BREATHING AND THE CONTROL OF HEART RATE

by

ERICA POTTER

School of Physiology and Pharmacology University of New South Wales

. : '.,;

' Ph.D . January 1981 ACKNOWLEDGEMENTS

I am very grateful to Professor W.E Glover for enabling me to undertake this project in the School of Physiology and Pharmacology, to the National Heart Foundation of Australia for its support of this project, to Miss Diane Madden for her untiring technical help, to Dr. Simon Gandevia for a stimulating period of collaboration, to Mrs. Judy Bokor who typed this thesis and especially to my supervisor, Associate Professor Ian Mccloskey for his encouragement, pa~ience and guidance throughout. STATEMENT REGARDING WORK DONE IN COLLABORATION Some of the results reported in chapter 3 have been published in a paper (Davis, Mccloskey and Potter, 1977) whicl included also some results from work done by Dr. Anne Davis. Dr. Davis and I did not perform our experiments together, and none of her results are included in this thesis.

Experiments involving recordings of heart rate in animals and man are described in chapters 4, 5 and 6; some of these were performed in equal collaboration with Dr. Simon Gandevia and were published with him (Gandevia, McCloskey and Potter, 1978 a, b). The collaborative work done with Dr. Gandevia was not included in his Ph.D thesis. The experiments on nerve recordings in the same chapters and elsewhere throughout this thesis were performed by me.

Associate Professor Ian Mccloskey supervised my work and collaborated in many experiments.

PUBLICATIONS ARISING FROM WORK DESCRIBED IN THIS THESIS

1. Davis, A.L., Mccloskey, D.I, and Potter, E.K (1977). Respiratory modulation of baroreceptor and chemo­ receptor reflexes affecting heart rate through the sympathetic nervous system. J .Physiol. 272 : 691 - 70: 2. Gandevia, S.C., McCloskey, D.I, and Potter, E.K (1978). Inhibition of baroreceptor and chemoreceptor reflexes on heart rate by afferents from the lungs. J. Physiol. 276 : 369 - 381. 3. Gandevia, S.C., Mccloskey, D.I, and Potter, E.K (1978). Reflex bradycardia occurring in response to diving, nasopharyngeal stimulation and ocular pressure, and its modification by respiration and swallowing. J. Physiol. 276 : 383 - 394. 4. McCloskey, D.I, and Potter, E.K (1981). Excitation and inhibition of cardiac vagal moto­ neurones by electrical stimulation of the carotid sinus nerve. J.Physiol. in press. 5. Potter, E.K (1981). Inspiratory inhibition of vagal responses to baro­ receptor and chemoreceptor stimuli in the dog. J. Physiol. in press. 1

TABLE OF CONTENTS

Page

TABLE OF CONTENTS 1

ABSTRACT 5

CHAPTER 1 INTRODUCTION (5 figs) 8

1. GENERAL INTRODUCTION 9

2. ACTIONS OF THE VAGUS 9

3. ANATOMY 11

(a) Cardiac vagal preganglionic 11 neurones (b) Properties of cardiac vagal 13 fibres

(c) Afferent pathways 15

4. REFLEXES THAT EXCITE THE VAGUS 18 (a) Baroreceptor reflex 18 (b) Chemoreceptor reflex 23 (c) Diving response 25 (d) Oculocardiac and other 27 reflexes (e) Ventricular receptors 28

(f) Juxtapulmonary capillary 29 receptors (g) Vasovagal syncope 31 5. SINUS ARRHYTHMIA 32

6. RESPIRATORY MODIFICATIONS OF 35 REFLEXES

(a) Carotid sinus nerve 35 stimulation

(b) Functional stimulation 39 2

Page

7. EXPERIMENTAL CONSIDERATIONS 42 (a) Identification of cervical 42 cardiac vagal efferent fibres (b) Effects of anaesthesia 43 CHAPTER 2 METHODS (2 figs) 47 1. WHOLE ANIMAL EXPERIMENTS WITH 48 OBSERVATIONS ON HEART RATE 2. NERVE RECORDING EXPERIMENTS 52 3. HUMAN EXPERIMENTS 59

CHAPTER 3 RESPIRATORY MODULATION OF BARO­ 61 RECEPTOR AND CHEMORECEPTOR REFLEXES MEDIATED BY THE VAGUS AND THE SYMPATHETIC NERVOUS SYSTEM (6 figs) 1. VAGAL RESPONSES 62 2. SYMPATHETIC EFFECTS 64 (a) Baroreceptor responses 66 (b) Chemoreceptor responses 69 3. DISCUSSION 71

CHAPTER 4 CENTRAL AND PERIPHERAL FACTORS 76 MODIFYING BARORECEPTOR AND CHEMORECEPTOR REFLEXES (11 figs) 1. CENTRAL MODULATION OF REFLEX 77 RESPONSIVENESS 2. PERIPHERAL MODULATION OF 81 REFLEX RESPONSIVENESS (a) Effects of stimuli timed with 81 respect to air flow into the lungs · fast ramps (b) Effects of stimuli timed with 85 respect to air flow into the lungs slow ramps (c) Effects of denervation of 88 the lungs 3. DISCUSSION 98 3

Page

CHAPTER 5 RESPIRATORY MODULATION OF OCULOCARDIAC 104 AND NASOPHARYNGEAL REFLEXES (5 figs)

1. OCULOCARDIAC REFLEX 105 2. NASOPHARYNGEAL STIMULATION 111

3. DISCUSSION 113 (a) Oculocardiac reflex 113

{b) Diving reflex 115 CHAPTER 6 RESPIRATORY MODULATION OF CARDIO­ 116 DEPRESSOR REFLEXES IN NORMAL HUMAN SUBJECTS . (4 figs) 1. OCULOCARDIAC REFLEX 118 2. DIVING RESPONSE 121

3. BREATH HOLDING IN NORMOXIC 121 AND HYPOXIC CONDITIONS

4. DISCUSSION 124

CHAPTER 7 ANALYSIS OF MECHANISMS OF 127 INSPIRATORY INHIBITION OF VAGAL ACTIVITY (8 figs) 1. RESPIRATORY EFFECTS OF VAGAL 129 TONE 2. PROPERTIES OF THE VAGAL 133 INHIBITORY PATHWAY (a) Interactions with central 133 inspiratory activity (b) Interac~ions with effects 137 from lung inflation

3. DISCUSSION 140 CHAPTER 8 COMPARISON OF VAGAL RESPONSES TO 149 ELECTRICAL STIMULATION OF THE CAROTID SINUS NERVE AND TO BRIEF BARORECEPTOR STIMULI (13 figs) 1. RESPONSES EVOKED BY SINGI.;E 150 ELECTRICAL STIMULI TO THE CAROTID SINUS NERVE (a) Variability of response 150 latencies 4

Page (b) Respiratory modulation of 151 variability of response latencies (c) Inhibition of vagal discharge 156 following responses elicited by single electrical stimuli 2. RESPONSES EVOKED BY MULTIPLE 165 STIMULI TO THE CAROTID SINUS NERVE (a) Vagal responses to pairs of 165 electrical stimuli (b) Vagal responses to trains of 165 electrical stimuli

3. RESPONSES TO BRIEF BARORECEPTOR 167 STIMULI (a) Single stimuli 167 (b) Pairs of baroreceptor stimuli 169

4. DISCUSSION 173 CHAPTER 9 GENERAL DISCUSSION (3 figs) 180 1. PHENOMENA STUDIED 181

2. METHODS 182 3. MECHANISMS 187 4. SIGNIFICANCE 192 5. APPLICATIONS 195 6. FURTHER EXPERIMENTS 196

REFERENCES 198 5

ABSTRACT

Heart rate alters with breathing. The neural basis of the relationship between the two was studied in the experiments described in this thesis.

There are several reflexes that slow the heart. Four were studied in detail here: the baroreceptor reflex, the chemoreceptor reflex, the diving response and the oculo­ cardiac reflex. The heart can be slowed by activation of cardiac vagal nerves or by withdrawal of cardiac sympathetic activity the sympathetic contribution to the slowing of the heart is relatively smaller and slower. Heart rate was studied in anaesthetised dogs and conscious human subjects. Dogs were chosen as experimental animals because even when anaesthetised they maintain a high level of cardiac vagal tone. More quantitative aspects of the interaction of breathing with the neural control of heart rate were studied in anaesthetised dogs by recording activity in single vagal efferent nerve fibres running to the heart.

It has been shown here that each of the reflexes studied can slow the heart only, or most effectively, in the expiratory phase of breathing. The bradycardia evoked by selective stimulation of these reflexes during expiration is due both to vagal excitation and sympathetic withdrawal. Selective stimulation of any one of these reflexes is inhibited in inspiration. This inspiratory inhibition has 6 two components a peripheral one caused by activation of intrapulmonary receptors stimulated during lung inflation, and a central one related to the cyclical activity of inspiratory neurones. Lung inflation alone, in the absence of central inspiratory activity can inhibit these reflexes, or central inspiratory activity alone in the absence of lung movements can also inhibit these reflexes.

Nerve fibre recordings in anaesthetised dogs showed differences in the mechanisms of inspiratory inhibition caused by lung inflation and central inspiratory activity.

Selective stimulation of arterial baroreceptors or chemo­ receptors was used to evoke reflex increases in vagal activity. Lung inflation inhibits reflexly-evoked increases in vagal discharge but leaves any resting vagal discharge relatively unaffected. Central inspiratory activity, on the other hand, inhibits both reflexly-evoked increases in vagal firing together with resting vagal discharge. These results are accommodated in two models of the vagal-excitatory pathway proposed in Chapter 7.

Briefly, these models have lung inflation inhibiting the baroreceptor-afferent to vaga~-efferent pathway relatively early in that pathway, while central inspiratory activity has its inhibitory action relatively late in the reflex pathway.

Electrical stimulation of the carotid sinus nerve was also used to study the central connectiomof baroreceptor and chemoreceptor reflexes. Although such stimulation is a 7 useful physiological tool in the investigation· of reflex pathways its excitatory action on the vagus differs in several important respects, documented here, from excitation caused by selective functional stimulation by baroreceptors and chemoreceptors.

In conscious human subjects three cardiodepressor reflexes were tested: the oculocardiac reflex, the diving response and bradycardia evoked by hypoxia. Each of these reflexes is inhibited by inspiration. This confirms the results obtained in anaesthetised dogs and justifies the use of anaesthetised dogs as suitable animal models in which to study the interaction of breathing with heart rate. Central inspiratory activity alone was tested here in human subjects by having them swallow, or take an inspiratory effort-against a closed glotis (with, consequently, little accompanying lung expansion). Both these manoeuvres inhibited the reflexes tested.

The inspiratory inhibition of cardiodepressor reflexes, therefore, appears to be general phenomenon as it is seen with all the cardiodepressor ~eflexes tested and in conscious human subjects as well as anaesthetised dogs. This inhibitory action of inspiration on heart rate may be best understood in terms of oxygen conservation and delivery. The results suggest applications in some simple clinical manoeuvres. Both these aspects of the results are discussed. 8

CHAPTER 1

INTRODUCTION 9

1. GENERAL INTRODUCTION

In 1845 two brothers named Weber showed that stimulation of the vagus slowed the heart. This was the first demonstration that inhibitory effects could be evoked through well defined neural pathways. It ultimately opened the way to Loewi's celebrated demonstration in 1921 of chemical transmission at vagal nerve endings, and established a fundamental property of the nervous system's influence on the circulation (Loewi, 1921; Dale, 1963).

The work described in this thesis deals with the natural behaviour of vagal preganglionic fibres travelling to the heart. In many instances this has been studied by observing heart rate, but more quantitative aspects of efferent vagal behaviour have been analysed by recording action potentials in single and few fibre vagal efferent preparations. In this first chapter the experimental background to this work is reviewed. My involvement in this work grew from research carried out in a B.Sc(honours) program ('Masters qualifying'), on the effects of angiotensin II on vagal pathways and on breathing (Lumbers, Mccloskey & Potter, 1979; Potter & McCloskey, 1979).

2. ACTIONS OF THE VAGUS

The vagus slows the heart. In their study of the action of the vagus nerve, Brown & Eccles (1934) showed 10 that a single electrical stimulus applied to the peripheral end of the vagus can slow the heart immediately, and that the effect is maintained for as much as several seconds before the heart returns to its resting rate of beating. Also, the precise timing of a delivery of vagal stimuli within the cardiac cycle determines which cardiac cycle is first affected, and by how much : detailed studies of these effects have been reported (e.g., Levy, Iano & Zieske, 1972;

Reid, 1969; Dong & Reitz, 1970; Levy & Martin, 1979).

In many animals, including man, the vagus has a powerful restraining influence on resting heart rate. However, resting heart rate is not simply the algebraic sum of the effects of the effects of vagal and sympathetic activity.

Samaan (1935) showed that vagal influences predominate when both the sympathetic nerves to the heart and the vagus are stimulated simultaneously. He also demonstrated the asymmetry of the vagal response slowing of the heart appears "at the first beat" following onset of vagal stimulation but the rate does not return to resting levels for 1 - 3 seconds after stimulation is stopped. In contrast, he noted that on stimulation of the sympathetic nerve to the heart, heart rate does not increase immediately but takes 7 - 12 seconds to reach its maximum level. This work has been extended by others (Warner & Cox, 1962; Warner &

Russell, 1969; Katona, Poitras, Barnett & Terry, 1970) in studies where these properties of the vagus and sympathetic nerves were used to predict heart rate in various situations. 11

Stimulation of the vagus also leads to decreased atrial contractility and decreased strength of atrial contraction.

