 was not administered. This was to provide several time controls for the effects of hypocapnia and the other effects of hyperventilation (e.g. the reduction of venous return). 37 I. Additional Procedures Two non-adrenalectomized animals were maintained and prepared as described except one leg was sympathetically denervated after the pumps were attached. CO2 was not administered and the animals were left for 2-4 hours normoventilating with room air enriched with 100% This was to study the changes in VR that may have occurred after denervation over the experimental time period. In one animal the flow rates were adjusted to be initially (before denervation) the same, while in the other animal flow rates were adjusted to obtain similar initial perfusion pressures. One adrenalectomized animal was maintained and prepared as described, except it did not undergo spinal transection. Afferent femoral nerve stimulation, carotid clamping, and administration of 5% CO2 (as described) were performed on this animal. 38 V . D A T A A N A L Y S I S A N D I N T E R P R E T A T I O N A . Perfusion Pressure and Vascular Resistance The PP response to increases in blood CO2 tensions was biphasic. Therefore PP was measured and recorded at the end of hyperventilation with 100% O2, at each peak and at the nadir of each experimental run. Vascular resistance (VR) values were calculated from those respective PP values using the following formula: VR(mmHg min ml"1)= PP (mmHg)/Blood Flow(ml min"1) B. Systemic Variables Increases in blood CO2 tensions most often produced a decrease followed by an increase in SAP. Therefore, heart rate (HR), systolic pressure (SP), diastolic pressure (DP) and mean arterial pressure (MAP) were recorded at the end of hyperventilation with 100% O2, at the minimum level of SAP during CO2 administration (minimum values) and at the subsequent maximum level SAP after 5 minutes of CO 2 administration (end values). HR, SP, DP and M A P were also measured and recorded during normoventilation with 100% 0 2. C. Experimental Groups and Statistical Comparisons The cats were divided into adrenalectomized and non-adrenalectomized groups. All cats had one leg sympathetically denervated, and were exposed to 5% and 10% CO2 thereby creating eight experimental situations. 39 Sample variances for each experimental situation was not significantly different using the Fmax-crit test to approximate this difference (Siegel, 1956). However, non-parametric statistics were chosen because sample sizes were 8 or less, and therefore normal distribution of the response values could not be ascertained reliably. Where each animal acted as its own control for all comparisons, the results were analyzed using the Wilcoxon paired signed ranks analysis for significance. For comparisons made between non-adrenalectomized and adrenalectomized animals, the results were analyzed using the Mann-Whitney analysis for significance. Differences were accepted as significant at p< .05 level for all statistical tests employed. The following comparisons were made: a) PP and VR values during hyperventilation with 100% 0 2 were compared to PI and P2 PP and VR values during hyperventilation with 5% and 10% C 0 2 . b) PI and P2 PP and VR values during lyperventilation with 5% C 0 2 were compared to those values during l^perventilation with 10% C 0 2 . c) All measured PP, VR, HR, SP, DP and M A P values in non-adrenalectomized cats were compared to those values for adrenalectomized cats. d) Values for HR, SP, DP and M A P during hyperventilation with 100% 0 2 were compared to those values during hyperventilation with 5% and 10% C 0 2 . e) Values for HR, SP, DP, and M A P during hyperventilation with 5% C 0 2 were compared to those values during 10% C 0 2 . f) Values for p H & and P a C 0 2 during hyperventilation with 0,5 and 10% C 0 2 were compared. g) Values for p H & and P & C 0 2 during hyperventilation with 0,5 and 10% C 0 2 for non-adrenalectomized and adrenalectomized animals were compared. 40 The prominent nadir values of PP and VR, and the minimum values of HR, SP, DP and M A P were used only to qualitatively describe trends in the P2 PP, end HR, end SP, end DP, and end M A P responses, and to illustrate the vasodilatory responses to CO2 administration. 41 VI. RESULTS A. Induction of Hypercapnic Acidosis In non-adrenalectomized and adrenalectomized cats hyperventilating with 5% CO2 and 10% CO2 produced significant differences in p H a and P &C02 values compared to those values for hyperventilating with 100% O2 (tables 1 and 2). The p H & and P &C02 values during hyperventilation with 10% CO2 were significantly greater than during hyperventilation with 5% C0 2. The p H & and PCO2 values in adrenalectomized cats were not significantly or observably different from those values in non-adrenalectomized cats. P^CC"2 reached hypercapnic levels within 25 seconds of CO2 administration, and at the end of each run required a similar amount of time to reach pre-CO^ administration levels. . B. Perfusion Pressures 1. General The PP of both legs were unstable and low when the pumps were initially attached. The PP increased and stablized 30-45 minutes after pump attachment. Hyperventilation in 100% O2 caused initial fluctuations in the PP of both legs in all animals. These pressures stabilized between 1-1.5 minutes, at lower levels than those seen during normoventilation with 100% ©2- In several animals, hyperventilation with 100% O2 was carried out for 7-7.5 minutes and this resulted in no further change in the PP of either leg, or the SAP of the animals (e.g. figure 2). After CC>2 administration was stopped, the PP of both legs required from 10-20 minutes to stabilize. Leaving the adrenals intact, and particularly inspiring 10% CO2 seemed to prolong this period. TABLE I PCO 2 (mm Hg) pH (pH u n i t s ) Hypervent i1 a t 1 on In 100% 0 2 16 .36 + .84 ( n - 8 ) 7 , .542 + .036 ( n - 8 ) Hypervent11 at1on 1n 5% C 0 2 37 . 4 8 + 1 . 0 3 a (n = 8) 7 , .314 + .020 a (n=8) Hypervent i1 a t i on i n 10% C 0 2 62 .23 + 3 . 2 3 a b (n = 7) 7 . 143 + . 0 2 3 a b (n = 7) V a l u e s f o r PC0 2 and pH In n o n - a d r e n a l e c t o m i z e d c a t s d u r i n g h y p e r v e n t i l a t i o n 100% 0 2, h y p e r v e n t i l a t i o n i n 5% C0 2, and h y p e r v e n t i l a t i o n i n 10% C 0 2 . A l l v a l a r e mean + S.E. S t a t i s t i c a l l y d i f f e r e n t from h y p e r v e n t i l a t i o n In 100% 0.. S t a t i s t i c a l l y d i f f e r e n t from h y p e r v e n t i l a t i o n i n 5% C0 2. 43 TABLE I I PCO 2 (mm Hg) PH (pH u n i t s ) H y p e r v e n t i l a t l o n i n 100% 0 2 17.16 + .66 (n=8) 7. .569 + .016 (n=8) H y p e r v e n t i l a t i o n i n 5% C 0 2 37 .96 + 1 .21 a (n = 8) 7 . 312 + .027 a (n=8) Hypervent11 at ion i n 10% C 0 2 64.20 + 1.55 a b (n-7) 7 . 179 + . 0 1 7 a b (n=7) V a l u e s f o r PC0 2 and pH i n a d r e n a l e c t o m i z e d c a t s d u r i n g h y p e r v e n t i l a t i o n i n 100% 0 2, h y p e r v e n t i l a t i o n i n 5% C0 2, and h y p e r v e n t i l a t i o n i n 10% C 0 2 . A l l v a l u e s a r e mean + S.E. a S t a t i s t i c a l l y d i f f e r e n t from h y p e r v e n t i l a t i o n i n 100% 0 2 . b S t a t i s t i c a l l y d i f f e r e n t from h y p e r v e n t i l a t i o n i n 5% C0_. Figure 2 Chart recording of the responses of systemic arterial pressure and perfusion pressures of both legs to 7 minutes of hyperventilation with 100% O^, in a non-adrenalectomized animal. PP den., Perfusion Pressure of the denervated leg; PP inn., Perfusion Pressure of the innervated; SAP, Systemic Arterial Pressure; PCO2, Partial Pressure of CO2 in mm Hg. 45 P C O o - 2 4 . 5 P C 0 2 - 1 5 . 6 \ P C 0 2 - 1 2 . 2 \ 46 2. Non-adrenalectomized Spinal Cats Increases in P a C 0 2 produced biphasic PP responses in the innervated leg (e.g. figure 3). Peak 1, and peak 2 PP responses to 5% and 10% C 0 2 were significantly different from the PP seen during hyperventilation with 0% C 0 2 (figures 4 and 5). Increases in P a C 0 2 produced similar biphasic PP responses in the denervated leg (e.g. figure 3). The peak 1 PP response to 5% and 10% C 0 2 , was present but was not significantly different from the PP response to 0% C 0 2 - Peak 2 PP response to 5% and 10% C 0 2 was significantly different from the PP seen during hyperventilation with 0% C 0 2 (figures 4 and 5). 3. Adrenalectomized Spinal Cats Increases in P a C 0 2 produced biphasic PP responses in the innervated leg (figures 6 and 7). The peak 1 and peak 2 PP responses to 5% and 10% C 0 2 were significantly different from the PP response to 0% C 0 2 . Increases in P a C 0 2 again produced biphasic PP respones in the denervated leg (figures 6 and 7). The peak 1 PP responses to 5% and 10% C 0 2 were present but only the response to 10% C 0 2 was significantly different from the PP response to 0% C 0 2 . Peak 2 PP responses to 5% and 10% C 0 2 were present but only the response to 5% C 0 2 was significantly different from the PP response to 0% C 0 2 All PI and P2 PP responses were observably (but not significantly) reduced in comparison to those responses in the non-adrenalectomized cats. C . Vascular Resistance and Pump Flow Rates Comparisons of VR responses during administration of 5% and 10% C 0 2 to VR responses to administration of 0% C 0 2 produced similar results to the comparisons of PP responses with one exception (figures 8,9, 10 and 11). In the denervated legs of the adrenalectomized animals 10% C 0 2 produced a very small PI VR response which was not significant (figure 10). Also, unexpectedly the values of V R in response to 0% C 0 2 were observably (but not significantly) greater in the denervated leg compared to the innervated leg unlike the PP responses which were similar. Adrenalectomy seemed to enhance this apparent difference. Figure 3 Chart recording of the responses of systemic arterial pressure, and perfusion pressures of both legs to hyperventilation with 10% inspired CO2 in a non-adrenalectomized animal. PP den., Perfusion Pressure of the denervated leg; PP inn., Perfusion Pressure of the innervated leg; SAP, Systemic Arterial Pressure; F j C C ^ , Inspired CO^; PI, Peak 1; P2, Peak 2; N , Nadir. 