It slows conduction velocity through the atrioventricular node. A question which has been much debated has been whether or not vagal endings innervate ventricular myocardium enabling vagally-mediated reductions of ventricular contractility. It now appears that such innervation does exist, but exerts only weak negative inotropic effects demonstrable in carefully controlled experimental situations (e.g., see Levy & Martin, 1979; see also Milnor, 1980).

3. ANATOMY

As a background to the study of natural activity of the vagus, it is necessary to consider the extent of afferent influences on the cardiac vagal motoneuronal pool and the neural pathways involved. Some of the important cardiovascular nuclei and their relative positions in the medulla are shown diagramatically in fig 1.1.

(a) Cardiac vagal preganglionic neurones. Early anatomical evidence, although sometimes conflicting, indicated that the dorsal motor nucleus of the vagus contained the cell bodies of cardiac vagal preganglionic neurones (see Mitchell & Warwick, 1955, for review). In the monkey (Mitchell & Warwick, 1955), dog (Gunn, Sevelius,

Puiggari & Myers, 1968) and rabbit (Jordan, Khalid,

Schneiderman & Spyer, 1979) the dorsal motor nucleus does 12

8

Fig 1.1. Diagrammatic representation of the relative positions of three major cardiac nuclei in the medulla, the nucleus of the tractus solitarius (NTS), the dorsal motor nucleus of the vagus (DMN) and the nucleus ambiguus (NA). The root­ lets of the vagus (X) and glossopharyngeal (IX) nerves are also shown as they emerge from the medulla. Panel A shows the extent of these nuclei within the medulla from their most rostral to caudal levels. Panel B shows a section through the medulla at the level of the obex. The NTS and the DMN both lie dorsally, the NA lies ventrally. 13 contain cardiac vagal cell bodies. In the dog some cardiac vagal notoneurones lie also in the nucleus ambiguus (Gunn, et al., 1968). Anatomical and electrophysiological studies in the cat have indicated that only tne nucleus ambiguus contains cardiac vagal preganglionic neurones (Szentagothai,

1952; Calaresu & Pearce, 1969; Kerr, 1969; Thomas &

Calaresu, 1974; McAllen & Spyer, 1976, 1978a; Spyer, 1979;

Spyer & McAllen, 1980). More recent studies have used the method of retrograde transport of horseradish peroxidase to trace central pathways. In these studies in the cat

(Sugimoto, Itoh, Mizuno, Nomura & Konishi, 1979; Geis &

Wurster, 1980) ~nd the rat (Nosaka, Yamamoto & Yasunaga,

1979) cardiac vagal preganglionic neurones have been found in both the dorsal motor nucleus and the nucleus ambiguus as well as in the area between these two nuclei.

Nosaka, et al (1979) suggested that vagal cells all lie dorsally early in development and then migrate ventrally.

The species differences in distribution of vagal motoneurones may reflect differing degrees of migration in the different species.

(b) Properties of cardiac vagal fibres. The vagus contains about 30,000 fibres. Eighty percent of these are afferent and twenty percent efferent of the

6,000 efferent fibres, approximately 2,000 are myelinated and 4,000 unmyelinated (Agostini, Chinnock, Daly & Murray,

1957). The cardiac branch of the vagus contains about

3,000 fibres, 2,500 of which were found not to degenerate after supranodose vagotomy (Agostini, et al., 1957). This 14 gives a figures, therefore, of only about 500 fibres (or < 2 percent of the total vagal population) efferent to the heart. Agostini, et al (1957) found that much of this efferent supply to the heart consisted of unmyelinated fibres. The function of such unmyelinated efferents is obscure as it is known that most cardioinhibitory fibres are p or small myelinated fibres (Middleton, Middleton & Grundfest, 1950; Daly & Evans, 1953; McAllen & Spyer,

1976).

The conduction velocity of cardioinhibitory fibres has been measured in many studies and ranges from 2.8 to 30m/s

(Heinbecker, 1930; Brown & Eccles, 1934; Middleton, et al.,

1950; Iriuchijima & Kumada, 1964; Jewett, 1964; Kunze,

1972; McAllen & Spyer, 1978a). By applying the conversion factor of 6 (Hursh, 1939), the diameter of these cardio~ inhibitory fibres ranges from 0.5 to 5_µ.m; i.e., confirming them to be fa, or small myelinated fibres. Only a few electrophysiological studies show that some cardioinhibitory fibres are unmyelinated (Heinbecker, 1930; Daly & Evans,

1953; McAllen & Spyer, 1976). Possibly, methodoligical problems have led to an under~stimate of the number of small myelinated fibres, as the numbers of such fibres reported varies greatly between animals, and it is usually only possible to demonstrate a· few degenerated small myelinated fibres after intracranial or supranodose vagotomy (Daly & Evans, 1953).

The cardioinhibitory vagal axons form part of the 15

parasympathetic outflow of the vagus from the medullary vagal nuclei (see section 1.3a) through the neck and thorax

to synapse on post-ganglionic cells in the heart itself.

Knowledge of the physiological properties of this intra­

cardiac synapse is clearly vital for understanding fully

the cardiac responses to vagal pre-ganglionic activity, but it has not been studied in detail, presumably because

of mechanical difficulties. Acetylcholine is the transmitter

at both pre- and post-ganglionic synapses and slows the

heart by its action on pacemaker cells (Burgen & Terroux,

1953; Hutter & Trautwein, 1956).

(c) Afferent Pathways. Medullary cardiac vagal motoneurones can be identified by 'backfiring' through

electrical stimuli delivered to cardiac vagal filaments

(Spyer, 1979). They are frequently identified also by

their pulse and respiratory rhythmicity. The pulse­

synchronous activity is attributable to arterial

baroreceptor inputs (McAllen & Spyer, 1978b). Both arterial

baroreceptors and arterial chemoreceptors influence vagal

activity. These receptors are innervated by the vagus

(cranial nerve X) and glossopharyngeal (cranial nerve IX)

nerves and synapse initially in the nucleus of the tractus

solitarius (Cottle, 1964; and see below). The nucleus of

the tractus solitarius is located dorsally in the medulla,

close to the obex (see fig 1.1). The glossopharyngeal nerve

synapses in the rostral and middle part of the nucleus,

while the vagus synapses in the middle and caudal part of

the nucleus (Cottle, 1964). The carotid baroreceptors and 16 chemoreceptors have been more extensively studied than their aortic counterparts because of their easier access.

Neurones in the nucleus of the tractus solitarius and its vicinity have been shown to respond to selective baroreceptor

(Trzebski, Peterson, Attinger, Jones & Tempest, 1962;

Biscoe & Sampson, 1970b; Muira & Reis, 1972; McAllen &

Spyer, 1972; Lipski, McAllen & Spyer, 1975; McAllen,

Jordan & Spyer, 1980) and chemoreceptor (Muira & Reis, 1972; Davies & Edwards, 1973; Lipski, McAllen & Trzebski,

1976) inputs. However, electrical stimulation of the carotid sinus nerve has most often been used to study the early connections of these pathways. Short latency responses have been evoked in the nucleus of the tractus solitarius (Humphrey, 1967; Sampson & Biscoe, 1968; Muira

& Reis, 1968, 1969; Seller & Illert, 1969; see also Biscoe & Sampson, 1970a, b; Lipski, McAllen & Spyer, 1975;

Trzebski, Lipski, Majcherczyk & Chruscielewski, 1975;

Lipski & Trzebski, 1975) confirming this nucleus as a primary relay station for afferents from the carotid sinus nerve.

Earlier studies, with recordings of medullary neurones with cardiac rhythm (Smith & Pearce, 1961; Salmoiraghi,

1962; Fussey, Kidd & Whitwam, 1967; Humphrey, 1967; Crill

& Reis, 1968), or degeneration studies of the glosso­ pharyngeal and vagus nerves (Ingram & Dawkins, 1945;

Cottle, 1964) had implicated the solitary tract and its nucleus as part of the baroreceptor pathway. Stimulation of this region (Calaresu & Pearce, 1965) mimicked the 17 response to carotid sinus nerve stimulation. An extensive study by Seller & Illert (1969) localised the first synapse of the primary afferent fibres from the carotid sinus nerve to the nucleus of ·the tractus soli tarius. Seller & Illert gave a detailed quantitative description of transmission at this synapse. Unmyelinated (or C) carotid sinus afferent fibres also terminate monosynaptically in the nucleus of the tractus solitarius (Jordan & Spyer, 1977). A monosynaptic input into the paramedian reticular nucleus was also proposed by Muira & Reis (1968, 1969) but other workers have been unable to find such an input (Biscoe &

Sampson, 1970a,b; Spyer & Wolstencroft, 1971) and it is suggested that unsatisfactory stimulating techniques allowing current spread to other afferent nerves may be the explanation for this discrepancy (see Lipski, McAllen & Spyer, 1975).

Electrical stimulation of the carotid sinus nerve has been an important tool for locating the first synapse in baroreceptor and chemoreceptor pathways (see above). The carotid sinus nerve contains both A (myelinated) and C (unmyelinated) baroreceptor and chemoreceptor fibres

(Fidone & Sato, 1969) so that selective stimulation of one or other receptor type by electrical means, while often claimed, is likely to be unattainable. Also evidence will be presented in this thesis (chapter 8) that the cardiac vagal motoneurones can behave differently in response to electrical and to functional stimulation. This has also been suggested by others for neurones earlier in the 18 pathway, for example, in the nucleus of the tractus solitarius (Lipski, McAllen & Trzebski, 1976). Thus some disadvantages attach to the use of non-specific electrical stimuli.

The other major area of baroreceptor and chemoreceptor sensory endings is located around the aortic arch.

Baroreceptor and chemoreceptor fibres run together in the aortic nerve in most species and in the dog and cat the aortic nerve is part of the vagosympathetic trunk (Douglas

& Schaumann, 1956; Edis & Shepherd, 1971). In the rabbit, the aortic nerve contains only baroreceptor fibres and is a separate nerve (Chalmers, Korner & White, 1967). Recent electrophysiological and autoradiographic studies have confirmed the earlier anatomical work (Ingram & Dawkins,

1945; Kerr, 1962; Cottle, 1964) that aortic afferents for the cat (Gabriel & Seller, 1970; Jordan & Spyer, 1978; Kalia & Welles, 1980) and the rabbit (Jordan & Spyer, 1978; Garcia, Jordan & Spyer, 1979; Wallach & Loewy, 1980) also terminate in the nucleus of the tractus solitarius.

4. REFLEXES THAT EXCITE THE VAGUS

(a) Barorecepto~ ~eflex. In 1924, H.E Hering first demonstrated the importance of the carotid sinus region in regulating blood pressure. Mechanical stimulation of the carotid sinus region or electrical stimulation of the carotid sinus nerve, caused hypotension and bradycardia. 19

These were reflex responses and were abolished by cutting the carotid sinus nerve. ·He also noticed that these nerves tonically restrained blood pressure because cutting them . led to systemic hypertension. Bronk & Stella (1935) recorded from baroreceptor fibres in the carotid sinus nerve and noted that there was a burst of impulses with each arterial pressure pulse. The mean discharge rate of carotid baroreceptors is higher for a pulsatile pressure than for a steady pressure of the same mean value (Ead,

Green & Neil, 1952; Heymans & Neil, 1958). The aortic arch baroreceptors also give pulse-synchronous discharges, although they do not have a markedly enhanced response to pulsatile over steady pressures in the way that the carotid baroreceptors do (Douglas, Ritchie & Schaumann, 1956;

Angel James & Daly, 1970) and, in general, have higher threshold pressures than the carotid receptors (e.g.,

Pelletier & Shepherd, 1973). Mean discharge rate of the arterial baroreceptors is a signal which has a high correlation with systolic, mean or diastolic blood pressure

(Arndt, Morgenstein & Samodelov, 1977). Whether the pulse synchronous nature of the discharge gives other information which the central nervous system uses (e.~, heart rate) is doubtful.

Koch (1929, 1931) was the first to show the relationship between carotid sinus pressure and systemic pressure. He constructed the now familiar sigmoid curves to describe the operating characteristics of this reflex as it affected

systemic blood pressure. He showed that there was a 20 threshold level for baroreceptor response of about 50 mm Hg pressure within the carotid sinus and that above about 210 mm Hg there was no further response. Curves of the type he constructed show that the maximal gain of the baroreceptor reflex is in the physiological range of blood pressure levels. Similar sigmoid curves are now commonly used also to correlate pulse interval against systolic pressure or mean arterial pressure, the slope of the curve indicating the sensitivity of -the baroreceptor - cardiodepressor reflex

(Smythe, Sleight & Pickering, 1969; Korner, 1971). These curves are used to compare the behaviour of the baroreceptor - cardiodepressor reflex in various situations and have been used, for example, to describe the contribution of higher centres to the functioning of the baroreceptor reflex (Korner, Shaw, West & Oliver, 1972; Korner, 1971, 1979).

Central connections of the baroreceptor pathway. From the nucleus of the tractus solitarius, a direct projection to the hypothalamus has been shown using autoradiographic techniques (Ciriello & Calaresu, 1980). Stimulation of the hypothalamus has long been known to inhibit the cardiac vagal limb of the baroreceptor reflex

(Abrahams, Hilton & Zbrozyna, 1960; Hilton, 1963; Hilton & Spyer, 1971; Gebber & Klevans, 1972; Lopes & Palmer,

1977; Coote, Hilton & Perez-Gonzalez, 1979). Korner and co-workers have made extensive studies on the contribution of suprabulbar centres to the control of heart rate through baroreceptor reflexes. They studied the baroreceptor 21 reflex in unanaesthetised rabbits with various lesions and found differences in the so-called 'set point' (a term which carries certain assumptions about reflex operation - see Streatfeild, Davidson & McCloskey, 1977, for discussion) and 'sensitivity' of the reflexes described by the curves in the vario~s preparations (reviewed by Korner, 1971, 1~8).

A functional rostral projection has also been suggested by Coleridge, Coleridge & Rosenthal (1976) who noticed a long lasting depression of pyramidal tract neuronal firing in response to a rise in carotid sinus pressure.

Similarly, inputs to the nervous system from other sensory receptors can affect the behaviour of the baroreceptor reflex (for example, see Korner, Uther & White, 1969;

Korner, 1970, 1971; Daly, 1972; Pelletier & Shepherd, 1975;

Kumada, Nogami & Sugawa, 1975). In general, the anatomical points of convergence have not been pin pointed, although the altered reflex behaviour seen following gross anatomical lesions has assisted analysis (Korner, Shaw, West, Oliver

& Hilder, 1973; Korner, West & Shaw, 1973). One particular interaction is the focus of much attention in this thesis it is the influence. of central and reflex respiratory inputs on the baroreceptor-cardiodepressor reflex. Inputs from arterial chemoreceptors are also discussed in section 1.4b, below.

Influences on to baroreceptor pathways from higher neural centres were mentioned above with reference to the hypothalamus in particular, and various brainstem 22

structures in general. There are other probable influences~ The complexity of pathways concerned make it difficult to perform experiments in many cases, however, and influences of emotional and complex behavioural factors - many of which may well exert powerful effects on baroreflex pathways - are poorly documented. Similarly, the effects of 'command'

- related factors (Freyschuss, 1970; Goodwin, Mccloskey & Mitchell, 1972) also deserve further study.

The basic reflex of interest here, the baroreceptor afferent to vagal efferent pathway, is not simple, but is probably a polysynaptic path with many potential points of influence. Electrical stimulation of the carotid sinus nerve excites neurones in the first relay station, the nucleus of the tractus soli tarius, within 3 - 21 ms (Trzebski, et al., 1975) but evoked potentials in the

cervical vagus do not appear for 29 - 120 ms ( Iriuchij ima & Kumada, 1964; Kunze, 1972; Potter, 1980; and see chapter 8 below). The responses of medullary vagal neurones in the nucleus ambiguus to electrical stimuli in the sinus nerve, and to pulses of pressure in the carotid sinus were

found by McAllen & Spyer (1978b) to have latencies of 20- 50 ms and 40 - 110 ms respectively : the additional latencies for pressure stimuli presumably occurring at receptor nerve endings. Only part of these delays can be attributable to conduction times. If the carotid sinus nerve is 5 cm from the nucleus of the tractus soli tarius, a delay of 0. 5 ms would apply to large ''A' fibres conducting at 100 m/s and 50 ms to unmyelinated · 'C' fibres. 23

Any further delay occurs within the central nervous

system. Clearly, a substantial, and probably largely

synaptic delay can be involved within the central nervous

system. Nevertheless, little is known of these later order

connections. McAllen, Jordan & Spyer (1980) have suggested

second order relays within the nucleus of the tractus solitarius, and Morest (1967) has suggested a potentially

rapid path of second order neurones projecting directly to the nucleus ambiguus. Unfortunately the fact remains, however, that too little is known of the baroreceptor to vagal pathway within the central nervous system.

(b) Chemoreceptor reflex. The afferent pathways of the chemoreceptors of the aortic and carotid bodies were outlined above.

Functional stimulation of arterial chemoreceptors, by

hypoxia, produces a primary reflex bradycardia, mediated

by both vagal excitation and sympathetic withdrawal in most species (Daly & Scott, 1962; Downing, Remensnyder &

Mitchell, 1962; Korner, 1965; Chalmers, et al., 1967; Carmody & Scott, 1974; Daly, Korner, Angell James & Oliver,

1978a; Korner & Other, 1970), systemic vasoconstruction, mediated by sympathetic excitation (Daly & Scott, 1962;

Downing, Remensnyder & Mitchell, 1962; Chalmers et al.,

1967; Korner & Other, 1970), and stimulation of breathing

(Heymans & Heymans, 1927; Bernthal, 1938; Daly & Scott, 1958, 1962, 1963; Daly, 1963; Daly, Korner, Angell James

& Oliver, 1978a). 24

Stimulation of aortic chemoreceptors produces similar primary cardiovascular responses to stimulation of the carotid chemoreceptors (Daly, Hazzledine & Howe, 1965;

Daly & Ungar, 1966; Angell James & Daly, 1969), but stimulation of breathing is much less effectively evoked from the aortic chemoreceptors (Daly & Ungar, 1966).

Because lung inflation is accompanied by an increase in heart rate (Anrep, Pascual & Rossler, 1936a) and peripheral vasodilatation (Salisbury, Galletti, Lewin & Rieben, 1959;

Daly, Hazzledine & Ungar, 1967; Daly & Robinson, 1968), the cardiovascular responses to chemoreceptor stimulation depend on the interaction of primary reflexes and the secondary reflexes initiated through intrapulmonary receptors by increased breathing. A further complication, and one which has been seldom recognised in this area, is that central inspiratory processes also oppose primary chemoreceptor reflexes (e.g., Anrep, Pascual & Rossler,

1936b; and see the results presented throughout this thesis). Thus, even by controlling respiratory movements

(for example, by paralysing an experimental animal and ventilating it constantly) only the secondary reflex opposition to chemoreceptor reflexes is avoided - any increase in the activity in central neural 'centres' for respiration can proceed, and may mask primary cardiovascular responses.

Early reports on the results of chemoreceptor stimulation were conflicting, probably because the pr~mary reflex 25 stimulation of ventilation varies between animal species and from one animal to another (MacLeod & Scott, 1964; Daly

1972; Daly, Korner, Angell James & Oliver, 1978a) and also because the response to mild hypoxia can differ from the response to severe hypoxia (Korner, 1965; Chalmers,et al.,

1967). Furthermore, during prolonged periods of hypoxia, the initial primary responses of stimulation of ventilation and bradycardia are attenuated (Carmody & Scott, 1974).

Only when ventilation is controlled are the primary reflex bradycardia (Bernthal, Green & Revzin, 1951; Daly & Scott,

1958, 1962) and systemic vasoconstriction seen (Bernthal,

1938; Daly & Scott, 1962; Daly & Ungar, 1966), and even then, as noted above, interference from central inspiratory activity remains likely.

(c) Diving response. It is well known that diving animals such as seals or whales, are able to endure long periods of submersion - in some whales over an hour

(see Anderson, 1966; Angell James & Daly, 1972a for review). They can do this largely because of a complex interaction of reflexes that enables them to restrict the blood supply to all but essential. organs, and to conserve the oxygen present by decreasing heart rate, and so oxygen consumption by the heart (see Irving, 1939 for review; also Andersen,1963). The diving response essentially consists of apnoea in the end expiratory position, bradycardia and peri.pheral vasoconstriction. The diving response can be observed in very many animals, not only in natural divers (although it is particularly marked 26

there), and is readily demonstrable in man (Scholander,

Hamel, Le Messurier, Hemmingsen & Garey, 1962; see also

Elsner, Franklin, Van Citters & Kenney, 1966; Daly &

Angell James, 1975; and also chapters 5 and 6). It occurs when the animal immerses its face in water and is evoked by stimulation of receptors around the nose (Andersen,

1963; Elsner, et al., 1966, Gooden, Stone & Young, 1974).

The bradycardia is vagally mediated (Richet, 1899).

Stimulation of nasal mucous membrane in the dog by water

also produces apnoea in the end expiratory position,

bradycardia and selective peripheral vasoconstriction

(Elsner, et al., 1966; Angell James & Daly, 1972b).

Maintenance of the diving response (bradycardia and

peripheral vasoconstriction) during prolonged submersion

depends on stimulation of arterial chemoreceptors (Jones

& Purves, 1970; Daly, Elsner & Angell James, 1977;

Elsner, Angell James & Daly, 1977). If arterial

chemoreceptor stimulation is withdrawn by temporarily

perfusing the chemoreceptors of a seal with blood of a

high pO2 , bradycardia is attenuated (Daly, et al., 1977). Participation of arterial baroreceptors in the maintenance

of the diving response has also been suggested (Andersen, 1966; Angell James & Daly, 1972; Angell James, Daly &

Elsner, 1978). The stimulation of ventilation that usually

accompanies chemoreceptor stimulation is inhibited by

nasal reflexes (Angell James & Daly, 1973; Elsner, et al.,

1977), and so 'diving bradycardia' is seen in spite of

increased chemoreceptor stimulation. 27

Diving bradycardia, once established, is reduced by inflation of the lungs even when this does not lead to reduction in established hypoxia or hypercapnia (Daly &

Angell James, 1975). This effect can be understood as a manifestation of the reflex masking of the 'primary' diving (trigeminal plus chemoreceptor) response through the excitation of pulmonary stretch receptors.

(d) 0culocardiac and other reflexes. When pressure is applied to the eyes, apnoea and bradycardia are seen. This reflex - the oculocardiac reflex - was first described in 1908 independently by Dagnini and

Aschner, and, though regarded as a physiological oddity, has been of use to clinicians as a simple means of enhancing vagal tone in order to treat supraventricular tachycardias. Aschner demonstrated that the afferent side of the reflex was contained in the ophthalmic branch of the trigeminal nerve and the efferent side was the vagus.

Since that time there have been many reports of cardiac arrest or arrhythmias during ophthalmic surgery (for example, Deacock & 0xer, 1962; Katz & Bigger, 1970; Apt,

Isenberg & Gaffney, 1973; Alexander, 1975). The slowing of the heart is accentuated by cessation of respiration (see

(bt(c) above) and, as with the diving response, bradycardia has been reported to be overcome by inflation of the lungs

(Aserinsky & Debias, 1963; see Chapters 6 and 7 following).

Presumably, lung inflation excites intrapulmonary receptors and these reflexly inhibit established vagal tone. 28

Other fo11Ds of nasopharyngeal stimulation evoke effects

like the diving response or the oculocardiac reflex.

McRitchie & White (1974) showed that nasopharyngeal

irritation by cigarette smoke evoked apnoea, bradycardia and vasoconstriction in unanaesthetised rabbits. Angell

James & Daly (1972b) reported that the passage of water through the upper respiratory tract in anaesthetised dogs evokes similar responses - despite the anaesthesia blunting and often abolishing, a 'diving' response to application of water only at the snout. Some prominence, also has been given to pharyngeal reflexes evoking apnoea and bradycardia in foetuses in response to the application of 'foreign' fluids (Johnson, Robinson & Sal~sbury, 1971). And finally,

it has been known for many years that gross electrical stimulation of the superior laryngeal nerve can evoke apnoea, and usually bradycardia (Koepchen, Wagner & Lux,

1961; Kordy, Neil & Palmer, 1975; Angell James & Daly, 1978;

Daly, et al., 1978b) especially when chemoreceptor stimulation occurs simultaneously (Kordy, et al., 1975;

Angell James & Daly, 1978; Daly, et al., 1978b).

(e) Ventricular receptors. The Bezold-Jarisch

reflex was first described in 1867 by Bezold and Hirt and was studied in detail by Jarisch and co-workers in the

1930s and 1940s (see Jarisch & Zotte11Dan, 1949). An

intravenous injection of a veratrum alkaloid evokes

bradycardia and hypotension, an effect which is blocked by

vagotomy. Later work by Dawes (1947) showed that the

endings sensitive to veratrum are located mainly in the 29 left ventricle. These fibres have since· been shown to be mechanoreceptors situated in the myocardium, in or near the epicardium, with fibres that have conduction velocity of about lm/s and so are mainly 'C' fibres (Sleight, 1964;

Coleridge, Coleridge & Kidd, 1964; Sleight & Widdicombe,

1965). These fibres have little tonic activity (Muers &

Sleight, 1972) and so are thought not to be significantly involved in the tonic regulation of blood pressure (Thoren,

1976). They are, however, sensitive to nicotine (Sleight,

1964) and adrenaline (Sleight & Widdicombe, 1965; Muers &

Sleight, 1972) and when stimulated by these agents, fire in systole. Because of this property, they have been claimed then to monitor intramyocardial tension and therefore left ventricular contractility (Sleight, 1976). The major stimulus to ventricular receptors is a rise in intramyocardial tension in the wall of the left ventricle and this may be brought about physiologically be several mechanisms including catecholamine stimulation to the heart working at low filling volumes, or with an increase in ventricular distension. This reflex may be physiologically important

in man, therefore, before and during exercise, and may also be involved in certain ill-understood syndromes

involving syncope (Sleight, 1976).

(f) Juxtapulmonary capillary receptors. In

studying ihe Bezold-Jarisch reflex described above, Dawes

(1947) also described another site of action for veratridine

- the lungs. He found that veratridine injected into the

pulmonary circulation caused an inhibition of breathing. 30

He also saw a bradycardia and hypotension but this effect was much weaker than the cardiovascular response elicited when the drug was injected into the coronary circulation.

Dawes, Mott & Widdicombe (1951) looked at over one hundred compounds for their effects on this reflex. They found that over thirty of the compounds studied could elicit the

Bejold-Jarisch reflex. Two, the most active they studied, were 2 r:J. -naphthyl ethyl isothio urea and phenyldiguanide.