48 Figure 4 Perfusion pressure values for F j C 0 2 = 0% (B) and 5% (Pi, N and P2) in the innervated and denervated legs of non-adrenalectomized cats. a Statistically different from hyperventilation with 100% 0 2 -Figure 5 Perfusion pressure values for FjCC>2= 0% (B) and 10% (PI, N and P2) in the innervated and denervated legs of non-adrenalectomized cats. a Statistically different from hyperventilation with 100% o2-5% CO 50 LZD Innervated Denervated 10% co 2 i i Innervated IWN Denervated K 1 2 0 S 6 CO w PU o CO fx, w 100-1 80 60 40 20 0 a 1 a J L B PI N P2 ! B 1 PI N P2 51 Figure 6 Perfusion pressure values for F jC02= 0% (B) and 5% (PI, N and P2) in the innervated and denervated legs of adrenalectomized cats. a Statistically different from hyperventilation with 100% o 2. Figure 7 Perfusion pressure values for F jC02= ® a n c * (PI, N and P2) in the innervated and denervated legs of adrenalectomized cats. a Statistically different from hyperventilation with 100% o2-Adrenalectomized 5% C0 2 52 CD Innervated L\\N Denervated 120 n Adrenalectomized r — | innervated 10% C0 2 | ^ Denervated 120 - i 100-80-60-40-20-0 a a I I B PI N P2 B PI N P2 53 Pump flow rates were the same in the innervated legs of non-adrenalectomized and adrenalectomized cats and both were observably but not significantly higher than the flow rates in the denervated legs (table 3). 54 Figure 8 Vascular resistance values for FjC02= 0% (B) and 5% (PI, N and P2) in the innervated and denervated legs of non-adrenalectomized cats. a Statistically different from hyperventilation with 100% o2-Figure 9 Vascular resistance values for FjCC>2 = 0% (B) and 10% (PI, N and P2) in the innervated and denervated legs of non-adrenalectomized cats. a Statistically different from hyperventilation with 100% Oo. 56 Figure 10 Vascular resistance values for FjC02 = 0% (B) and 5% (PI, N and P2) in the innervated and denervated legs of adrenalectomized cats. a Statistically different from hyperventilation with 100% o 2. Figure 11 Vascular resistance values for F j C 0 2 = ® a n ( * ^ a n c * ^ ) in the innervated and denervated legs of adrenalectomized cats. a Statistically different from hyperventilation with 100% o2-Adrenalectomized 5% COo 57 I I Innervated ES Denervated 58 TABLE I I I Pump flow r a t e s (mls/min) I n n e r v a t e d Leg Denervated Leg No n - a d r e n a l e c t o m i z e d 1 2 + 1 1 0 + 1 (n=8) (n=8) A d r e n a l e c t o m i z e d 1 2 + 1 9 + 1 (n=S) (n=G) V a l u e s f o r pump flow r a t e s f o r the i n n e r v a t e d and de n e r v a t e d l e g s i n non-a d r e n a l e c t o m i z e d and a d r e n a l e c t o m i z e d c a t s . A l l v a l u e s a r e mean + S.E. 59 D. Systemic Hemodynamics 1. General Cervical (C2) spinal transection resulted in a transient increase followed by a decrease in SAP to approximately 65-70 mm Hg. SAP increased to stable levels after 10-30 minutes (tables 4 and 5, values for normoventilated animals) and remained more or less at these levels for the duration of the experiment. Recovery was improved if the transection was carried out slowly over 2-3 minutes and occurred more than one hour after surgery began. Hyperventilation with 100% 0 2 resulted in significant decreases in systolic pressure (SP), diastolic pressure (DP), and mean arterial pressure (MAP) and an observable (but not significant) increase in heart rate (HR) when compared to values of these variables during normoventilation with 100% 0 2 (table 4 and 5). The values of these systemic variables during normoventilation and hyperventilation with 100% 0 2 in adrenalectomized cats were not significantly or observably different from those values in non-adrenalectomized cats. After C 0 2 administration was stopped, the SAP required 10- 20 minutes to stabilize. Again, leaving the adrenals intact, and particularly inspiring 10% C 0 2 seemed to prolong this period. 2. Non-adrenalectomized Spinal Cats Increases in P a C 0 2 resulted in decreases (minimum values) followed by increases (end values) in all systemic variables (table 4). Only the end values were tested for statistical significance (see Data Analysis and Interpretation). Hyperventilation with 5% C 0 2 resulted in significant increases in DP and M A P , a non-significant increasing trend in SP, and no significant or observable change in HR compared to values for these variables during hyperventilation with 100% 0 2 . Hyperventilation with 10% C 0 2 resulted in significant increases in SP and M A P , a non-significant but observable increase in DP, and no significant or observable change in HR 60 compared to values for these variables during hyperventilation with 100% O^-The SP value during 10% C O 2 administration was also significantly larger than the SP value during 5% C O 2 administration. 3. Adrenalectomized Spinal Cats Increases in P aCC>2 resulted in decreases in all systemic variables (minimum values) followed by increases in SP, DP, and M A P and a decrease in HR (end values), (table 5). Only the end values were tested for statistical significance (see Data Analysis and Interpretation). However, hyperventilation with 5% and 10% C O 2 resulted in no significant differences in SP, DP and M A P compared to values for these variables during hyperventilation with 100% O2. Hyperventilation with 5% and 10% C O 2 did result in significant decreases in HR compared to values for HR during hyperventilation with 100% The value of HR during 10% C O 2 administration was also significantly lower than the HR value during 5% C O 2 administration. E . Additional Procedures In two animals, the changes in VR over time were monitored (table 6). The PP, in both legs of both animals, were unstable after pump attachment and after denervation of one leg.. The PP of both legs dramatically fluctuated during the denervation of one leg, in both experiments. In animal 1, where the flow rates were the same before denervation, VR was also initially similar and remained near the same throughout the experiment. Overall, the V R of both legs decreased during the experiment. In animal 2, where the PPs were the same before denervation, V R was initially similar. The VR of both legs in this animal increased during the experiment. However, while the VR of the innervated leg increased by 2.65 PRU, the VR of the denervated leg increased by 4.12 PRU as compared to the values of VR before denervation. 61 TABLE IV HR SP DP MAP (beats/mi n) (mm Hg) (mm Hg) (mm Hg) Normovent11 at 1 on 164+18 108+ 6 a 70+ 6 a 90+ 6 a In 100% 0 2 (n = 5) (n=7) (n = 7) (n = 7) Hypervent11 at Ion 170+11 84+ 8 46+ 8 60+ 8 i n 100% 0 2 (n = 7) (n=8) (n=8) (n = 8) Hypervent11 at 1 on 158+12 76+ 8 40+ 4 56+ 6 i n 5% C 0 2 (min) (n = 6) (n = 8) (n=8) (n = 8) Hypervent i1 at 1 on 164+14 92+ 8 56+ 8 a 70+ 8 a i n 5% C 0 2 (end) (n = 6) (n=8) " (n = 8) (n = 8) Hypervent11 at i on 160+16 72+ 6 40+ 4 54+ 6 i n 10% C 0 2 (min) (n = 4) (n = 7) (n = 7) (n = 7) Hypervent11 at Ion 164+20 106+10 a t J 62+ 10 82+ 8 a i n 10% C 0 2 (end) (n = 4) (n = 7) (n = 7) (n = 7) V a l u e s of HR, SP. DP and MAP of no n - a d r e n a l e c t o m i z e d c a t s d u r i n g n o r m o v e n t i l a t i o n w i t h 100% 0 2, h y p e r v e n t i l a t i o n w i t h 5% C 0 2 (minimum and end v a l u e s ) , and h y p e r v e n t i l a t i o n w i t h 10% C 0 2 (minimum and end v a l u e s ) . A l l v a l u e s a r e mean + 5.E. S t a t i s t i c a l l y d i f f e r e n t from h y p e r v e n t i l a t i o n i n 100% 0 2 . S t a t i s t i c a l l y d i f f e r e n t from h y p e r v e n t i l a t i o n i n 5% C 0 2 ( e n d ) . TABLE V HR SP DP MAP (beats/mln) (mm Hg) (mm Hg) (mm Hg) Normoventi1 at i o n 158+ G 1 16+ 6 a 82+ 4 a 98+ 6' 1n 100% 0 2 (n-7) (n=8) (n=8) (n = 8) Hypervent11 at 1 on 170+ 6 82+ 8 48+ 6 60+ 6 1n 100% 0 2 (n = 7) (n=8) (n=8) (n=8) Hypervent11 at 1 on 156+ 8 72+10 46+ 8 58+ 8 1n 5% C 0 2 (m1n) (n=7) (n = 8) (n = 8) (n=8) Hypervent11 at 1 on 154+ 8 a 82+10 50+ 8 66+10 In 5% C 0 2 (end) (n = 7) (n=8) • (n = 8) (n=8) Hypervent11 at 1 on 148+ 8 72+10 44+ 8 56+ 8 1n 10% C 0 2 (m1n) (n = 6) (n = 7) (n = 7) (n = 7) Hypervent11 at i o n 138+10 3 b 86+14 54+10 68+12 i n 10% C 0 2 (end) (n=5) (n=6) (n = 6) (n=6) V a l u e s of HR, SP, DP and MAP of a d r e n a l e c t o m i z e d c a t s d u r i n g n o r m o v e n t i l a t i o n w i t h 100% 0 2 > h y p e r v e n t i l a t i o n w i t h 5% C 0 2 (minimum and end v a l u e s ) , and h y p e r v e n t i l a t i o n w i t h 10% C 0 2 (minimum and end v a l u e s ) . A l l v a l u e s a r e mean + S.E. S t a t i s t i c a l l y d i f f e r e n t from h y p e r v e n t i l a t i o n i n 100% 0 2 . S t a t i s t i c a l l y d i f f e r e n t from h y p e r v e n t i l a t i o n i n 5% C0_ (end). 63 effects were not as markedly affected by lowering pH. This could have also explained the greater ability of sympathetic stimulation to cause vasoconstriction when the pH is lowered compared to exogenous N A administration. Local acidosis has also depressed the response of isolated strips of saphenous veins from dogs to sympathetic nerve stimulation (Vanhoutte and Clement, 1968). Acidosis (decreasing pH from 7.4 to 7.1) depressed the response curve to nerve stimulation but did not affect the response curve to exogenous N A . This degree of acidosis also decreased the q release of [ °H]NA to the same extent as it depressed the contractile response to nerve stimulation in saphenous veins previously incubated with the labelled transmitter. Therefore these authors suggested that the impairment of vasoconstrictor activity by decreasing pH was a result of impaired neurotransmission. In the pithed rat, acidosis reduced the alpha-1 mediated pressor response to phenylephrine but favored alpha-2 mediated pressor responses to xylazine (Grant et al., 1985). The opposite effects were seen with alkalosis. In this study, combinations of hypoxia and hypercapnia dramatically reduced the response to phenylephrine, but did not affect the response to xylazine. Finally, metabolic acidosis may enhance the conversion of dopamine (DA) to N A and A (Tajimi et al., 1983). These results have shown that the hypercapnic acidosis may affect the production, release, uptake and post-synaptic effects of N A . 6 4 Stimulation of the central stump of the sectioned femoral nerve caused marked alterations in the PP of the innervated leg but little change in the PP of the innervated leg, in one intact, adrenalectomized cat (figure 11). In this animal, carotid clamping and administration of 5% CO2 caused larger changes in the PP of the innervated leg (table 7). 