They found that these compounds, as well as evoking the

Bezold-Jarisch reflex, also reflexly halted breathing in expiration and they localised the receptors responsible for this response to the lungs. Paintal (1954), in his search for gastric stretch receptors, also used phenyl­ diguanide and isolated two pulmonary afferent fibres from the right vagus that were sensitive to phenyldiguanide.

He suggested that these fibres might be responsible for the reflex effects described by Dawes, et al (1951). The sensory endings of these fibres were located close to the pulmonary capillaries and are called 'juxtapulmonary capillary' or simply J-receptors. Paintal (see Paintal,

1973, for review) subsequently studied these receptors in some detail and found that they have mostly unmyelinated axons with conduction velocities of 0.8 - 7m/s and run in the vagus. The physiological stimulus for these endings is pulmonary congestion, or specifically, the increase in interstitial volume consequent on a rise in pulmonary capillary pressure. They are usually silent, but can be stimulated experimentally by injection of phenyldiguanide into the right atrium or right ventricle. When stimulated 31

J-receptors cause reflex apnoea (in the end-expiratory position) hypotension and bradycardia. An interesting 'side effect' of their stimulation (which.may come to be recognised as their major action, in time) is the wide­ spread inhibition of motoneurones throughout the body

(Deshpande & Devanandan, 1970; Paintal, 1973).

(g) Vasovagal syncope. Vasovagal syncope or fainting is sometimes induced by severe haemorrhage, tilting or emotional reactions. This response is due to marked vagally mediated bradycardia, peripheral vasodilation, and the consequent fall in blood pressure (Barcroft &

Edholm, 1945) due to release of sympathetic constrictor tone alone or activation of sympathetic cholinergic vasodilator nerves to the vessels (Blair, Glover, Greenfield

& Roddie, 1969). Cardiac receptors have been shown to play a role in initiating at least some forms of vasovagal syncope (Oberg & White, 1970) possibly through the activation of ventricular 'C' fibres (Oberg & Thoren, 1972; see also ,(e) above). Thoren (1979) suggested that the mechanism of activation of these ventricular 'C' fibres is an mincreased sympathetic outflow and a lowered ventricular filling" which gives a "powerful contraction around an almost empty chamber", and that such circumstances might thereby trigger a vasovagal syncopal response.

Such suggestions, however, appear to derive from a view that all cardiovascular adjustments are reflexly-mediated.

While it might well be that certain centrally-generated emotional responses use reflex mechanisms, such as those 32 mentioned here, as intermediate steps, there seems to be only a philosophical impediment to the view that vasovagal responses are evoked from within the CNS by signals primarily generated there.

5. SINUS ARRHYTHMIA

Inspiration is accompanied by an increase in heart rate. This has been known at least since the time of Hering

(1871). It was not until 1936, however, that the mechanisms underlying the variation in heart rate with breathing were elucidated. Anrep, et al (1936a, b) described "two distinct mechanisms underlying the respiratory cardiac arrhythmia,

a reflex and a central mechanism". They showed that resting vagal activity, me~sured by recording heart rate could be

inhibited either by a reflex mechanism activated by

inflation of the lungs or by activity in central inspiratory

centres in the absence of lung movements. Rijlant (1936a,b)

showed the inhibitory action of central inspiratory centres

(as indicated by phrenic nerve activity) by recording from the cardiac vagus in the chest in a paralysed dog (see fig 1.2). Because sinus arrhythmia persists in the

paralysed animal, tachycardia being synchronous with phrenic

nerve activity, it has been suggested that central inspiratory

inhibition of the vagus and excitation of cardiac sympathetics

are responsible for sinus arrhythmia (Joels & Samueloff, 1956). 33

. ..., .• ,~.~~------~ Nnl ,-,.. £.... , ..... ""'~-1n~M!lti i,i,.t~ I'!'! 1!1!~"1111, H11'f!-.l;l·~-~lll~Oil~I 1.. b;t;t~...... nc.1~ ..... l i-- c..s,...i..:.., ...

-,- J- X

ti;. I. - Chirn; n,,,,·1tl1inr.. cur-,r,:. lli,...-.:&iun de 1,,11s I,·, 1-.u•1o·a11, ,h• ...... u:,:.1,hi•111•· ••roil. S..-c:lion Jr. 10111 I.:• J';am•· .. 11, urtl,<•·~•111••llti,111•·•.' l',,l~i;r:unme ~.alhoili,1m·.

From R ijlant ( 1936).

Fig 1.2. Dog, anaesthetised with morphine and paralysed

with curare. Top panel shows phrenic nerve

activity. Middle panel shows activity in a

filament dissected from the cardiac vagus in

the chest. Bottom panel shows the electro­

cardiogram. During phrenic nerve activity

(and central inspiratory activity) discharge in

the cardiac filament is inhibited. When the

phrenic nerve is silent, activity in the cardiac

filament increases and this is accompanied by

bradycardia. This figure is reproduced from

Rij lant, 1936 b. 34

There have been several other mechanisms proposed to

account for the variations in heart rate with breathing.

The most notable of these being the Bainbridge reflex.

Bainbridge (1920) proposed that "each inspiration increases the diastolic filling of the heart and gives rise to reflex acceleration of the pulse rate". Recently, Linden (1973) has shown that stimulation of left and right atrial receptors leads to an increase in heart rate, mediated by

sympathetic excitation rather than vagal withdrawal. The work of Melcher (1976) gives some support to the hypothesis that increased atrial filling during inspiration can contribute to the inspiratory increase in heart rate. It has also been suggested that the changes in blood pressure

1throughout the respiratory cycle might, through the

arterial baroreceptor reflex, evoke cyclical changes in heart rate (e.g., Matthes & Ebeling, 1948). The arterial pressure oscillations that accompany breathing are not sufficient alone to account for sinus arrhythmia (Melcher,·

1976) but the exaggerated sinus arrhythmia seen when blood. pressure is raised has been demonstrated to be facilitated by the arterial baroreceptor reflex (Przybyszewski &

Trzebski, 1980). The arterial chemoreceptors have also been shown to facilitate sinus arrhythmia because, if the

arterial chemoreceptors are briefly inhibited by having

a subject take a single breath of oxygen, expiratory

bradycardia is reduced. If chemoreceptors are stimulated

by hypoxia and hypercapnia, however, sinus arrhythmia is

increased (Tafil & Trzebski, 1980). 35

It is clear from the references just reviewed in brief that resting heart rate, and therefore vagal activity are

influenced by many factors. Apart from neural sources of

inhibition associated with inspiration itself.- a lung reflex and a central inhibitory mechanism - cyclical changes in cardiovascular and respiratory variables can also contribute. Melcher's (1976) recent review deals with these effects.

6. RESPIRATORY MbDIFICATIONS OF CARDIODEPRESSOR

REFLEXES

(a) Carotid sinus nerve stimulation. Electrical , stimulation of the carotid sinus nerve was used by Koepchen, et al (1961) to look at the interaction of breathing with

inputs delivered along carotid sinus afferents. Their

important paper, in German, has received little recognition

in the English language literature until recently. Koepchen

and his colleagues, in anaesthetised but unparalysed dogs,

showed that the primary reflex bradycardia evoked by stimulation of the carotid sinus nerve was seen only when

stimuli were delivered to the carotid sinus nerve in the

expiratory phase of breathing - that is, during expiratory

air flow or in the expiratory pause. Similar stimuli

delivered to the carotid sinus nerve during inspiration

evoked no reflex bradycardia (see also Iriuchijima &

Kumada, 1964; and fig 1.3). 36

10sec

,_;, '"J':''' l,1•'•'' '1.''IJ'•, 'ci'.'''" '' OI-J,~ 1' '.I)). ,I.Ill,.,_.,

11 : . .i I, ' '1 ·":'- ' I I I : ' I ' : : . ! I ! ; I i ' ; : I . I ' I ' , 'I : f: ,, : . ' I '. /I / : '' I ' .. , . :' ',· ;i. ·:·, :' . '; '' 1·: ' : ., ,:,I ' ' ; i .-.' '

1 I: ·1 ; l,'\1· i1-· '.::: i/ij;' . l,1' '/\11/':1··. ·::;!_;( -~·-·. i 1 ''..· I . I 1 • .. ,H I f J' . .I. : • ,. ' ; / I l ; r ' ., ' i I f ' , :•• . . ,,- II . ' , : :: :. ' I t1 ' ' ' 'I '1·' ,, f I 1 ; / i ; :i r ; 1I · :, :HI I I I,· ;: /. I i i J : i ;,1,, : i , 1 , '; i . 1 , U I , i I I , 1 1 ,:, r ; 1, 11 '. i l I ·; I ; I I I I I,.. • : • I ' • ' I ' ' •. ' • From Koepchen et al (1961)

Fig 1.3. Dog, anaesthetised with morphine and chloralose.

Top trace shows electrocardiogram (EKG); second

trace shows movements of the thorax (Thoraxumfang);

third trace shows the timing of electrical

stimuli to the carotid sinus nerve within the

respiratory cycle (Sinusnervenreiz); fourth

trace shows arterial blood pressure (art~rieller

druck) and the bottom trace shows the heart

period (the time between heart beats) in seconds

(Herzschlagabstd). Stimuli given during

inspiration (second and third stimuli)have no

effect on heart period. Stimuli given in

expiration (first and fourth stimuli) evoke an

increase in heart period (i.e., bradycardia).

Reproduced from Koepchen, et al (1961). 37

They showed also that stimuli given some seconds apart, even when in separate but succeeding expiratory pauses, summated in their effects on heart rate. Subthreshold expiratory stimuli, which alone were not sufficiently strong to cause a decrease in heart rate, when given as a pair in an expiratory pause summated to slow the heart.

Such pairs of subthreshold stimuli given during inspiration, or one in inspiration and another in expiration, did not summatein this way. See fig 1.4. Thus, the effect of a stimulus given in the expiratory phase of breathing, raises central excitability of cardiodepressor pathways for several seconds, but stimuli given during inspiration do not raise excitability levels at the time the stimuli are given, nor in succeeding expiratory phases.

Koepchen, et al (1961) also made observations on the reflex effects of brief, intracarotid pressure pulses, which they used as baroreceptor stimuli. Again, stimuli delivered in expiration, but not those delivered in inspiration, slowed the heart. These observations were confirmed in paralysed animals in which the cycling of central respiratory neurones was monitored by recording from a phrenic nerve. The inspiratory refractoriness of the baroreceptor-cardiodepressor reflex could thus be attributed, at least in part, to the activity of central

inspiratory mechanisms.

In 1976, Lopes & Palmer briefly reported that the bradycardia usually evoked by electrical stimulation of the 38 . II I p I I I = EKG I I I I I I I I I I I r I ' I f I I I I I L

l ~ 1/Jorrrxumfong l 11 I I 11 I I

I '. I ' ; . '. ' .. I I ' '.. '. ' : ' i I i I : .i , I j ; ; ' ! ; ; ,. I I , : Ii .. : I I I 'I ! ; ., . ; j ' ' . ' . I ' .. ' 10sec From Koepchen et al (1961)

Fig 1.4. Dog, anaesthetised with morphine and chloralose.

Top trace shows electrocardiogram (EKG); second

trace shows movements of the thorax (Thoraxumfang);

third trace shows the timing of stimuli to the

carotid sinus nerve within the respiratory cycle

(Sinusnervenreiz), and bottom trace shows the heart

period (Herzschlagabstand). Panel 'a' shows that

single subthreshold stimuli to the carotid sinus·

nerve, given in successive expirations, have no

effect on heart period. Panel 'b' shows that

two such subthreshold stimuli given in the same

expiratory pause can summate to cause an increase

in heart period. Panel 'c' shows one subthreshold

stimulus given in inspiration followed by a

subthreshold stimulus in the following expiration

do not summate - that is, there is no change in

heart period. Panel 'd' shows one subthreshold

stimulus given in expiration followed by one

subthreshold stimulus given in the following

inspiration also do not summate to affect heart

period. Reproduced from Koepchen, et al (1961). 39 carotid sinus nerve could be blocked by sustained inflation of the lungs as well as by central inspiratory activity.

Their report is unconvincing, however, as it seems likely that quite large, unphysiological inflation pressures were used. Although they gave no figures on which to judge this matter, their figure le, for example, shows a lung inflation that sufficiently impeded venous return for. arterial pressure to fall by approximately 60 mm Hg within

10 seconds. Such large inflations can ·Stimulate tachycardia directly through excitation of intrapulmonary receptors

(Anrep, et al. ,1936a), and the secondary cardiovascular change just mentioned would also contribute to tachycardia.

Models of possible vagal excitatory pathways were suggested by Koepchen, et al (1961), and these are discussed in chapter 9, together with models suggested by others, and those suggested in this thesis.

(b) Functional stimulation. Haymet & Mccloskey

(1974, 1975) used functional stimulation of the arterial baroreceptors and chemoreceptors to study the responsiveness of these reflexes throughout the respiratory cycle. They showed that, as with electrical stimulation of the carotid sinus nerve (Koepchen, et al., 1961; Iriuchijima & Kumada,

1964) inspiration inhibited the separate baroreceptor and chemoreceptor reflexes.