65 TABLE VI ANIMAL 1 ANIMAL 2 i n n l e g den l e g Inn l e g den l e g f l o w (mls/mln) 11.5 11.2 13.0 14.5 P P (mm Hg) b e f o r e 104 104 78 80 VR (PRU) b e f o r e 9.04 9.29 6.00 5.52 0 hrs 7.65 7.50 6.92 6.48 3 h r s 5.74 5.89 8.65 9.64 V a l u e s f o r f l o w r a t e s , p e r f u s i o n p r e s s u r e s of each l e g b e f o r e d e n e r v a t i o n , and v a s c u l a r r e s i s t a n c e s of each l e g b e f o r e d e n e r v a t i o n , and 0 and 3 hours a f t e r d e n e r v a t i o n , f o r two n o n - a d r e n a l e c t o m i z e d c a t s . €6 TABLE VII P e r f u s i o n P r e s s u r e Change (mm Hg) I n n e r v a t e d l e g d e n e r v a t e d l e g C a r o t Id C1 amp 1ng 17 I n s p i r1ng 5% C 0 o 26 12 V a l u e s f o r changes 1n p e r f u s i o n p r e s s u r e i n response to c a r o t i d clamping, and i n s p i r i n g 5% C0_ (a peak 1 change), i n one a d r e n a l e c t o m i z e d c a t . Figure 12 Chart recording of the perfusion pressures of both the innervated and denervated legs of one adrenalectomized non-spinal cat in response to afferent nerve stimulation. Two sets of parameters were used, one which produces a pressor effect, the other which produces a depressor effect, in the non-spinal cat. PP den., Perfusion Pressure of the denervated leg; PP inn., Perfusion Pressure of the innervated leg; SAP, Systemic Arterial Pressure. y 68 pressor depressor PP den. 90 (mm Hg) 70 c ] 1 00 PP inn. a / % (mm Hg) 80 SAP (mm Hg) 40 69 VII. DISCUSSION A. Discussion of the Methods 1. The Experimental Preparation a. Effects of Vascular Isolation The method of isolation in these experiments was very effective in separating the hindlimb and upper body vascular areas. The technique could have caused anoxic damage to the spinal cord, but this is unlikely for two reasons. Firstly, the areas most prone to ischemic damage when spinal cord blood flow is compromised are the so-called watershed areas of the thoracic spinal cord. After vascular isolation the lumbar spinal cord areas would have received adequate blood flow from collateral circulation. Secondly, carotid clamping and afferent femoral nerve stimulation resulted in prominent changes in hindlimb PP, in the intact animal after undergoing this method of vascular isolation. Therefore, it is unlikely that the spinal cord was damaged by the isolation technique. However, the neural activity could have been affected by vascular isolation and the reduced resting SAP of the spinal animal may have further compromised the lumbar spinal cord blood flow. b. Effects of Spinal Transection-Spinal Shock Spinal shock has been used to describe the lack of motor activity after spinal transection (Janig, 1985), and its intensity and duration generally increases as one ascends the evolutionary scale. Sympathetic reflex activity has usually taken much longer to recover than motor activity (Downman and McSwiney, 1946). Full recover}' of SND has taken days in the hindlimb post-ganglionic nerves of the cat (Janig, 1985), but some SND has been found in some post-ganglionic nerves after only hours (Alexander, 1945; Coote and Sato, 1978; Meckler and Weaver, 1985). In the present experiments, recovery of a stable SAP required 15-30 minutes after spinal transection. This recovery likely was not due to the concomitant recovery of basal SND, since most previous studies have demonstrated the lack of sympathetic support for 70 SAP sooner than 2 days (e.g. Ardell et al, 1982). The initial recovery of basal SAP in these experiments was likely due to a combination of vagal withdrawal increasing HR and increased humoral support for both the heart and blood vessels. The resting SAP usually dropped no lower than 60-70 mm Hg, after spinal transection. In previous experiments, the administration of alpha-chloralose in cats has resulted in an elevation of SAP lasting 40-50 minutes (Mukherjee, 1957). This author found that the loss of blood was too great if spinal transection was performed during this period and resulted in a resting SAP of 30-40 mm Hg, which increased to only 60 mm Hg even after 4 hours. If spinal transection was performed 45-60 minutes after anaesthesia, the pressure returned to 80-90 mm.Hg within 3 hours and to over 100 mm Hg after 4 hours. Since spinal transection would have resulted in extensive vasodilation, fluid administration alone during the period of shock would not adequate^ correct SAP. These results were also seen in the experiments presented in this thesis. Three animals i underwent surgery very quickly after alpha-chloralose administration and lost substantial amounts of blood. Their blood pressure remained very low despite large infusions of dextran-saline, they did not respond to administration of CO2 and their condition quickly deteriorated. These results supported the contention of Mukherjee that the amount of hypotension produced during spinal transecton would have affected the degree of spinal shock. Reflex responses depend on the state of recover of the animals after spinal shock which in turn depends on time (Downman and McSwiney, 1946; Mukherjee, 1957; Coote and Downman, 1966). Our experiments began approximately 3-4 hours after spinal transection, but spinal reactivity was probably still somewhat diminished. Despite identical protocols, the animals probably could have undergone different degrees of recovery after spinal transection, and this may have contributed to the variability in the responses to CO2. 2. The Experimental Protocol 71 a. Hyperventilation Positive pressure hyperventilation caused a marked reduction i n S A P which could have caused changes in sympathetic nerve activi ty. Sympathetic nerve activity has been shown to generally increase in response to decreases i n S A P although i t is not clear whether this would have enhanced or reduced reflex responses to CO2 i n our experiments. However , in the acute cervical monkey spinal cord blood flow did not change in response to changes in S A P between 50 and 100 m m H g (Kobrine et a l . , 1976). Therefore, changes in spinal cord blood flow probably did not occur during the hyperventi lat ion in our experiments and were therefore not responsible for the reduced P P responses observed. Hypervent i la t ion has also decreased the pressor effects of exogenous catecholamines and increased the effects of adrenalin on the contractions of the nicti tating membrane, in the acute spinal cat (Gowdey and Patel , 1964). Whether i t was the associated hypocapnia that was responsible for this change in responsiveness was not clear. Therefore, in our experiments the P P responses to hypo- and hypercapnia may have been affected by a s imi lar mechanism. Hypervent i la t ion has also increased the femoral resistance by a hypocapnia induced central effect in the acute cervical spinal cat (Gowdey and Patel , 1964). However , P P consistently fell in our study during hyperventi lat ion (as did S A P ) . The respiratory rate was over 100 per minute while the p H a values were approximately 7.8 in the experiments by Gowdey and Patel , compared to 40 per minute and 7.5 in our experiments. Those authors kept respiratory stroke volume constant using only the increase in respiratory rate to produce hypocapnia. In our experiments, hypocapnia was produced by increasing the respiratory stroke volume in addition to increasing respiratory rate. Respiratory rates greater than 40 per minute were found to severely depress resting S A P when using corresponding respiratory stroke volumes necessary to produce hypocapnia. 72 The PP is the driving pressure of the blood into the hindlimbs, and is an indication of vascular resistance assuming both the flow and the venous pressure are constant (see data analysis and interpretation). The two effects of hyperventilation were hypocapnia and a mechanical reduction of venous return. Hypocapnia may produce reflex vasodilation via the sympathetic nervous system (which would have decreased PP) and vasoconstriction due to the washout of CC"2 (which would have increased PP). Decreasing venous return produces an increase in central venous pressure (CVP). Assuming no vasoconstriction or vasodilation, an increase in C V P would have resulted in an increase in the driving pressure during constant flow perfusion i.e. an increase in PP. The decrease in PP observed in these experiments was a reflection of all these factors, the predominant effect being the reflex vasodilatory action of hypocapnia. Interestingly, the drop in SAP observed was much greater than the drop in PP probably because the decreased venous return also decreased cardiac output. Decreased cardiac output would have reduced SAP but would not have affected PP, since the hindlimbs were perfused at constant flow. In the experiments of Gowdey and Patel, a reduction of femoral blood flow was observed which indicated an increase in femoral resistance. The higher respiratory rate these authors used during hyperventilation produced a greater alkalosis and probably a greater increase in C V P than in our experiments. The decrease in femoral blood flow these authors observed again reflected a balance of factors. The greater direct vasoconstrictory action of CO2 washout and the greater CVP, could have predominated over the reflex vasodilatory influence of hypocapnia. The reduction of venous return would have also reduced cardiac output and therefore could have contributed to the reduced femoral blood flow they observed. However, this was ruled out by the authors who observed the decrease in femoral flow in the presence of constant driving pressure. The mechanical effects of hyperventilation could have caused effects on the PP after 1-1.5 minute stabilization period. The administration of CO2 may have simply masked these effects. Also, it could be argued that the proper comparison for the P2 73 response (i.e. the PP value between approximately 3 and 5 minutes of CO2 administration) would have been a PP value between approximately 3 and 5 minutes of hypocapnia after the 1-1.5 minute stabilization period. In other words, the hypocapnia also may have had effects on PP that did not become evident until after the 1-1.5 minute stabilization period. Therefore several animals were hyperventilated for 7-7.5 minutes but this resulted in no further