Black, Mccloskey & Torrance (1966, 1971) showed that the

carotid chemoreceptors are extremely sensitive to sudden 40

changes in pC02 , although responses to sustained changes

are not dramatic. In parallel to this study, Black &

Torrance (1967, 1971; see also Eldridge, 1972) showed that a sudden retrograde injection into the carotid artery of saline equilibrated with co2 stimulated inspiration when given during inspiration but had little effect if given in expiration. Using similar brief injections of co2 - equilibrated saline to stimulate arterial chemoreceptors selectively, Haymet & McCloskey (1975) showed that these stimuli evoked prompt and pronounced decreases in heart rate when given in expiration and little or no change in heart rate when given in inspiration. A brief pulse of pressure applied to the carotid sinus (as used by Koepchen, et al, 1961) stimulates the arterial baroreceptors selectively. The stimulus, when given in expiration, evokes a prompt and marked decrease in heart rate and has

little or no effect when given in inspiration. By recording

directly from baroreceptor and chemoreceptor fibres in

the carotid sinus nerve, Haymet & Mccloskey (1975) were

able to confirm the selectivity of their stimuli.

These findings highlight the early.dilemma faced by Daly

and others in studying responses to sustained chemoreceptor

stimuli. In giving maintained hypoxic stimuli to the

chemoreceptors they were looking at the results of

interactions of conflicting reflex and central effects.

By giving brief stimuli, Haymet & Mccloskey (1975) were

able to confine stimuli to any selected part of the

respiratory cycle and to observe very prompt heart rate 41 responses before respiratory 'side effects' became prominent.

Where the respiratory rate is high under anaesthesia, it has sometimes proved difficult to achieve precise inspiratory or expiratory timing of brief chemoreceptor stimuli. This was the experience of Daly, et al (1978a) in the anaesthetised monkey, where the respiratory rate was of the order of 30 breaths /minute. In the same experimental animals, however, the same workers (1978b) reported marked bradycardia in response to chemoreceptor stimuli delivered when inspiratory efforts were inhibited by stimulation of the superior laryngeal nerve.

The demonstrations of this respiratory modulation of the baroreceptor and chemoreceptor reflexes were first achieved in anaesthetised dogs and cats. Since then, similar results have been obtained in conscious human subjects for arterial baroreceptor stimulation (Melcher,

1976; Eckberg & Orshan, 1977, 1980; Trzebski, 1980). Neck suction has been shown to stimulate arterial baroreceptors in man (Ernsting & Parry, 1957; Thron, Brechmann, Wagner

& Seller, 1967). Brief pulses of pressure given only in expiration evoke a prompt bradycardia, as described above in animal experiments. Similar stimuli given in inspiration, have markedly less effect.

The brief selective stimuli described by Haymet &

Mccloskey (1975) have been used in the work described· in 42 this thesis to examine the mechanism of inspiratory inhibition of cardiodepressor reflexes. Other cardiodepressor reflexes have also been studied here in animals and man.

7. EXPERIMENTAL CONSIDERATIONS

(a) Identification of cervical cardiac vagal

efferent fibres. Recordings from cardiac vagal efferent fibres were first accomplished in 1936 by Rijlant

(see fig 1.2) in an experiment performed in a paralysed dog with an open chest. The technical difficulties faced by Rijlant in recording from the cardiac vagus (i.e., open chest, heart movements, induced electrical activity from the electrocardiogram) were not overcome until 1963 when Jewett established methods for identifying cardiac vagal efferent fibres in the cervical vagus, so allowing the recording from such fibres away from the heart and in spontaneously breathing animals.

Jewett (1963, 1964) compared the activity in efferent filaments of the cervical vagus wi_th activity he recorded in fibres that had been anatomically identified as going to the heart. He was able to identify fibres in the cervical vagus with similar properties to cardiac fibres in the thoracic vagus.

Jewett encountered and identified a variety of fibres in the efferent vagus. Fibres firing in synchrony with 43 inspiration were most easily encountered and probably supply laryngeal muscles, but expiratory firing, bronchoconstictor and probable gastrointestinal efferents were also found.

The fibre types (i-vii) that Jewett categorised are summarised in the table in fig 1.5.

Activity in vagal efferent fibres reflects arterial baroreceptor influences. Discharge is increased when blood pressure is raised and decreased when blood pressure is lowered. This change in activity with changes in blood pressure has been shown to be correlated with changes in heart rate (Jewett, 1963, 1964; Katona, et al., 1970).

Sometimes also, pulse modulation is very marked. Cardiac vagal fibres are typically silent during inspiration and this is due to the inhibitory effects of both lung receptors and activity in central inspiratory centres

(Anrep, et al., 1936a, b; and this thesis). These characteristics form the basis of the identification of cardiac vagal efferent fibres in the neck, and eliminate the need to open the chest for anatomical identification of fibres. The use of these characteristics in the present study is described in the Methods s~ction (chapter 2).

(b) Effects of anaesthesia. Many physiological experiments on animals require the use of anaesthetics.

This was so for the majority of the animal experiments described in this thesis, particularly those involving nerve recordings. Chloralose was the major anaesthetic used here, usually after thiopentone induction (see 44

CLASSIFICATION OF VAGAL EFFERENT FIBRES {Jewett, 1964)

CHARACTERISTIC FIRING PATTERN FUNCTION

J; t with t BP 2. with BP TYPE I t t CARDIOINHIBITORY 3. Pulse modulation 4. Silent in inspiration 1. t with t BP 2. t with BP TYPE II t BRONCHOCONSTRICTOR 3. Pulse modulation 4. No respiratory modulation TYPE Ill 1. lnspirotory firing INNERVATES LARYNX TYPE IV 1. Expiratory firing INNERVATES LARYNX t. t with t BP TYPE V 2. t with t BP CARDIOINHIBITORY (?) 3. little tonic activity 1 •Pulse synchronous 2. Silenced by carotid occlusion ABERRANT (?) C.SINUS TYPE VI 3. t with t BP BAROREC EPTORS 4. t with t BP t.No effects from BP TYPE VII (?) 2. No respiratory modulation

Fig 1. 5. Table summarising types of fibres found in

the efferent vagus by Jewett, with

characteristic firing patterns related

to possible physiological role. 45

Methods, chapter 2). The potential disturbances to physio­ logical function caused by anaesthesia are clearly matters of concern.

It has been shown that pentobarbitone has a vagolytic action and so increases heart rate in an anaesthetised rabbit (Korner, Uther & White, 1968). Further studies on the rabbit (Korner, Langsford, Starr, Uther, Ward & White, 1968) showed that both barbiturate and chloralose anaesthesia modified the cardiovascular response to hypoxia by effectively abolishing the vagal response while minimally altering the sympathetic response. In the dog, however, chloralose has little effect on resting heart rate or blood pressure (Cox, 1972a), although pentoba:rt.bitone has similar effects to those described above in the rabbit (Cox 1972b). Responses to changes in blood ·pressure may be exaggerated in such animals (see also Brown & Hilton, 1956; Armstrong,

Porter & Langston, 1961),, although this is not always borne out by comparisons between such responses in anaesthetised and unanaesthetised animals (Thames & Kontas,

1970; Scher & Young, 1970). In man, various anaesthetics not including chloralose, were tested and were found in general to depress the sensitivity of the baroreceptor­ cardiodepressor reflex (Bristow, Prys-Roberts, Fisher,

Pickering & Sleight, 1969).

While such studies indicate that quantative aspects of reflex function may be altered by anaesthetics, there is little to suggest that results obtained under anaesthesia 46 give qualitatively incorrect pictures of reflex behaviour.

In a recent authoritative review, Korner (1979) has given prominence to findings reported in abstract form by

Kirchheim, Gross & Brandstetter (1974) suggesting that respiratory modulation of cardiodepressor reflexes is markedly exaggerated under anaesthesia, and that in the unanaesthetised state such modulation ''is of little consequence in baroreflex control of heart rate". The demonstration in conscious human subjects of clear respiratory modulation of heart rate responses to pulses of neck pressure (described above: see Melcher, 1976;

Eckberg & Orshan, 1977, '1980; Trzebski, 1980) and of more complex cardiodepressor responses, as will be described here in chapter 6, make such a view difficult to sustain. 47

CHAPTER 2

METHODS 48

1. WHOLE ANIMAL EXPERIMENTS WITH OBSERVATIONS ON HEART RATE

Experiments in this category were performed on nineteen dogs of both sexes, weighing 9 - 19 kg. The animals were anaesthetised with intravenous thiopentone (Pentothal, Abbott, 25mg/kg), followed by a warmed intravenous chloralose solution (~-chloralose, British Drug Houses, 80mg/kg). In each dog the trachea was cannulated low in the neck and nylon canrtulae were inserted with their tips facing towards the heart into the lingual and external carotid arteries on both sides. The tips of these cannulae lay close to each other facing into the carotid sinus. A nylon cannula was inserted in a femoral vein for further administration of anaesthetic or drugs. Rectal temperature was maintained at 37 - 39°c.

A further twenty one dogs were studied for the effects of respiratory influences on the sympathetic control of heart rate (Chapter 3). These animals were prepared similarly to those described above, except that they were anaesthetised with intravenous pentobarbitone (Nembutal, Abbott, 35mg/kg- no premedication or induction), and cannulae were inserted into lingual and external carotid arteries on both sides.

Arterial pressure was measured from either a lingual or external carotid artery, using a Statham P23 AC transducer, and was recorded on one channel of a Grass polygraph pen 49

recorder. In some animals a signal proportional to tracheal air flow was obtained by passing a nylon tube (2 mm internal diameter) into the trachea and measuring the pressure drop between its tip and the atmosphere, using a Statham P23 Db transducer. This signal was then recorded on the Grass polygraph. Alternatively, respiratory activity was measured by a bag-in-box method similar to that described by Donald & Christie (1949) in which the animal inspired through a valve from a bag enclosed in an airtight box into which expired air was led; pressure in the box, which was related to the tidal volume of the breath taken, was measured using a Grass PT5A volumetric pressure transducer. The electrocardiogram and the beat­ by-beat heart rate (Grass 7P4 cardiotachometer, triggered from the e.c.g) were also recorded.

In paralysed animals respiratory activity was recorded from the central end of the cut and de-sheathed phrenic nerve through platinum electrodes, and this phrenic neural discharge was integrated using a Grass 7P 3B preamplifier ('leaky' integrator; time constant b.05 sec). To produce paralysis intravenous D-tubocurarine (1 - 2 mg/kg) was given slowly until all respiratory movements ceased: the animals were then ventilated on pure oxygen with a Starling 'Ideal' pump, usually adjusted so that some phrenic nerve activity remained. Observations on the reflexes tested were made during periods when the pump was temporarily stopped.

The lungs were denervated for some of the experiments 50

described in Chapter 4, in eight dogs, after a control series of observations had been made. This was done on the right-hand side by opening the chest through the fourth intercostal space and cutting the pulmonary vagus close to the hilum of the lung. On the left-hand side the vagus was cut in the neck, the right vagus being left intact. Evidence of successful pulmonary denervation was the abolition of the Hering-Breuer inflation reflex. This method is described by Daly & Scott (1958). Resting heart rates were little altered after pulmonary denervation.

Four reflexes were tested in anaesthetised dogs. The baroreceptor reflex, the chemoreceptor reflex, the oculocardiac reflex and the reflex elicited by nasopharyn­ gal stimulation. This last reflex is similar to the diving reflex or diving response (see Introduction). Brief baroreceptor and chemoreceptor stimuli were given as

described by Haymet & Mccloskey (1975). Baroreceptor stimuli were delivered by sudden retrograde injections of 2 - 5 ml of air-equilibrated saline into the external carotid artery, after first clampin·g the common carotid artery below the carotid sinus. The chemoreceptor stimuli were provided by sudden retrograde injections into the

external carotid artery of small volumes ( 0. 2 ... 0. 5 ml) of warmed saline equilibrated with C02. Injections of similar small volumes of saline equilibrated with air were always without reflex effect. These small volumes did not change carotid sinus pressure, and have been shown previously not

to affect baroreceptor nerve endings (Haymet & Mccloskey, 51

1975). Stimuli were delivered during the inspiratory or expiratory phases of breathing (as judged from the record of phrenic nerve activity), or during the inflation of the lungs and at various times after inflation while the lungs were maintained in the inflated state by a constant positive pressure, usually of 8 - 10 mm Hg. The lungs were inflated by blowing into a tube attached to the tracheal cannula (a record of achieved pressure on an oscilloscope showed the experimenter the pressure achieved). Stimuli were also delivered during deflation of the lungs from this pressure (see Chapter 4).

The oculocardiac reflex was elicited by placing the thumbs on the closed eyelids of the animal and pressing.

No measurement of the pressure used was made, but similar pressures were applied to awake human subjects in other experiments (see below) without causing great discomfort.

Nasopharyngeal stimulation was achieved by a method similar to that described by Angell James & Daly (1972).

A cuffed, endotracheal tube was inserted into the trachea pointing rostrally and advanced until its tip lay in the nasopharynx. The tube was then tied firmly into the trachea, and the cuff was inflated. Tap water at room temperature was led into this tube when nasopharyngeal stimulation was required, and was collected through a rubber glove tied over the muzzle as it flowed out of the nose and mouth. The rate of flow of water was usually

11/min. 52

2. NERVE RECORDING EXPERIMENTS

Forty-four further animals were used for the experiments in which recording of single preganglionic cardiac vagal fibres were made. These were prepared for surgery in a slightly different way from those described in section 1 above. This was done in an attempt to preserve the high vagal tone that is typically seen in conscious dogs. Both anaesthesia and surgery favour stimulation of the sympathetic nervous system and withdrawal of vagal tone (see Introduction). Although chloralose anaesthesia is said by some to enhance vagal tone (possibly there is an elevated blood pressure and decreased heart rate, suggesting increased baroreceptor activity: Jewett, 1963), in the experiments in which nerve recordings were made, cardiac vagal fibres were often difficult to locate and so were presumed to be silent because of predominating sympathetic activity.

Because of the more extensive surgical preparation of the animal for nerve recording than for the experiments described in section 1 above, it was assumed that there was some sympathetic stimulation associated with this also. It has been part of conventional experimental physiological wisdom since the experiments of Anrep et al

(1936a) that morphine stimulates the cardiac vagus, so all animals were premedicated with morphine sulphate

(1 - 2 mg/kg) subcutaneously. Thirty minutes to an hour

later they were anaesthetised with a warmed intravenous chloralose solution(~ -chloralose, British Drug Houses, 53

60 - 100 mg/kg) after induction with thiopentone (15 mg/kg). In each dog the trachea was cannulated low in the neck and a nylon cannula was placed in the femoral vein for further doses of anaesthetics (as described in section 1 above). A balloon-tip cannula was inserted through the femoral artery and advanced so that the inflatable balloon lay in the upper abdominal aorta. This was used to raise central arterial pressure mechanically by obstruction of the abdominal aorta, to enable ease of identification of cardiac vagal fibres (see below). Rectal temperature was kept between 37° - 39°C.

In most animals nylon cannulae were inserted into the external carotid and lingual arteries on the right hand side, so that their tips lay close together, facing into the carotid sinus. One cannula, usually the lingual cannula, was used for recordings of arterial pressure within the carotid sinus. Through the other cannula, selective baroreceptor or chemoreceptor stimuli were given as described in section 1 above. Breathing and heart rate were also measured as described above in section 1, and were recorded on a Grass polygraph. In those animals in which carotid sinus nerve stimulation was used,only the lingual artery was cannulated for measurement of arterial blood pressure. The rest of the gross surgical preparation of this group of animals was as described above.

Each animal was prepared for nerve recording in the 54 following way. Through a mid-line incision in the neck the entire pharynx and larynx were removed from just above the sternum up to the level of the hyoid bone, and the skin flaps were raised to make a pool which was then filled with liquid paraffin. The vagus nerve on the right side was then cut, its central end desheathed and laid across an earthed, stainless steel plate with a blackened upper surface. The vagus nerve on the right side was used as there are several reports (see Hamlin &

Smith, 1968, or Scott & Reid, 1955) that the right vagus has a more powerful cardiac slowing effect. If correct this may reflect the anatomical distribution of vagal endings, or mean that the right vagus contains more cardioinhibitory fibres. However, there are also reports that there is no difference in the bradycardia evoked by stimulating either right or left vagus (see Loeb &

Vassalle, 1978; Hamlin & Smith, 1968).

The nerve was then divided into filaments under a microscope with fine jeweller's forceps. Neural activity in the filaments was recorded by lifting them, one by one, on to fine stainless steel electrodes which were connected to preamplifiers (Neurolog NL 103/106 : variable band pass, usually between 50 Hz - 1 KHz) and then to a speaker for audio monitoring and to a storage oscilloscope.

Nerves were considered to be cardiac efferents if they demonstrated a cardiac rhythm (- often not pronounced, and discernible only on the loudspeaker when systemic 55

arterial pressure was raised: see below) and an inhibition of their activity during inspiration (Jewett,

1963t 1964). Sometimes, especially when vagal tone was high, this inhibition was incomplete but was still pronounced (see also Trzeb.ski, 1980 and figs 4.1 and 4.2).

Confirmation of this characterisation was always obtained in at least one, and usually both of two ways: (i) a response could be evoked in the nerve by single electrical stimuli applied to the central end of the carotid sinus nerve (Iriuchijima & Kumada, 1963, 1964), or (ii) the nerve increased its discharge while maintaining its cardiac and respiratory rhythms, in response to a

'mechanical' elevation of blood pressure produced by inflating the aortic balloon. Records of cardiac vagal efferent activity were obtained by direct photography from the oscilloscope screen. In some experiments spikes were led to a spike trigger (Neurolog NL200) which produced pulses that were then processed using a Neurolog

NL750 Averager. Post-stimulus histograms of cardiac vagal activity were accumulated using the carotid sinus nerve stimulus or baroreceptor puise to trigger the averager, and histograms were then displayed on the oscilloscope screen and photographed.

The vagal filament was considered to contain a single active axon if: (a) the amplitude of the spikes varied by no more than the existing noise level; (b) if spikes had similar shapes when inspected at fast sweep speed on the oscilloscope (for example see fig 8.4); (c) if 56

spikes were never superimposed; and (d) if a clear refractory period free of further spikes always succeeded each spike. These criteria are made clear by reference to figs 2.1 and 2.2, which show the influence of noise level on apparent spike amplitude, and the refractory period

(criterion (d) above), respectively.

The dogs used in experiments involving recordings from cardiac vagal fibres were paralysed with pancuronium

(Pavulon, Organon : 40 - 80µg/kg). Curare was not used as it stimulates histamine release (Koelle, 1975) and this leads to hypotension with consequent loss of vagal tone. A ganglion-blocking action of curare may also contribute to the hypotension (Koelle, 1975).

Respiratory activity was recorded from the central end of the cut and desheathed phrenic nerve using platinum electrodes, as described in section 1 above. The paralysed animals were ventilated on pure oxygen using a

Starling 'Ideal" pump, adjusted so that phrenic nerve activity was not entirely suppressed. Intra-tracheal pressure was recorded in these animals during artificial ventilation.

For experiments requiring stimulation of the carotid sinus nerve a Grass SD9 isolated, square wave stimulator and platinum electrodes were used. The nerve was identified anatomically and then confirmed as the carotid sinus nerve if both heart rate and blood pressure 57

A B

Fig 2.1. Illustration of significance of criteria for

definition of single nerve fibres. Recording of

activity in a single active axon of a respiratory

motoneurone in a dog. Panel A shows two traces at

a fast sweep speed to demonstrate that an individual

action potential can be found at varying positions

in the noise level, thereby altering the apparent

spike height. Panel B shows the effect of noise

level on amplitude of the same action potentials at

a slower sweep speed: as the noise level increases

(achieved here by varying the filtering frequencies)

the variability of the apparent amplitude of the

action potential increases. Note that the variability

of amplitude does not exceed the background noise 58 CVE : natural activity

BP~lOO

BP__,150

lOms

Fig 2.2. Dog, anaesthetised with chloralose. Records of

activity in a single cardiac vagal efferent

fibre at two levels of blood pressure. The

sweep of a storage oscilloscope was triggered

repeatedly from the action potentials and many

such sweeps were superimposed, then photographed.

It can be seen that all spikes are similar in

shape and vary in amplitude by less than the

noise level. No spikes occur in the natural

or effective refractory period after a triggering

spike. At the higher rate of discharge (lower

panel), achieved by inflating an intra-aortic

balloon to raise arterial pressure, a natural

refractory period is preserved, although

reduced in duration. 59

decreased in response to its stimulation at 30 Hz, 1 ms, ,1ov. Usually the nerve was not cut before being stimulated: in some experiments, however, it was crushed between the carotid sinus and the stimulating electrode

(see chapter 8). For the experiments on vagal responses the nerve was stimulated at 1 ms, 1 - 10 V, and ~1 Hz unless otherwise stated. The stimulating voltage used was the smallest necessary to evoke a maximal fall in blood pressure during the test stimulation at 30 Hz (c. f

Seller & Illert, 1969). When required, such stimuli could be delivered only in expiration or only in inspiration. In some experiments the stimuli to the carotid sinus nerve were triggered from the electrocardio gram.

3. HUMAN EXPERIMENTS

Experiments were performed on fifteen healthy subjects of both sexes. Electrocardiograms were measured from conventional limb leads, and beat by beat heart rate was obtained using a Grass 7P4 cardiotachometer triggered from the e.c.g. Both were recorded on the polygraph. For both diving reflex and oculo-cardiac reflex studies subjects were usually seated. Diving reflexes were elicited by the subjects immersing the nose and mouth in a small bowl of cold water. Oculo-cardiac reflexes were elicited by an experimenter placing the thumbs on the subject's closed eyelids and pressing. The pressure was 60

firm, but subjects were instructed to tell the experimenter as soon as any discomfort occurred; all the results reported here were obtained in experiments in which the subjects reported no discomfort.

In the diving experiments the subjects, of course, stopped breathing while their faces were immersed. In the experiments on the oculo-cardiac reflex, subjects were instructed to stop breathing while the eye pressure was applied (see Chapter 6). Although no consistent differences between reflexes elicited with the breath held in inspiration and those in which it was held in expiration were noted, subjects were asked usually to commence breath-holding at the normal end-expiratory point. In the course of a 'dive' or of application of eyeball pressure during a breath-hold, subjects performed one of two simple manoeuvres in response to an instruction from an experimenter. The subjects were required either to swallow or to make a single inspiratory effort against a closed glottis ('false breath'). Chapter 6 gives further details.

In another series of experiments the effects on heart rate of voluntary breath-holding during normoxic and hypoxic conditions were compared.

All of these experiments were performed with the written consent of the subjects, and in accordance with the ethical requirements of this University. 61

CHAPTER 3

RESPIRATORY MODULATION OF BARORECEPTOR AND CHEMORECEPTOR REFLEXES MEDIATED BY THE VAGUS AND THE SYMPATHETIC NERVOUS SYSTEM 62

It is known (see Introduction) that brief selective stimulation of either carotid arterial baroreceptors or chemoreceptors (see Methods) evokes reflex bradycardia only, or most readily, when such stimuli are given in the expiratory phase of the respiratory cycle. Similar stimuli given during the inspiratory phase of the respiratory cycle are ineffective (Koepchen, Wagner & Lux, 1961; Haymet & Mccloskey, 1975). Inspiration is inhibitory to the reflex bradycardia evoked by stimulation of the arterial baroreceptors or chemoreceptors. See fig 3.1 which is reproduced from Haymet & Mccloskey (1975).

In this chapter experiments are described which show that the bradycardia evoked in expiration, from stimulation of arterial baroreceptors or chemoreceptors, has both vagal and sympathetic components.

1. VAGAL RESPONSES

Responses to brief baroreceptor and chemoreceptor stimuli were always examined first ·with the vagi intact.

The findings of Haymet & McCloskey (1975) were confirmed there was a prompt and pronounced bradycardia evoked when the stimuli were given in expiration, but little or no change in heart rate when they were given in inspiration.

In three animals propranolol was then given (usually 1 mg/kg) in a dose which was sufficient to abolish the 63

Tidal 100 vol [ iI ml 0

ECG

RAP mm Hg

BP mmrlg

Time (sec} I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I

Fig 3.1. Dog, anaesthetised with chloralose. Records of

tidal volume (inspiration upwards), electro­

cardiogram, right atrial pressure, and arterial

pressure are shown. At A, Band C injections of

co2 - saline were made into the carotid bifurcation to provide brief chemoreceptor stimulation. When

these stimuli were deliver~d in expiration (A and

C) they evoked a prompt slowing of the heart, and

an expiratory effort. When the stimuli were

delivered in inspiration (B), they evoked an

increased inspiratory effort but no change in

heart rate. As a control, the chest was squeezed

at D to mimic the thoracic and atrial pressure

changes seen with the chemoreceptor stimuli given

in expiration, but this did not affect heart rate.

(From Haymet & Mccloskey, 1975). 64 heart rate response to administration of 10 - 20 )lg isoprenaline. Baroreceptor and chemoreceptor stimuli were again given and evoked prompt and marked bradycardia when given in expiration. Stimuli given in inspiration were ineffective. An example of this can be seen in fig 3.2. Subsequent vagotomy (2 animals) or administration of atropine (1 animal) abolished the responses to both baroreceptor and chemoreceptor stimuli.

2. SYMPATHETIC ·EFFECTS

Responses to brief baroreceptor and chemoreceptor stimuli were examined before vagotomy in eighteen dogs.

The findings of Haymet & McCloskey (1975) were again confirmed. The animals were then vagotomised and given atropine (1 mg repeated hourly to ensure removal of vagal effects). Electrical stimulation of the cardiac ends of the cut vagi was then without effect on heart rate. There then remained a slight sinus arrhythmia. This could be attributed to the waxing and waning of sympathetic tone with the respiratory cycle because in all animals it was abolished at the end of the experiments by administration of propranolol (1 mg/kg). It was found that this sympathetically mediated sinus arrhythmia was most pronounced when the respiratory rate was slow. This was achieved by keeping the level of anaesthesia deep, and by maintaining the animal at a slightly cool temperature

(35 - 370c). 65

ecg Heort rote /rrvn

CSm Pressure mm Hg

tt...... e i e e i e 10 sec

Fig 3.2. Dog anaesthetised with chloralose: IV propranolol. These records show reflex effects of brief chemo­ receptor and baroreceptor stimuli delivered at different points in the respiratory cycle - during expiration marked 'e', or during inspiration marked 'i'. Electrocardiogram, heart rate, carotid sinus blood pressure and tracheal air flow are shown. In the panel on the left are shown the effects of three successive chemoreceptor stimuli . ( injections of O. 5 ml co2 - equilibrated saline into the carotid bifurcation): the first and third stimuli, which were given during the expiratory phase of breathing, evoked a prompt reflex brady­ cardia; the second stimulus, given during inspiration, did not affec·t the heart rate. In the panel on the right are shown the effects of three successive baroreceptor stimuli (injections of approx 3 ml air-equilibrated saline into the carotid bifurcation after clamping the common artery): the first and third stimuli, which were given during the expiratory phase of breathing, evoked a prompt reflex bradycardia; the. second stimulus, given during inspiration, did not affect heart rate (the sensitivity of the tracheal air flow trace was altered between records). 66

Brief selective baroreceptor or chemoreceptor stimuli were again given to see whether sympathetic withdrawal could contribute to the reflex bradycardia seen when arterial baroreceptors or chemoreceptors are stimulated.