changes in PP, SAP or P a C 0 2 . The effects of increasing CO2 versus decreasing CO2 on cardiovascular variables probably do not follow the same temporal pattern (Lioy and Trzebski, 1984). These authors found that the effects of reducing CO2 on SAP were very rapid while the effects of increasing CO2 on SAP had a slower onset i.e. the CO2 SAP responses exhibited hysteresis. Reducing the P aC02 to a particular value for 7 minutes could result in different SAP values than increasing the P &C02 to this same value for 7 minutes. Comparisons of the PI and P2 hypercapnic PP responses to the PP value produced by reducing P aC02 to hypocapnic values for 1-1.5 minutes, were therefore reasonable. This had the additional advantage of reducing the total period of hyperventilation (as opposed to a 5 minute control period) and therefore the period of reduced SAP. The hyperventilation could have shifted all PP responses in comparison to PP responses in a normoventilated animal. This limitation was accepted in order to obtain PP responses to a range of P aC02 values from hypocapnia to hypercapnia.. b. Constant Blood Flow Method The blood flows of the hindlimbs were lower in these experiments as compared to previous similar experiments (Lioy et al., 1978; Hanna et al., 1979; Rohlicek and Polosa, 1986). However, the responsiveness of the hindlimbs in this study indicated that the hindlimb vessels and neural tissue were in good condition and not hypoxic. Literature values for skeletal blood flow vary from 2-5 mis per lOOg in humans (Shepherd, 1983) and 2-26 mis per lOOg in isolated cat soleus muscle (Shepherd, 1983; Bockman et al., 1980; Whalen et al., 1973; Hilton et al., 1970). The reason for this variability is not resolved, 74 but may be caused by methods used to isolate the circulation to the muscle. Also, the use of different muscles which have different compositions of muscle types, would have different metabolic requirements and consequently different blood flow requirements. The larger muscles of the cat have lower values for resting blood flow. There are also large variations in skin blood flow depending on the temperature (Roddie, 1983). In our experiments, the temperature was maintained at 3 7 ° C and blood flow was approximate^ 4 mis per lOOg assuming that the hindlimbs represent 20% of the bodyweight. It should be remembered that the hindlimb PP was initially set to be at the level of resting SAP and hindlimb blood flow was secondary to this PP level. In our experiments, constant hindlimb blood flow was used to allow PP to be an indicator of VR, as opposed to the use of constant hindlimb PP while monitoring changes in blood flow to indicate changes in VR. The problem with the method used in these experiments was the presence of a constant flow (which is not physiological) and that changes in the PP could have had direct effects on the vasculature. Alternatively, a constant pressure method of assessing changes in VR could have been used. However, with the constant pressure method changes in tissue perfusion could directly affect the vasculature. Further experiments of this kind could include using both methods to assess changes in VR under conditions of constant flow and constant pressure perfusion. In our experiments, the PP of both legs was adjusted to be the same at the beginning of the experiment under basal conditions. However, the blood flows were sometimes different and this could have contributed to the different responses of the innervated and denervated legs to CO^- Further experiments of this nature could include experiments whereby the blood flows to each leg would be the same at the beginning of the experiment under basal conditions. Although the different PP of each leg could then affect the response (as opposed to the effects of different flows), contributions of both these effects could be investigated. The problem of the denervated legs having higher initial VR values could not have been circumvented, regardless of which method was used. 75 VR is defined as the ratio of the driving pressure (the difference between arterial and venous pressure) and flow. In this preparation the flow was constant, while the assumption was made that C V P was constant and therefore any changes in PP would reflect changes in V R due to vasoconstriction in the hindlimb arterial bed. The changes in PP could have resulted from changes in C V P i.e. the body venous bed, in response to CO2 administration. However, other similar experiments showed that C V P did not change during inhalation of similar levels of CO2 in the acute cervical spinal cat (Rohlicek and Polosa, 1986). It was possible that regional changes in the venous circulation could still have occurred in response to CO2 and these could then have altered the PP responses of the hindlimbs. c. Hyperoxia In these experiments the levels of P aC02 were kept above 150 mm Hg to avoid the effects of systemic hypoxia. Also, this would have eliminated the afferent inputs from peripheral chemoreceptors (Korner, 1974). A V P release by hypercapnia may be mediated in part by intact afferent peripheral chemoreceptor input (Rose et al., 1984). Hyperoxia, at levels comparable to those in this study has caused an increase in efficacy of angiotensin II and A V P , and has caused both increases and decreases in alpha 1 mediated vasoconstriction (Dusting and Rand, 1975; Grant et al., 1985,; Dai and Wong, 1986; Grant et al., 1986). These effects could have occurred in our experiments but would have been constant throughout each experiment and between experiments. B. Discussion of the Results These experiments were designed to investigate whether acute spinal cats could demonstrate a spinal component of CO2 mediated sympathetic neurogenic vascular tone. Increasing PaCC>2 resulted in biphasic hindlimb PP responses, and this suggests that CO2 has complex effects on the hindlimb vasculature in acute spinal cats. These results suggested that CO2 exerts vasoconstrictive effects on the hindlimbs by altering sympathetic neurogenic outflow, most likely by exciting sympathetic vasoconstrictor 76 activity, by increasing adrenal activity and, possibly, by increasing the activity of other hormone systems. CO2 probably altered sympathetic neurogenic outflow to the hindlimb vasculature. The short latency of the PI response and its reduction by denervation suggested a sympathetic neurogenic origin. In the first 1-1.5 minutes of CO2 exposure, SAP occasionally demonstrated a rapid increase primarily in DP suggesting an increase in TPR and a change in sympathetic neurogenic activity. The use of systemic hypercapnia in these experiments did not allow determination of where C 0 2 was acting, except to say that in terms of the neurogenic response the structures involved were not the classical chemosensitive areas. Conceivably CO2 (and H + ) could have acted on sympathetic afferents, spinal cord inhibitory or excitatory interneurons, pre- and/or post-ganglionic neurons, or the neuro-effector junction. Decreased uptake and increased efflux of N A at the neuro-effector junction in response to acidaemia have been demonstrated (Dusting and Rand, 1975; Verhaeghe et al., 1978). In our experiments, it is possible that the increase in PaCC"2 caused an initial efflux in NA from the nerve terminals which then caused the PI response. However, in previous experiments this phenomenon seemed to occur only when the vasoconstrictive effects of exogenous catecholamines and sympathetic stimulation were severely compromised (Dusting and Rand, 1975). Hypercapnia also caused increase in the activity of SPNs of the cervical sympathetic trunk of acute spinal cats (Zhang et al, 1982). This suggests that CO2 acts on the SPNs and/or structures antecedent to them. No evidence of hypercapnic excitation of spinal cord interneurons in the spinal animal is available, but this is an attractive possibility. Chemical stimulation of cardiac afferents has also caused increases in SPN activity, in acute spinal cats (Malliani et al, 1975). Chemosensitive areas have been identified in the sympathetic ganglia (Bulgyin, 1983), and hypercapnia has synchronized post-ganglionic nerve activity in chronic spinal 77 cats (Ardell, 1982). Also, CO2 induced increases in splanchnic nerve activity have been described in the acute spinal cat (Gootman and Cohen, 1981). However, CO2 has usually failed to alter the activity of post-ganglionic nerves in acute spinal animals (Alexander, 1945; Meckler and Weaver, 1985). Histaminergic (Lioy and White, 1973), cholinergic and non-histaminergic non-cholinergic vasodilator post-ganglionic neurons have been demonstrated in many animals including the cat (Bell, 1983; Janig, 1985). Vasoactive intestinal polypeptide (VIP) has been shown to exist in the sciatic nerve, and infusions of VIP have resulted in marked vasodilation in the cat hindlimb (Jarhult et al., 1980). Whether CO2 produced its neurogenic action by causing the reduction of activity in post-ganglionic vasodilatory fibres (by direct CO2 depression or CO2 induced inhibition) or the increase in activity in post-ganglionic vasoconstrictor fibres could not be determined from these experiments and the evidence for these possibilties in the acute spinal animal is scanty. CO2 excitation of sympathetic afferents, or spinal excitatory or inhibitory neurons, which then influence SPNs and post-ganglionic vasoconstrictor fibres is a probable mechanism for the vascular responses observed. There is evidence from Zhang and co-workers that SPN activity is increased during hypercapnia and during chemical stimulation of sympathetic afferents in the acute spinal animal. Precisely how this activity is generated, whether it is expressed in post-ganglionic nerves and destined for the blood vessels, and how this information is processed along the pathway is not known. Since C 0 2 has a generally depressant effect on nerves and since vasodilator fibres and reflexes are present in the acute spinal cat (Dorokhova et al., 1974), the withdrawal of vasodilator tone during CO2 inhalation is also possible. Finally, it should be remembered that CC<2 could excite or depress excitatory and inhibitory neurons at every point in the pathway. The post-ganglionic neurons in the hindlimb of the cat include vasoconstrictor fibres to the muscle and skin which have reacted differently to the same stimuli in the chronic 78 spinal condition (Kummel, 1983). The present experiments did not differentiate between the effects of CO2 on cutaneous versus muscle vascular resistance. Increasing P aC02 resulted in a transient fall in PP in all situations. This was likely due to the direct vasodilatory action of CO2 (Somylo and Somylo, 1970). DP, M A P , and particularly SP also demonstrated decreases in response to CO^. The drop in SAP could have been responsible for the PI response, since decreases in SAP have been shown to increase SND in the acute spinal animal (Alexander, 1945; Fernandez de Molina et al., 1965; Meckler and Weaver, 1985). However, in our experiments the PP often increased prior to the drop in SAP. Also, the cervically transected spinal cord can autoregulate its blood flow in response to changes in SAP from 55 to 100 mm Hg (discussed previously). Therefore, it was not likely that the PI neurogenic response was due to decreased spinal cord perfusion (and hypoxia) secondary to CO2 induced hypotension. Increasing P &C02 produced significant P2 responses in both innervated and denervated legs which suggested a primarily hormonal response. Adrenalectomy reduced this response (observably but not statistically) indicating that perhaps adrenal vasoconstrictor hormone release occurred during CO2 exposure. Adrenalectomy abolished the PI response of the denervated leg which suggested the adrenals may also have played a role in this early response. Adrenalectomy also reduced but did not eliminate the PI response in the innervated leg. This further demonstrated the importance of both the adrenal and the sympathetic neurogenic systems in the PP response to CO^. During CO2 administration catecholamines were likely released since HR was significantly reduced in the adrenalectomized cats but was unchanged in the non-adrenalectomized cats. The direct vasodilatory effects of CO2 likely reduced the expression of its catecholaminergic (and neurogenic) vasoconstrictive effects. However, the positive inotropic effects of catecholamine release (and increased neurogenic activity) would have remained virtual^ unopposed hence the non-significant increase in DP but significant increase in SP. 79 The decrease in HR (and possibly inotropism) may have been due to the direct effect of CO2 or H + on the cardiac pacemaker (Suutarinen, 1966; Lagneaux and Remacle, 1981). Also, in these experiments the cats were not vagotomized. Therefore, CO2 could have affected peripheral chemoreceptors which would have then activated the vagus and reduced HR and SAP. Catecholamine (and neurogenic) induced increases in SAP and HR could have also been partially compensated for by baroreceptor mediated vagal activity to the heart. The lack of a significant change in HR in the non-adrenalectomized cats would have been a reflection of the opposing effects of the vagus and catecholamines. The lack of a significant increase in SAP in the adrenalectomized cats was probably a result of direct and vagally mediated decrease in HR unopposed by catecholamine release, in response to neurogenic and possibly non-adrenal hormone vasoconstriction. The reduction of the SAP response to CO2 after adenalectomy was due to the reduced SP response, whereas DP was not as affected. This also supported the idea of catecholamine release in response to CO2 in the non-adrenalectomized cats, since catecholamines can have a particularly marked effect on SP. After inhalation of CO 2 the amount of time for SAP and PPs to stabilize was proportionally longer in the non-adrenalectomized animals. Catecholamine release would have been consistent with this observation. After inhalation of 10% CO2 the time required for SAP and PPs to stabilize was also longer than after inhalation of 5% CO2. Perhaps, a more severe hypercapnia releases more adrenal catecholamines. An observably larger P2 response to 10% CO 2 in the non-adrenalectomized animals (but not in the adrenalectomized animals) supported this contention. Removal of the adrenal glands constituted not only removal of catecholamines but also the removal of other vasoactive hormones. Neuropeptide Y (NPY) and neurotensin have been found in the cat adrenal glands (Allen, 1983), and manj' other vasoactive peptides have been found in the adrenal glands of a variety of other species. Therefore, it was possible that CO2 could have affected the release of any of these adrenal hormones 80 particularly N P Y , a potent vasoconstrictor often co-localized with N A in peripheral nerves. Adrenal hormone release has been implicated in other cardiovascular reflexes in spinal animals such as H R increases in response to aortic stretch (Lioy et a l . , 1974) and S A P increases in response to sciatic stimulation (Sherrington, 1906), bladder distension (Brooks, 1933) and hemorrhage (Brooks, 1935). Severe hypercapnic acidosis has resulted in catecholamine release in spinal cats which was attenuated by destruction of the spinal cord which suggested the involvement of spinal sympathetic "centres" (Tenney, 1956). Severe increases in C O 9 which resulted in increases in catecholamines in spinal dogs (Morris and M i l l a r , 1962; Cantu et a l . , 1966) and isolated dog adrenal glands (Nahas et a l . , 1967) suggested that CO2 could have a direct effect on the adrenal medulla. Therefore, in our experiments both direct adrenal and sympathetic activation could have caused catecholamine release. In previous experiments using intact cats, s imi lar changes in P &C02 have resulted in P P changes which were not significantly altered in adrenalectomized cats (Hanna et a l . , 1979). The present results suggest that the adrenals are of part icular importance in the cardiovascular responses and in the maintenance of C0 2 mediated vascular tone of spinal animals. Our results suggested catecholamine release by C0 2 but did not provide information as to whether N A or A is pr imar i ly released. Previous experiments have suggested that in the intact cat activation of the baroreceptors released pr imar i ly N A while activation of peripheral chemoreceptors released pr imar i ly A (Critchley et a l . , 1973; M a l y g i n a , 1961). Greater release of N A would have resulted presumably in a greater vasoconstrictive effect in the hindlimbs (Malygina , 1961). In these experiments, it can be said that adrenal hormones were released in amounts and proportions which caused overall hindlimb vasoconstriction. The presence of the P2 response in the denervated leg of adrenalectomized animals could suggest the C 0 2 mediated release of non-adrenal vasoconstrictor hormones. The significant decrease in H R and substantial reduction in S P in response to increasing 81 P aC02 in the adrenalectomized cats implied that these hormones had no cardiac effects. The P2 response of the innervated leg could have reflected the amount of time necessary for the neurogenic activity to plateau, since it can take minutes for the CSF to equilibrate with the plasma (Loeschke, 1982). However, AVP and angiotensin II are possible candidates for increasing hindlimb vascular resistance in response to increased P aC02> particularly in situations of reduced sympathetic activity such as that which occurred with spinal transection in these experiments (see introduction). There is not much evidence for augmented release of these substances by hypercapnia before 10-15 minutes, but reduced (perhaps "insignificant") amounts may have been sufficient to directly affect the hindlimb vasculature. Small amounts of AVP have also been shown to affect neurogenic activity by acting on the spinal cord (Riphagen at al., 1985). Also, afferent renal nerve stimulation for 5 minutes has resulted in significant increases in pAVP (Caverson and Ciriello, 1987). These authors also found that the secondary SAP response which occurred approximately within 1 minute of renal afferent nerve stimulation, could be blocked using an AVP Vy antagonist. These results suggest that AVP could probably be released within minutes of CO2 administration, in quantities sufficient to cause vasoconstriction. In our experiments there may have been some residual sympathetic innervation not eliminated by the lumbar sj^mpathectom}- performed. This could have explained the reduced but present PI response and the slightly reduced but present P2 response in the denervated leg. The full response of this residual activity may have required more time to become fully expressed in the denervated leg again because CSF CO2 may have required minutes to equilibrate (Loeschke, 1982). Hence, the P2 response of the denervated leg in adrenalectomized cats could have been the result of this residual sympathetic activity and not non-adrenal vasoconstrictive hormones. However, the lumbar denervation was tested in one adrenalectomized animal with an intact spinal cord and was shown to eliminate the PP responses to afferent femoral nerve stimulation, and reduce carotid clamping and CO2 administration. Possibly, the spinal animal could have mediated its response to CO2 82 through pathways different from those used to carry efferent information from the afferent stimuli tested in the intact animal. However, most vasoconstrictor and vasodilator fibers to the hindlimbs exit from T12-L5, with the most exiting from L1-L4 (Sonnenschein et al., 1978). Therefore, our method of denervation should have eliminated most of the ipsilateral fibres. Post-ganglionic somata have not been found in the contra-lateral ganglia (McLachlan and Janig, 1983), and therefore probably did not contribute to the reduced responses seen in the denervated legs. However, the left side of the cord has been shown to give rise to both ipsilateral and contralateral cardioacceleratory fibres as low as L4 (Faden et al., 1976). Therefore, C O 2 could have caused some contralateral excitaton of the "denervated" leg but this was probably not an important factor. The PI and P2 responses of the denervated leg could also be explained by spillover of NA from the nerve terminals of the intact and possibly the sectioned lumbar and thoracic post-ganglionic nerves. Since systemic hypercapnia would be expected to affect all levels of the sympathetic spinal cord, a substantial amount of NA could have been liberated into the circulation which would then have been perfused through the hindlimbs. However, the