(a) Baroreceptor responses. In ten of eighteen dogs tested there was a slowing of at least 15 beats / minute (range 15 - 25) whenever baroreceptor stimuli were timed to occur in expiration, but less than 10 beats/ minute (range 5 - 10) when the stimuli were timed to occur in inspiration (for these comparisons the heart rates at corresponding points in the two respiratory cycles before stimulation were taken as controls). Typical responses of this type are seen in fig 3.3.

In the remaining animals, there was either no response to the brief baroreceptor stimuli whenever delivered (four dogs) or a small (5 beats/ minute) and variable response, which bore no demonstrable relation to the respiratory cycle (four dogs).

The responses to baroreceptor stimulation were typically much slower after vagotomy than when the vagi were intact. Slowing usually commenced about 1 - 2 seconds and was most marked about 5 - 7 seconds after the stimulus was given. Fig 3.4 shows the time course of a typical response, which can be compared with the very rapid responses seen when the vagi were intact (fig 3.2). 67

Air flow insp up m-nrn1117TD"11i7-n11111117, I I I I I I I I I I I I I I I I I I I I Heart I I I I rote 200( I I /min i I I i 160 I I I I I I I I I I I I I I I I C Sinus 150f Pressure a mm Hg J A .. l tA AL 50 e - "'I e i

20 sec

Fig 3.3. Vagotomised dog, anaesthetised with pentobarbitone

and given atropine. Record shows tracheal air flow

heart rate and carotid sinus blood pressure (both

common carotid arteries clamped). Four brief

baroreceptor stimuli were delivered, by simul­

taneous injections of approx 3 ml air-equilibrated

saline _into both carotid sinuses, at the markers.

The first and third stimuli (marked 'e') were

delivered during the expiratory phase of breathing

and slowed the heart rate more than the second and

fourth stimuli (marked 'i') which were delivered

during inspiration. 68

ecg

Hecrt rate 200[ 160 /rrin t I I I. I I I I I I I It I I I I C Sinus 5 sec /['''"'''""''''"""'::::·""'''',i .. ,,,, .. Pressure ~o[ nm Hg - 50

Resp rnovt _f\ insp up -----~----J\-

Fig 3.4. Vagotomised dog, anaesthetised with pentobarbitone

and given atropine. Record shows tracheal air

flow, heart rate, carotid sinus blood pressure,

and respiratory movements. A brief pulse of

pressure in the carotid sinus, caused by the

sudden injection of approx 3 ml air-equilibrated

saline, was a baroreceptor stimulus delivered

during the expiratory pha~e of breathing. The

time course of the reflex slowing of the heart

evoked by the stimulation is shown .. 69

The bradycardia evoked by stimulation of arterial baroreceptors was due to sympathetic withdrawal because administration of propranolol (1 mg/kg) abolished it.

(b) Chemoreceptor responses. Brief stimuli were delivered to the carotid arterial chemoreceptors by retrograde injections of 0. 2 - 0. 5 ml C(½-equilibrated saline into the external carotid arteries in the same eighteen vagotomised dogs in which the brief baroreceptor stimuli were given. In all these animals the stimuli were delivered simultaneously into both carotid bifurcations.

In six of these animals, chemoreceptor stimuli given during expiration evoked a reflex slowing of the heart of at least 15 beats / minute (range 15 - 20), while stimuli given during the inspiratory phase of breathing evoked ·. reflex responses of less than 5 beats / minute (range 0 - 5).

These same animals gave similarly modulated responses when baroreceptor stimuli were applied during different phases of the respiratory cycle. Records obtained from one of these animals are shown in fig 3.5.

In the experiments using chemoreceptor stimuli, marked changes in breathing were frequently evoked. It thus became difficult to compare the exaggerated sinus arrhythmia immediately following stimulation with that which preceded it. The six animals described above in which respiratory modulation of changes in heart rate was clearly shown were notably poor in their ventilatory responses to chemoreceptor stimulation. The most common 70

A:. flow insp up rrrm,n77777177771171·11117 I I I I I I I I I I I I I I I Heartrate l158J .-.-.,._- , I /min

C Sinus ~ P--essure mm Hg A + A A e I e i 20 sec

Fig 3.5. Vagotomised dog, anaesthetised with pentobarbitone

and given atropine. Record shows tracheal air flow,

heart rate and carotid sinus blood pressure. Four

brief chemoreceptor stimuli were delivered by

simultaneous injections of 0. 5 ml Co2-equilibrated saline into both carotid sinuses, at the·markers.

The first and third stimuli (marked 'e') were

delivered during the expiratory phase of breathing,

and slowed the heart more than the second and

fourth stimuli (marked 'i') which were delivered

during inspiration. 71 problem encountered in the other animals was that stimuli delivered during the expiratory pause evoked an immediate inspiratory effort. The tachycardia associated with such premature inspirations -probably masked any immediate direct reflex effects on heart rate. In many animals, however, a marked bradycardia occurred in the expiratory pause following the premature inspiratory effort (see fig 3.6). All of the changes in heart rate evoked by arterial chemoreceptor stimulation were abolished following administration of propranolol (lmg/kg).

3. DISCUSSION "(CHAPTER 3)

The experiments described here show that the effectiveness of carotid baroreceptor and chemoreceptor stimuli in evoking reflex changes in heart rate through the sympathetic nervous system depends on the phase of the respiratory cycle in which the stimuli are given. The sympathetic efferent components of these reflexes are thus modulated by the respiratory cycle in a manner similar to the vagal components (Koepchen, et al., 1961; Haymet & Mccloskey, 1975). Sympathetic modulation in the absence of vagal effects, and vagal modulation in the absence of sympathetic effects have been demonstrated here. The responses mediated by withdrawal of sympathetic tone are slower and smaller than vagally mediated reflex changes in the same direction, and are considerably more difficult to demonstrate than the vagal effects. 72

Air flow insp up

Heart rate 180[ /min 160

C Sinus Pressure mm Hg

10 sec

Fig 3.6. Vagotomised dog, anaesthetised with pentobarbitone

and given atropine. Record shows tracheal air flow,

heart rate and carotid sinus blood pressure. At

the marker, a brief bilateral chemoreceptor

stimulus was given. The stimulus was given during

the expiratory phase of breathing, but immediately

evoked a large premature breath. There was a

bradycardia following the evoked breath. 73

There have been reports that it is difficult to demonstrate a sympathetically based sinus arrhythmia following vagotomy or the administration of atropine (e.g., Anrep, et al., 1936b). Most of the animals in this study showed the phenomenon, probably because efforts to slow the respiratory rate meant that there was time for the rather sluggish sympathetic effects to develop fully with each breath. The same slow respiratory rate probably also gave time for the sympathetic reflex effects to become apparent. If this is so, the responses described here are likely to be much less marked in animals with a more normal respiratory rate. Certainly, it was difficult to demonstrate responses in faster breathing animals. Often responses were seen only after the respiratory rate had been slowed by moderate cooling or by deepening the level of anaesthesia. Because of the difficulty in demonstrating sympathetic effects on heart rate, only baroreceptor and chemoreceptor reflexes were studied.

Respiratory effects on reflexes involving the sympathetic nervous system have been described previously. Seller,

Langhorst, Richter & Koepchen (1968) showed that a more pronounced vasodilatation in the vascular bed of the gracilis muscle was evoked by electrical stimulation of the carotid sinus nerve applied in the expiratory phase of breathing, than by similar stimuli given in inspiration. It has also been found that electrical stimuli given to the carotid sinus nerve during the expiratory phase of breathing gives a stronger inhibition of abdominal, 74

cervical and lumbar sympathetic neural activity than

stimuli given during inspiration (Seller, et al., 1968;

Richter, Keck & Seller, 1970). Difficulty in interpreting

these results arises, however, because both baroreceptor

and chemoreceptor afferents would have been excited by electrical stimulation of the sinus nerve, and these are known to have opposite effects on sympathetic vascular

tone: chemoreceptor stimulation -causes sympathetic vasoconstriction (Daly & Scott, 1963), and baroreceptor

stimulation withdraws sympathetic tone (Koizumi, Seller,

Kaufman & Brooks, 1971). Seller, et al., (1968) argued

that the chosen intensities of electrical stimulation

applied to the sinus nerves in their experiments were

such as to stimulate baroreceptor afferents alone, without

exciting chemoreceptor afferents. This is supported by

their finding little evidence of respiratory responses to

their stimuli. Nevertheless, it is known that baroreceptor

and chemoreceptor afferents are represented in both the myelinated and the unmyelinated fibre groups of the sinus

nerve (Fidone & Sato, 1969), and complete selectivity in

any form of electrical stimulation would appear impossible.

Indeed, in other studies (e.g., Black & Torrance, 1971;

Eldrid6e, 1972), electrical stimulation of the sinus nerve has been employed specifically to study the effects of

excitation of the chemoreceptor fibres within it. If

Seller, et al (1968) are correct in assuming that their

electrical stimuli involved mainly baroreceptor afferents,

then their results are consistent with those presented

here: if they stimulated mainly chemoreceptors, then their 75 results and these are at variance. By looking at the sympathetic cont-rol of heart rate, however, these considerations have been avoided. They do not apply here because stimulation of either baroreceptors or chemo­ receptors will evoke a similar sympathetic effect on heart rate, namely, sympathetic withdrawal (e.g., Bronk, 1933;

Daly & Scott, 1958).

The respiratory modulation of cardiac sympathetic activity which is described here was clearly not imposed by phasic afferent traffic travelling along the vagi, because the vagi were cut in these experiments. Nevertheless, it remains possible that the modulation could be altered, perhaps augmented, by such afferent inputs in intact animals. An augmentation of the effect might indeed be expected from the work of Daly & Scott (1958), who showed that the activation of intrapulmonary receptors by inflating the lungs could accelerate the heart by both vagal and sympathetic reflex mechanisms. 76

CENTRAL AND PERIPHERAL FACTORS MODIFYING BARORECEPTOR AND CHEMORECEPTOR REFLEXES 77

Brief selective baroreceptor or chemoreceptor stimuli evoke a prompt reflex bradycardia only when given in the expiratory phase of the respiratory cycle. It was shown in chapter 3 that both vagal excitation and sympathetic withdrawal contribute to this reflex bradycardia. It was shown that the vagally mediated effects are prompt and marked, in contrast to the sympathetically mediated effects which are relatively small, difficult to demonstrate and follow a typically much slower time-course. In the remainder of the studies described in this thesis, the predominant vagal effects are the main focus of attention.

1. CENTRAL MODULATION OF REFLEX RESPONSIVENESS

A question arising from the results presented to this point is, what mechanism is responsible for inspiratory inhibition of the baroreceptor and chemoreceptor reflexes? Some indication of the answer has been given in the work of Koepchen, et al (1961) and Davidson, Goldner & Mccloskey (1976) who studied respiratory modulation of reflexes in paralysed animals. In paralysed animals the cyclic effects of lung inflation are absent. Koepchen, et al (1961) showed that the cardiodepressor effects of brief baroreceptor stimuli or of brief trains of electrical stimuli delivered to the sinus nerve are wholly or almost wholly inhibited during the activity of central inspiratory neurones (as monitored by recording the phrenic neurogram). 78

Davidson, et al (1976) confirmed the findings for brief baroreceptor stimuli, and showed that the effects of brief chemoreceptor stimuli were similarly modulated. Confirmation of the results of Davidson, et al (1976) provided the starting point for the results now reported. Further examples of the effects described by Davidson, et al (1976) are seen in figs 4.10 and 4.11, and are dealt with in more detail now.

In the studies described here, the findings of Davidson, et al (1976) have been extended by recording directly from cardiac vagal efferent fibres. In paralysed dogs, while the lungs were motionless (respiratory pump temporarily stopped), brief selective baroreceptor or chemoreceptor stimuli were given during the expiratory phase of central respiratory activity: this was indicated by the phase of silence in the phrenic neurogram. Such stimuli evoke a prompt increase in vagal discharge. Typical examples are illustrated in the records shown in figs 4.1 and 4.2. Similar stimuli given during central inspiratory activity (i.e., phrenic discharge) evoke little or no vagal firing. Furthermore, any prevailing tonic activity in the vagal efferent fibres is similarly inhibited during the period of phrenic activity. These effects are also shown in figs 4.1 and 4.2 (right panels).

Activity in central inspiratory centres, while the lungs remain at resting volume, can inhibit both tonic vagal discharge and the increments in discharge which are evoked 79 Phrenic

if tracheal p. [~ Vagal a.p's lrll/lllW•l~l'l~k~~lili.~ U,Jijil~IU i1u1 ,1~:·1\1111:~,~~i~ 250 ~------)l,______.,,._.,,.._, _____

0

2s

Fig 4.1. Dog, anaesthetised with chloralose, paralysed with pancuronium, respiratory pump temporarily halted. Records of 'integrated' phrenic neural activity (indicating central inspiratory activity), intra­ tracheal pressure, activity in a single cardiac efferent fibre dissected from the cervical vagus, and blood pressure measured in the carotid sinus, are shown. Cardi·ac vagal responses to brief, selective baroreceptor stimuli are shown. The left hand panel shows the response when a stimulus is given while the phrenic nerve is silent and the lungs are motionless. Note the prolonged response evoked in the vagal fibre (see chapter 7 also). The right hand panel shows a similar stimulus given during central inspiratory activity (indicated by phrenic discharge), but while the lungs remain motionless: the reflex increase in vagal discharge previously evoked by stimulation of baroreceptors is inhibited, together with tonic vagal discharge. 80

Phrenic

if tracheal p.