PI response of the innervated leg would have been expected to appear sooner than in the denervated leg. This appeared to be the case in several experimental runs, but did not occur consistently. The occasional early rise in SAP often appeared earlier than the rise in PP, but again this did not occur consistently. However, since the sensitivity of the hindlimbs after spinal transection may be increased, the effects of even small amounts of liberated NA on both hindlimbs should not be disregarded. The non-significant but apparent increase in the VR of the denervated leg was not expected. This basal increase in VR may have been responsible for the reduced responses of the denervated leg to increased P CO2. The vessels of the denervated leg would have initially been more constricted and would not have been able to further constrict as much as the innervated leg in response to the same C 0 2 stimulus. However, this leg was still able to produce quite large increases and decreases in PP indicating that the ability of the 83 denervated vessels to respond to CO2 was not compromised. Experiments on two animals denervated after pump attachment demonstrated a possible increase in VR after approximately 3 hours. However, in these two experiments the instability of the PP of both legs after pump attachment and after denervation made the differences obscure and uncertain. Therefore, the presence, cause and significance of the higher VR of the denervated legs has not been elucidated but several possible explanations are possible. The increase in the VR of the animals exposed to CO2 did not become apparent until the equilibration period after the pump placement. During this period of time the denervated hindlimb may have become hypersensitive to circulating humoral vasoconstrictors. The increase in VR seemed to be especiallj' marked in the denervated leg of the adrenalectomized animals. Adrenalectomy could have increased the hypersensitization of the hindlimbs to non-adrenal humoral vasoconstrictors. Another possibility was that sympathetic vasodilatory neurons (discussed previously) which were not active in the intact animal (Janig, 1985), became active in the spinal animals although there has been no evidence for this phenomenon. Perhaps spinal transection removed a tonic inhibitory influence originating above the transection, resulting in a tonic vasodilatory influence. This basal vasodilatory influence would then have been eliminated by lumbar denervation. Finally, the right leg of all cats was denervated and one side may have had a higher VR. Paw preference has been shown in dogs (Tan, 1987) and cats (Cole, 1955). In dogs, the incidence of left handers was suggested to be approximately 17.1%, similar to that found in man (Tan, 1987). Paw preference could have meant differences in sympathetic innervation to mediate differential basal and reflex changes, in blood flow. Differences in sympathetic lumbar innervation have been observed in humans, and cardioacceleratory nerves were found to have a right side preponderance (Faden et al., 1976). 84 No significant differences were found in the PP of the hindlimbs when comparisons of 5% versus 10% inspired CO2 were made despite significant differences in pH a and PaCC"2- This was likely a result of the excitatory versus the depressant effects of CO2, and increasing its concentration would have increased these opposing effects. In addition to the greater sympathetic excitation of 10% CO2, this would have caused increased hormone release, greater direct vasodilatory effects, greater inhibition of NA transmission, decreased post-junctional efficacy of NA (and A)and a direct depression of neural activity (see introduction). The PP of the innervated hindlimbs would have then represented the balance of these effects. 10% inspired CO2 would also have increased the alternative factors involved such as NA spillover and residual nerve activity resulting in a similar balance being attained in the denervated leg. Variations in these factors would also increase the variability of the PP responses and the variation in responses was greater for 10% co 2. There were differences in HR of adrenalectomized animals and SP of non-adrenalectomized animals in response to 5% and 10% C0 2- These results illustrated the opposing abilities of the factors which control heart function in the spinal non-vagotomized preparation namely the catecholamines and the vagus nerve. Inspired 10% C 0 2 produced P a C 0 9 values which were quite high and not likely to be encountered "normally". Its effects illustrated the importance of the adrenal gland in response of spinal animals to stressful amounts of C0 2. Hyperventilation with 5% CO2 produced PaCC*2 values which were approximately normal (Fink and Schoolman, 1963; Herbert and Mitchel, 1971). Therefore, the sympathetic nervous system, the adrenals and possibly other hormonal systems can be involved to different degrees in the maintenance of resting vascular tone in response to both physiological and pathological P aCU2 levels in the acute spinal animal. Previous experiments have shown that increases in CO2 have resulted in increases in sympathetic neurogenic tone which were abolished by hexamethonium in the vascularly 85 isolated cross perfused hindlimbs of the acute C l spinal cat (Rohlicek and Polosa, 1986). These results did not contradict the results of the experiments presented in this study . However, those authors suggested that vasoactive humoral substances were not likely released because the SAP of the donor cat was not affected. The amount of collateral flow in the experiments presented in this thesis was indicated by pressures of 5-7 mm Hg obtained after stopping the hindlimb pumps for approximately 30 seconds. The method of vascular isolation included elimination of collateral flow by the epigastric artery, which was not done on the studies by Rohlicek and Polosa. In their experiments, the amount of collateral flow between the upper and lower body of the experimental animal (indicated by pressures of 15 mm Hg or less after stopping the hindlimb pumps) could have been large enough to allow passage of amounts of humoral substances sufficient to cause changes in hindlimb vascular resistance. The donor animal presumably had intact cardiovascular reflexes and could have effectively damped the pressor response to vasoactive humoral substances. These substances need not have been present in large-quantities to have caused significant changes in PP. and furthermore would have been broken'down and diluted during the course of passage from the experimental to the recipient animal. Therefore whether or not vasoactive humoral substances were released in response to increasing P a C 0 2 could not be ascertained in their study. The experiments presented in this thesis did not distinguish between the individual effects of C 0 2 and H + . Both have been implicated in other studies of C 0 2 mediated increases in catecholamines, renin, angiotensin II and A V P release, and increases in sympathetic activity. C . Concluding Comments These experiments have demonstrated the effects of hypocapnia, normocapnia, and hypercapnia on the vascular resistance of the blood vessels of the hindlimb of the acute spinal cat. C 0 2 produced a certain amount of hindlimb vascular tone in this situation, which is dependent not only on sympathetic neurogenic factors but probably also on 86 hormonal factors. This was not surprising since the resting tone in the normal animal would have also been dependent on these factors. In situations where the sympathetic nervous system was compromised it would have been expected that hormonal systems become more important in the maintenance of vascular tone. Each vascular region would have had its own particular ratio of neurogenic and hormonal tone. Whether or not these experiments have demonstrated a spinal portion of the CO2 mediated neurogenic tone seen in the intact animal is not clear. Some investigators have chosen to study spinal animals in order to elucidate the spinal segments of autonomic reflexes which is an extension of the Sherringtonian approach used in the study of motor control (Lind, 1983; Janig, 1985). Changes in spinal cord plasticity and blood flow have been problems for all experiments in spinal animals, and therefore the reflexes observed were not. "normal". The use of spinal animals has been very useful in research in motor control, and has also been useful in the study of the organization of the sympathetic nervous system (Malliani et al., 1983). The concept that the spinal cord acts as an integrator of peripheral and supraspinal inputs to produce a sympathetic output and that the resulting cardiovascular control is dynamic rather than homeostatic (Malliani et al., 1983) is logical. However, vascular tone is generated not only by the sympathetic nervous system but also by hormonal systems which are of particular importance in spinal animals as demonstrated by these experiments. Animals have not evolved to deal physiologically with the cardiovascular system altered by spinal transection. Therefore, hormonal systems involved in vascular regulation react in a pathological fashion after spinal transection. This would be expected to directly affect the vessels and alter the vascular responses of effector organs such as the hindlimbs, in response to stimuli such as CO2. To better understand the contributions of the sympathetic nervous system, all systems controlling the vasculature should be monitored. Therefore, these experiments have shown that the spinal animal can generate both neurogenic and hormonal vascular responses to various levels of CO2. That these 87 responses represented the spinal portion of the response of the intact animal to low, normal and elevated levels of CO2 was not clear in light of the limitations described. These experiments have demonstrated the responses of the spinal animal to low, normal and elevated levels of CO 2- These could also be relevant to clinical problems of spinal patients, such as CO2 retention and autonomic hyperreflexia. 88 REFERENCE LIST Alexander, R.S. (1945). The effects of blood flow and anoxia on spinal cardiovascular centres. American Journal of Physiology 143, 689-708. Alexander, R.S. (1946). Tonic and reflex functions of medullary sympathetic cardiovascular centres. Journal of Neurophysiology 9, 205-217. Altura, B . M . , and Altura, B.T. (1977). Vascular smooth muscle and neurohypophyseal hormones. Federation Proceedings 36, 1853-1860. Anderson, R.J . , Rose, C . E . , jr., Berns, A.S. , Erickson, A . L . , and Arnold, P .E. (1980). Mechanism of effect of hypercapnic acidosis on renin secretion in the dog. American Journal of Physiology 238, F119-F125. Ardell, J . L . , Barman, S.M., and Gebber, G.L. (1982). Sympathetic nerve disharge in the chronic spinal cat. American Journal of Physiology 243, H463-H470. Balis, G.U. , and Monroe, R.R. (1964). The pharmacology of chloralose. Psychopharmacologia 6, 1-30. Barman, S.M. (1984). Spinal cord control of the cardiovascular system. In Nervous Control of Cardiovascular Function, ed. Randall, W.C. , pp. 321-325. Oxford University Press. Beacham, W.S., and Perl, E.R. (1964). Background and reflex discharge of sympathetic preganglionic neurones in the spinal cat. Journal of Physiology 172, 400-416. Beacham, W.S., and Perl, E.S. (1964). Characteristics of a spinal reflex. Journal of Physiology 173, 431-448. Bell, C. (1975). Vasodilator nerves in regional circulatory control. Clinical and Experimental Pharmacology and Physiology Suppl. 2, 49-53. Bell, C. (1983). Vasodilator neurons supplying skin and skeletal muscle of limbs. Journal of the Autonomic Nervous System 7, 257-262. Bernard, C. (1863). Lecons sur la Physiologie et la Pathologie du Systeme Nerveux, Volume 1. Balliere, Paris. Berns, A.S. , Anderson, R.J . , and McDonald, K . M . (1979). Effect of hypercapnic acidosis on renal water excretion in the dog. Kidney International 15, 116-125. Bockman, E . L . , McKenzie, J . E . , and Ferguson, J . L . (1980). Resting blood flow and oxygen consumption in soleus and gracilus muscles of cats. American Journal of Physiology 239, H516-524. Brooks, C M . (1933). Reflex activation of the sympathetic system in the spinal cat. American Journal of Physiology 106, 251-265. Brooks, C M . (1935). The reaction of chronic spinal animals to hemorrhage. American Journal of Physiology 114, 30-39. 89 Brown, A . M . , and Berman, P . R . (1970). Mechanisms of excitation of Aplysia neurons by-carbon dioxide. Journal of General Physiology 56, 543-558. Bulgyin, L A . (1983). A consideration of the general principles of the organization of sympathetic ganglia. Journal of the Autonomic Nervous System 8, 303-330. Cantu, R.C. , Nahas, G.G. , and Manger, W.C. (1966). Effect of hypercapnic acidosis and of hypoxia on adrenal catecholamine output of the spinal dog. Proceedings in Social, Experimental and Biological Medicine 122, 434-437. Caverson, M . M . , and Ciriello, J . (1987). Effect of stimulation of afferent renal nerves on plasma levels of vasopressin. American Journal of Physiology 252, R801-R807. Chalazanotis, N . (1963). Effects of changes in P C Q 2 a n < ^ ^02 o n t f i e rhythmic potentials from giant neurons. Annals of the New York Academy of Sciences 109, 451-479. Ciriello, J . , Rohlicek, C.V. , and Polosa, C. (1985). 2-Deoxyglucose uptake in the central nervous system during systemic hypercapnia in the peripherally denervated cat. Experimental Neurology 88, 673-687. Coote, J . H . , and Downman, C.B.B. (1966). Central pathways of some autonomic reflex discharges. Journal of Physiology 183, 714-729. Coote, J . H . , and Sato, A. (1978). Supraspinal regulation of spinal reflex discharge onto cardiac sympathetic nerves. Brain Research 142, 425-437. Coren, M . E . , Cross, B.A. , Forsling, M . L . , and Jenkins, M.P. (1987). The time course of the vasopressin response to hypercapnic acidosis in mechanically ventilated dogs. Journal of Physiology 382, 59P. Dai, S., Wong, H . , and Ogle, C.W. (1986). Effects of hypoxaemia and hyperoxaemia on some cardiovascular responses of rats to adrenaline. Archives Internationales de Physiologic et Biochimie 94, 323-329. Dorokhova, M.I., Medvedev, O.S., Reznikova, Y . A . , and Tsyrlin, V . A . (1974). Evidence for inhibitory vasomotor control at the spinal level. Bulletin of Experimental and Biological Medicine 77, 95-97. Downman, C.B.B., and McSwiney, B.A. (1946). Reflexes elicited by visceral stimulation in the acute spinal animal. Journal of Physiology 105, 80-94. Dusting, G.J . , and Rand, M . J . (1975). Interactions between the hydrogen ion concentration and the vasoconstrictor responses to catecholamines and sympathetic nerve stimulation. Clinical and Experimental Pharmacology and Physiology Suppl. 2, 43-48 Faden, A . L , Jacobs, T, and Woods, M . (1978). Cardioacceleratory sites in the cat spinal cord. Experimental Neurology 61, 301-310. Fernandez De Molina, A. , and Perl, E.R. (1965). Sympathetic activity and the systemic circulation in the spinal cat. Journal of Physiology 181, 82-102. Fink, B.R., and Schoolman, A. (1963). Arterial acid base balance in unrestrained waking cats. Proceedings of Social, Experimental and Biological Medicine 112, 328-330. 90 Ford, R., Hainsworth, R., Rankin, A . J . , and Soladoye, A.O. (1985). Abdominal vascular responses to changes in carbon dioxide tension in the cephalic circulation of anaesthetized dogs. Journal of Physiology 358, 417-431. Frankel, H . L . , Lightman, S.L., Poole, C . J . M . , and Williams, T . D . M . (1986). Increased sensitivity to the pressor response of A V P in subjects with mid cervical cord transection. Journal of Physiology 365, 39P. Fukuda, Y . , See, W.R., and Honda, Y. (1980). H +-sensitivity and pattern of discharge of neurons in the chemosensitive areas of the ventral lateral medulla of rats in vitro. Pflugers Archives 388, 53-61. Gavras, I., Hatinoglou, S., and Gavras, H . (1986). The adrenergic system and the release and pressor action of vasopressin. Hypertension 8 (suppl.II), 163-167. Gilbert, S.G. (1976). Pictorial anatomy of the cat, University of Toronto Press. Gowdey, C.W., and Patel, Y . J . (1964). Responses to adrenaline and noradrenalin in cats during hypoxia, hyperventilation and hyperoxia. Archives of International Pharmacodynamics 130, 67-84. Grant, T . L . , McGrath, J . C , and O'Brien, J.W. (1985). The influence of blood gases on alpha ^ and alpha 2 adrenoceptor-mediated pressor responses in the pithed rat. British Journal of Pharmacology 86, 69-77. Grant, T . L . , and McGrath, J .C. (1986). The effects of arterial blood gas tensions on pressor responses to angiotensin II in the pithed rat. Journal of Physiology 372, 437-444. Gregor, M . , and Janig, W. (1977). Effects of systemic hypoxia and hypercapnia on cutaneous and muscle vasoconstrictor neurones to the cat's hindlimb. Pflugers Archives 368, 71-81. Hainsworth, R., Macgregor, K . H . , Rankin, A . J . , and Solado5re, A . O . (1984). Cardiac inotropic responses from changes in carbon dioxide tension in the cephalic circulation of anaesthetized dogs. Journal of Physiology 357, 23-25. Hammer, M . , and Skagen, K. (1986). Effects of small changes of plasma vasopressin on subcutaneous and skeletal muscle blood flow in man. Acta Physiologica Scandanavica 127, 67-73. Hanley, D.F . , Wilson, D.A. , and Traystman, R.J . (1986). Effect of hypoxia and hypercapnia on neurohypophyseal blood flow. American Journal of Physiology 250, H7-H15. Hanna, B.D., Lioy, F . , and Polosa, C. (1979). The effect of cold blockade of the ventral medullary chemoreceptors on the C 0 2 modulation of vascular tone and heart rate. Canadian Journal of Physiology and Pharmacology 57, 461-468. Hanna, B.D., Lioy, F . , and Polosa, C. (1981). Role of carotid and central chemoreceptors in the C 0 2 response of sympathetic preganglionic neurons. Journal of the Autonomic Nervous System 3, 421-435. Herbert, D.A. , and Mitchel, R.A. (1971). Blood gas tensions and acid base balance in awake cats. Journal of Applied Physiology 30, 434-436. 91 Hilton, S.M., Jeffries, M.G. , and Vrbova, G. (1970). Functional specializations of the vascular bed of the soleus. Journal of Physiology 206, 543-562. Horeyseck, G . , and Janig, W. (1974). Reflex activity in post-ganglionic fibres within the skin and muscle in chronic spinal cats. Experimental Brain Research 21, 155-168. Iwawoto, G.A. , Waldrop, T .G. , Kaufman, M.P. , Botterman, B.R., Rybicki, K . J . , and Mitchell, J . H . (1985). Pressor reflex evoked by muscular contraction: contribution by neuraxis levels. Journal of Applied Physiology 59, 459-467. Janig, W. (1985). Organization of the lumbar sympathetic outflow to skeletal muscle and skin of the cat hindlimb and tail. In Reviews in Physiology, Biochemistry, and Pharmacology ed. Vol. 102, pp. 121-202. Janig, W., Krauspe, R., and Wiedersatz, G. (1983). Reflex activation of post-ganglionic vasoconstrictor neurones supplying skeletal muscle by stimulaton of arterial chemoreceptors via non-nicotinic synaptic mechanisms in the sympathetic ganglia. Pflugers Archives 396, 95-100. Jarhult, J . , Hellsrand, P., and Sundler, F . (1980). Immunohistochemical localization and vascular effects of vasoactive intestinal polypeptide in the skeletal muscle of the cat. Cell and Tissue Research 207, 55-64. Johnson, R.H. , Crampton Smith, A. , and Walker, J . M . (1965). Heart rate and blood pressure of spinal cats inspiring C 0 2 before and after injection of hexamethonium. Clinical Science 36, 257-265. Kaneko, Y . , Page, I.H., and McCubbin, J .W. (1964). Hemodynamic studies in normotensive and renal hypertensive chronic spinal dogs. American Journal of Physiology 206, 562-566. Kobrine, A . L , Doyle, T . F . , Newby, N . , and Rizzoli, H .V. (1976). Preserved autoregulation in the rhesus spinal cord after high cervical section. Journal of Neurosurgery 44, 425-428. Korner, P.I. (1971). Integrative neural cardiovascular control. Physiological Reviews 51, 312-355. Krnjevic, K. , Randic, M . , and Siesjo, B.K. (1965). Cortical C 0 2 tension and neuronal excitability. Journal of Physiology 176, 105-122. Kummel, H . (1983). Activity in sympathetic neurons supplying skin and skeletal muscle of spinal cats. Journal of the Autonomic Nervous System 7, 319-329. Kurz, K.D. , and Zehr, J . E . (1978). Mechanisms of enhanced renin secretion during C 0 2 retention in dogs. American Journal of Physiology 234, H573-H581. Liard, J . F . , Deriaz, O., Schelling, P., and Thibonnier, M . (1982). Cardiac output distribution during vasopressin infusion or dehydration in concious dogs. American Journal of Physiology 243, H663-H669. Lind, A.R. (1983). Cardiovascular adjustments to isometric contractions; static effort. In Handbook of Physiology, Volume III Peripheral Circulation and Organ Blood Flow, part 2, eds. Shepherd, J .T . , Abboud, F . M . , and Geiger, S.R., pp. 947-966. American Physiological Society. 92 Lioy, F . (1986). Neurally mediated hemodynamic effects of CO^. Journal of the Autonomic Nervous System Suppl., 151-154. Lioy, F . and Polosa, C. (1971). Passive vasodilation of the hindlimb after sympathetic preganglionic stimulation. Journal of Physiology 213, 55-66. Lioy, F. and White, K.P. (1973). -^C-Histamine release during vasodilation induced by lumbar ventral root stimulation. Pflugers Archives 342, 319-324. Lioy, F . , Hanna, B.D., and Polosa, C. (1978). CO2 dependent component of neurogenic vascular tone in the cat. Pflugers Archives 374, 187-191. Lioy, F . , Hanna, B.D., and Polosa, C. (1981). Cardiovascular control by medullary surface receptors. Journal of the Autonomic Nervous System 3, 1-7. Lioy, F . , Malliani, A. , Pagani, M . , Recordati, G. , and Schwartz, P. (1974). Reflex hemodynamic responses initiated from the thoracic aorta. Circulation Research 34, 74-84. Lioy, F . , and Trzebski, A. (1984). Pressor effect of CO2 in the rat: different thresholds of the central cardiovascular and respiratory responses to CO2. Journal of the Autonomic nervous System 10, 43-54. Malliani, A. , Lombardi, F . , Pagani, M . , Recordati, G. , and Schwartz, P.J. (1975). Spinal cardiovascular reflexes. Brain Research 87, 239-246. Mannard, A. , and Polosa, C. (1973). Analysis of background firing of single sympathetic preganglionic neurons of cat cervical nerve. Journal of Neurophysiology 36, 398-408. Matalon, S., Nesarajah, M.S., Krasney, J .A. , and Farhi, L . E . (1983). Journal of Applied Physiology: Respiratory, Environmental, and Exercise Physiology 54, 803-808. Mathias, C.J . and Frankel, H . L . (1983). Clinical manifestations of malfunctioning sympathetic mechanisms in tetraplegia. Journal of the Autonomic Nervous System 7, 303-312. Meckler, R.L. and Weaver, L . C . (1985). Splenic, renal, and cardiac nerves have unequal dependence upon tonic supraspinal inputs. Brain Research 338, 123-135. Millar, R.A. (1960). Plasma adrenaline and noradrenaline during diffusion respiration. Journal of Physiology 150, 79-90. Miller, J . M . , Downey, J .A . , and Darling, R.C. (1969). The cardiovascular effects of breathing 5 per cent carbon dioxide. Archives in Physical Medicine 50, 442-447. Millhorn, D .E . (1986). Neural respiratory and circulatory interaction during chemoreceptor stimulation and cooling of the ventral medulla in cats. Journal of Physiology 370, 217-231. Mitchell, J . H . (1985). Cardiovascular control during exercise; central and reflex mechanisms. American Journal of Medicine 55, 34D-41D. Morris, M . E . , and Millar, R.A. (1962). Blood pH/plasma catecholamine relationships; respiratory acidosis. British Journal of Anaesthesiology 34, 672-681. Mukherjee, S.R. (1957). Effect of bladder distension on arterial blood pressure and renal circulation in acute spinal cats. Journal of Physiology 138, 300-306. 93 Nahas, G.G. , Zagury, D., Milhaud, A. , Manger, W . M . , and Pappas, G.D. (1967). Acidemia and catecholamine output of the isolated canine adrenal gland. American Journal of Physiology 213, 1186-1192. O'Brodovich, H . M . (1984). Plasma catecholamines during hypoxemia and hypercapnia. American Journal of Physiology 250 letter, H341-H342. O'Brodovich, H . M . , Stalcup, S.A., Pang, L . M . , and Mellins, R.B. (1982). Hemodynamic and vasoactive mediator response to experimental respiratory failure. Journal of Applied Physiology: Respiratory, Environmental and Exercise Physiology 52, 1230-1236. Page, R.B., Funsch, D.J. , Brennan, R.W., and Hernandez, M . J . (1981). Regional neurohypophyseal blood flow and its control in adult sheep. American Journal of Physiology 241, R36-R43. Paller, M.S. , and Linas, S.L. (1984). Role of angiotensin II, alpha adrenergic system and arginine vasopressin on arterial pressure in the rat. American Journal of Physiology 246, H25-H30. Papajewski, W., Klee, M.R., and Wagner, A. (1969). The action of raised CO2 pressure on the excitability of spinal motoneurons. Electroencephalography and Clinical Neurophysiology 27, 618. Polosa, C., Lioy, F . and Hanna, B.D. (1983). The role of the ventral medulla in the control of sympathetic activity by systemic arterial CO2. In Central Neurone Environment, ed. Shlaefke, M . E . , Koepchen, H.P., and See, W.R., pp. . Springer Verlag. Preiss, G. , and Polosa, C. (1977). The relation between end tidal CO2 and discharge patterns of sympathetic preganglionic neurons. Brain Research 122, 255-267. Raff, H . , Shinsako, J . , Keil, L . C . , and Dallman, M . F . (1983). Vasopressin, A C T H , and corticosteroids during hypercapnia and graded hypoxia in dogs. American Journal of Physiology 244, E453-E458. Raff, H . , Shinsako, J . , and Dallman, M . F . (1984). Renin and A C T H responses to hypercapnia and hypoxia after chronic carotid chemodenervation. American Journal of Physiology 247, R412-R417. Riphagen, C .L . , and Pittman, Q.J. (1985). Cardiovascular responses to intrathecal administration of arginine vasopressin in rats. Regulatory Peptides 10, 293-298. Roddie, I.C. (1983). Circulation to skin and adipose tissue. In Handbook of Physiology, Volume III Peripheral Organ Blood Flow, part 1, eds. Shepherd, J .T . , Abboud, F . M . , and Geiger, S.R., pp. 285-318. American Physiological Society. Rohlicek, C V . and Polosa, C. (1986). Neural effects of systemic hypoxia and hypercapnia on hindlimb vascular resistance in acute spinal cats. Pflugers Archives 406, 392-396. Rose JR., C E . , Althaus, J .A . , Kaiser, D .L . , Miller, E .D. , and Carey, R .M. (1983). Acute hypoxemia and hypercapnia: increase in plasma catecholamines in concious dogs. American Journal of Physiology 245, H924-H929. Rose JR., C . E . , Anderson, R.J. , and Carey, R.M. (1984). Antidiuresis and vasopressin release with hypoxemia and hypercapnia in concious dogs. American Journal of Physiology 247, R127-R134. 94 Rose JR., C .E . , Godine JR., R.L. , Rose, K . Y . , Anderson, R.J. , and Carey, R.M. (1984). Role of arginine vasopressin and angiotensin II in cardiovascular responses to combined acute hypoxemia and hypercapnic acidosis in concious dogs. Journal of Clinical Investigation 74, 321-331. Rose JR., C . E . , Kimmel, D.P., Godine, R .L . , Kaiser, D .L . , and Carey, R . M . (1983). Synergistic effects of acute hypoxemia and hypercapnic acidosis in concious dogs. Circulation Research 53, 202-213. Shechzer, P .H. , Egbert, L .D. , Linde, H.W., Cooper, D.Y. , Dripps, R.D., and Price, H . L . (1960). Effect of C 0 2 inhalation on arterial pressure, E C G and plasma catecholamines and 17-OH corticosteroids in normal man. Journal of Applied Physiology 15, 454-458. Shepherd, J .T . (1983). Circulation to skeletal muscle. In Handbook of Physiology, Volume III Peripheral Organ Blood Flow, part 1, eds. Shepherd, J .T. , Abboud, F . M . , and Geiger, S.R., pp. 319-370. American Physiological Society. Shirahata, M . , Nishino, T., Honda, Y. , Itoh, K. , and Yonezawa, T. (1985). Effects of hypercapnia on renal nerve activity. Japanese Journal of Physiology 35, 391-399. Siegel, S. (1956). Non-parametric statistics for the behavioural sciences, McGraw Hill Book Company. Soladoye, A .O. , Rankin, A . J . , and Hainsworth, R. (1985). Influence of C 0 2 tension in the cephalic circulation on hindlimb vascular resistance in anaesthetized dogs. Quarterly Journal of Experimental Physiology 70, 527-538. Staszewska-Barczak, J . (1978). Participation of the sympathetic and the renin-angiotensin systems in blood pressure control during hypercapnia in the anaesthetized dog. European Journal of Pharmacology 49, 441-444. Stjernberg, L . , and Wallin, B .G. (1983). Sympathetic neural outflow in spinal man. Journal of the Autonomic Nervous System 7, 313-318. Sulzyck, P., and Trzebski, A. (1976). The local effect of pH changes on the cerebrospinal fluid on the ventrolateral areas of the medulla oblongata and spinal cord surfaces on activity of cardiac and vertebral sympathetic nerves. Acta Physiologica Polonica 27, 9-17. Suutarinen, T. (1966). Cardiovascular response to changes in arterial carbon dioxide tension. Acta Physiologica Scandinavica 67, suppl. 266, 1-75. Tabrizchi, R, King, K . A . , and Pang, C.Y. Vascular role of vasopressin in the presence and absence of influence from angiotensin II or alpha-adrenergic system. Canadian Journal of Physiology and Pharmacology 64, 1143-1148. Tajimi, K. , Kosugi, I., Hamamoto, F. , and Kobayashi, K. (1983). Plasma catecholamine levels and hemodynamic responses of severely acidotic dogs to dopamine infusion. Critical Care Medicine 11, 817. Tenney, S.M. (1956). Sympatho-adrenal stimulation by carbon dioxide and the inhibitory effect of carbonic acid on epinephrine response. American Journal of Physiology 187, 341-346. Tenney, S.M. (1960). The effect of carbon dioxide on neurohormonal and endocrine mechanisms. Anaesthesiology, 21, 674-685. 95 Walker, B.R. (1987). Cardiovascular effect of V j vasopressinergic blockade during acute hypercapnia in concious rats. American Journal of Physiology 252, R127-R133. Wang, B.C. , Sundet, W.D., and Goetz, K . L . (1984). Vasopressin in plasma and cerebrospinal fluid of dogs during hypoxia or acidosis. American Journal of Physiology 247, E449-E455. Weissman, M . L . , Rubinstein, E . H . , and Sonnenschein, R.R. (1976). Vascular responses to short-term systemic hypoxia, hypercapnia and asphyxia in the cat. American Journal of Physiology 230, 595-601. Whalen, W.J . , Buerk, J .D. , and Thuning, C.A. (1973). Blood flow-limited oxygen consumption in resting cat muscle. Journal of Physiology 224, 763-768. Wurster, R.D., and Randall, W.C. (1979). Cardiovascular responses to bladder distension in patients with spinal transection. American Journal of Physiology 228, 1288-1292. Zhang, T. -X. , Rohlicek, C.V. , and Polosa, C. (1982). Responses of sympathetic preganglionic neurons to systemic hypercapnia in the acute spinal cat. Journal of the Autonomic Nervous System 6, 381-389. 

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
https://iiif.library.ubc.ca/presentation/dsp.831.1-0096857/manifest

Comment

Related Items