Vagal a.p's

tli:mi111a: ,,. ,, 11 :nut .,J.1 •

2s

Fig 4.2. Dog, anaesthetised with chloralose, paralysed with pancuronium, respiratory pump temporarily halted. Records of 'integrated' phrenic neural activity (indicating central inspiratory activity), intra­ tracheal pressure, and activity in a single c~rdiac efferent fibre dissected from the cervical vagus, are shown. Cardiac vagal responses to brief, selective chemoreceptor stimuli are shown. The left hand panel shows the response when a stimulus is given while the phrenic nerve is silent and the lungs are motionless. Note the prolonged response evoked in the vagal fibre (see chapter 7 also). The right hand panel shows a similar stimulus given during central inspiratory activity (indicated by phrenic discharge), but while the lungs remain motionless: the reflex in~rease in vagal discharge previously evoked by stimulation of chemoreceptors is inhibited, together with tonic vagal discharge. 81 by baroreceptor or chemoreceptor stimuli.

The results presented to this point indicate that central inspiratory activity can inhibit cardiodepressor reflexes from baroreceptors and chemoreceptors. They give no indication about whether or not other effects on these reflexes are brought about as a result of lung inflation.

In 1975, Haymet & McCloskey showed that both reflexes remained effective when the lungs were held inflated at the end - inspiratory position. They concluded from this that the refractoriness of vagal mechanisms during inspiration is not due to the activation of slowly adapting mechanoreceptors in the lungs or thorax. However, they did observe that "participation of intrapulmonary or thoracic receptors, particularly of a rapidly adapting kind, cannot be ruled out" by such results.

The experiments now to be described pursued this matter. Again, experiments were carried out in paralysed dogs (21 dogs) in periods of temporary ( '-1 minute) cessation of artificial ventilation, when centr.al inspiratory activity could be monitored through a phrenic neurogram and lung inflations imposed independently of it.

2. PERIPHERAL MODULATION OF REFLEX RESPONSIVENESS

( a) Effects of stimuli timed with respect to air flow into the lungs : fast ramps. Rapid inflations of 82 the lungs were achieved by an experimenter blowing into a tube connected to the tracheal cannula. Inflations in which the tracheal pressure was raised from atmospheric pressure by 6 - 10 mmHg in 1 - 2 seconds were delivered during the expiratory phase of breathing. Chemoreceptor or baroreceptor stimuli delivered during these ramps of pressure evoked no reflex bradycardia. There was no evidence of inspiratory activity in the phrenic electroneurograms during these ramps of pressure. The process of lung inflation, in the absence of evidence of central inspiratory activity, was therefore shown to be sufficient to block both reflexes

(see figs 4.3, 4.4).

This inhibition of the baroreceptor or ·chemoreceptor reflex by inflation of the lungs is transient. If the lungs are held inflated and brief baroreceptor or chemoreceptor stimuli are again given, they become increasingly more effective. Graded responses were observed in all tests, related tothe delay between attainment of the peak of pressure and delivery of the stimulus. Stimuli usually evoked a slight bradycardia as soon as 1 second after air flow into the lungs was completed. By 5 - 15 seconds after the inflow of air, baroreceptor and chemoreceptor stimuli again evoked the full reflex bradycardia (see figs 4 .. 3,

4.4). Thus, the phenomenon observed by Haymet & Mccloskey

(1975) in unparalysed dogs was confirmed here in paralysed animals: during prolonged periods of inflation (Hering­

Breuer inflation apnoea) both baroreceptor and chemoreceptor stimuli still evoked prompt bradycardia. As noted above, . 83

10 sec

Fig 4.3. Dog, anaesthetised with chloralose and paralysed

with D-tubocurarine. Integrated phrenic activity,

heart rate, carotid sinus blood pressure and

intratracheal pressure, recorded during a period

of temporary cessation of artificial ventilation,

are shown. The effects of baroreceptor stimuli

(pulses of intracarotid pressure at markers) on

heart rate can be seen. A control baroreceptor

stimulus given in the expiratory phase of

breathing evokes a prompt and large bradycardia

(panel at left). A stimulus given during an

increase in intratracheal pressure of 6 mm Hg in

1 - 2 sec evokes little or no bradycardia ( second

panel). Stimuli given after the incr~a~e in

intratracheal pressure, but while the lungs remain

inflated, regain .their effectiveness with time

(three panels at right). 84

fltnric mchcrge JVV\ _I\ ~ )\___ Heat vv-- \j'vv- rate ~ry /rrin 1J'1'J .. ..

C SinJs pres54.ft w,,.,. mm Hg l-,- ~ ~!-~,siJ .,.ro-tnxheal • • 'Y pre5Sl.re nm Hg J _n _f7 J7 10 sec

Fig 4.4. Dog, anaesthetised with chloralose and paralysed

with D-tubocurarine. Integrated phrenic activity,

heart rate, carotid sinus B.P and intratracheal

pressure, recorded during a period of temporary

cessation of artificial ventilation, are shown.

The effects of chemoreceptor stimuli (intracarotid)

injections of C02-saline at markers) on heart rate

can be seen. A control chemoreceptor stimulus given

in the expiratory phase of breathing evokes a

prompt and large bradycardia (panel at left). A

stimulus given during an increase in intratracheal

pressure of 6 mm Hg in 1 - 2 sec evokes 1 it tie or .no

bradycardia (second panel). Stimuli given after the

increase in intra tracheal pressure, .but while the

lungs remain inflated, regain their effectiveness

with time (two panels at right). 85

Haymet & McCloskey took this as evidence that tonically

active mechanoreceptors in the lungs and thorax excited by

inflation were not sufficient by themselves to block the baroreceptor and chemoreceptor reflexes.

The effects of lung inflation were also studied in single

vagal preganglionic fibres. Brief selective baroreceptor

or chemoreceptor stimuli given during expiration (phrenic

silence) evoke a prompt increase in discharge of cardiac

vagal efferent fibres (see figs 4.5, 4.6 (left panels)).

Responses to baroreceptor or chemoreceptor stimuli

delivered during lung inflation (and phrenic silence) are

inhibited. No reflex increase in discharge is seen (see

figs 4.5 and 4.6 (right hand panels)). Resting vagal

discharge, however, remains relatively unaffected. This

is a point of contrast with the effects of central

inspiratory activity. Lung inflation inhibits reflexly

induced increments in vagal discharge, while tonic vagal

firing is much less affected. Central inspiratory activity

inhibits both tonic and reflexly induced vagal firing

equally effectively. This difference is analysed further

in chapter 7.

(b) Effects of stimuli timed with respect to air flow

into the lungs: slow ramps. The effects of slower ramps of tracheal pressure were also tested. Pressure was

raised again by 6 - 10 mmHg, commencing during the

expiratory phase of breathing, but inflations were made 86 Phrenic

if tracheal p. [~ ------, ...

0

2s

Fig 4.5. Dog, anaesthetised with chloralose, paralysed with pancuronium, respiratory pump temporarily halted. Records of 'integrated' phrenic neural activity (indicating central inspiratory activity), intra­ tracheal pressure, activity in a single cardiac efferent fibre dissected from the cervical vagus, and blood pressure measured in the carotid sinus, are shown. (Note that these records come from the same vagal fibre as was illustrated in figs 4.1 and 4.2, so the reader can compare the effects of central inspiratory activity and lung inflation) Cardiac vagal responses to brief, selective baroreceptor stimuli are shown. The left hand panel shows the response when a stimulus is given while the phrenic nerve is silent and the lungs are motionless. Note the prolonged response evoked in the vagal fibre. The right hand panel shows a similar stimulus given during lung inflation, but while phrenic discharge is absent: this also fails to evoks a reflex increase in discharge, but note that tonic vagal activity is relatively unaffected. 87

Phrenic

if tracheal p. [~0 _ __,J./ Vogel a.p's *1~1~~111111~~-~... 11 •••r uauuu

2s

Fig 4.6. Dog, anaesthetised with chloralose, paralysed with pancuronium, respiratory pump temporarily halted. Records of 'integrated' phrenic neural activity (indicating central inspiratory activity), intra­ tracheal pressure, and activity in a single cardiac efferent fibre dissected from the cervical vagus are shown. (Note that these records come from the same vagal fibre as was illustrated in figs 4.1 and 4.2, so the reader can compare the effects of central inspiratory activity and lung inflation). Cardiac vagal responses to brief, selective chemo­ receptor stimuli are shown. The left hand panel shows the response when a stimulus is given while the phrenic nerve is silent and the lungs are motionless. Note the prolonged response evoked in the vagal fibre. The right hand panel shows a similar stimulus given during lung inflation, but while phrenic discharge is absent: this also fails to evoke a reflex increase in discharge, but note that tonic vagal activity is relatively unaffected. 88 more slowly so that the inflations proceeded for 2 - 15 seconds. Again there was no evidence of inspiratory activity in the phrenic electroneurograms during the inflations. Chemoreceptor and baroreceptor stimuli were delivered during these inflations: most commonly the stimuli were given at about the mid-point of the ramp of pressure. In all tests the effectiveness of both stimuli was clearly related to the rate of air flow into the lungs: for slow ramps of pressure (taking about 10 seconds to reach peak) the reflexes were as prompt and as complete as in the control conditions (i.e., as in the expiratory phase, without inflation); for ramps of intermediate duration (taking 3 - 6 seconds to reach peak), the reflexes were prompt, but typically evoked a smaller bradycardia than in the control conditions (see figs 4.7 1. and 4. 8).

(c) Effects of denervation of the lungs. The results described to this point indicate that lung inflation alone in the absence of central inspiratory activity is sufficient to inhibit the baroreceptor and chemoreceptor cardiodepressor reflexes. This inhibition is due to aphasic mechanism and the inhibitory effects are not maintained if the lungs are held statically inflated. See also the brief note by Trzebski (1980). To see whether the sensory receptors responsible for this inhibition lie in the lungs, the lungs were surgically denervated in eight dogs by cutting the vagal supply to the lungs (see Methods) leaving the supply to the heart 89

Phrenic discharge __f\___/1 _fL_ _(\__ _)1 Heart ~ rate ]J\( A/ /rrin • • 1y

I C Sinus 200[ / I i pressure I ' nm Hg 50 _J~ ----l__,, _JL -l-Ii

lntra-tracheoJ 'f' 'f' 'f' presue nvn Hg :[ Jl _n _/l 10 sec

Fig 4.7. Dog, anaesthetised with chloralose and paralysed with D-tubocurarine. Integrated phrenic nerve activity, heart rate, carotid sinus blood pressure and intratracheal pressure, recorded during a period of temporary cessation of artificial ventilation, are shown. The effects of baro­ receptor stimuli (pulses of intracarotid pressure at markers) on heart rate can be seen. A control baroreceptor stimulus given in the expiratory phase of breathing evokes a prompt bradycardia (panel at left). A stimulus. given during an increase in intra tracheal pressure of 6 mm Hg in 1 - 2 sec evokes little or no bradycardia (second panel). A stimulus given during a slower rise in intratracheal pressure (2 - 5 sec) evokes a slight bradycardia (third panel). A stimulus given during even slower rises in intratracheal pressure ( 10 - 15 sec) evokes a prompt bradycardia similar to the control response (fourth panel). (In this and following figures the heart rate trace is triggered beat-by-beat by the e.c.g and refers to the beat preceding its registration: i.e. , it lags behind the heart rate change by one beat. ) 90

Phrenic discharge JL JL jl__

Heart rate ~ /min /\v Jl\I' • • C Sinus ¾ pressure , tll 1 :a:~l;J!1un1ul , ,11% S i,1_1.1 ~!m,1,,1iimm~~ -i1,,1i~~!!!.1 mm Hg ,1~l~~l,l~(~i~~t!~ 1 ~im:11:.~ 111•' '""" '\Jrl I

Intra-tracheal "' pressure mm Hg l J7l _/l _i1 10 sec

Fig 4.8. Dog, anaesthetised with chloralose and paralysed with D-tubocurarine. Integrated phrenic nerve activity, heart rate, carotid sinus blood pressure and intratracheal pressure, recorded during a period of temporary cessation of artificial ventilation, are shown. The effects of chemo­ receptor stimuli (intracarotid injections of co2- saline at markers) on heart rate can be seen. A control chemoreceptor stimulus given in the expiratory phase of breathin~ evokes a prompt bradycardia (panel at left). A stimulus given during an increase in intratracheal pressure of 6 mm Hg in 1 - 2 sec evokes little or no bradycardia (second panel). A stimulus given during a slower rise in intratracheal pressure (2 - 5 sec) evokes a slight bradycardia (third panel). A stimulus given during even slo~er rises in intratracheal pressure ( 10 - 15 sec) evokes a prompt bradycardia similar to the control response (fourth panel). 91 intact. In this way cardiac vagal effects on heart rate would be preserved. When the lungs have been denervated, lung inflation no longer evokes a Hering-Breuer inflation reflex. This can be seen in the bottom panels of figs 4.9 and 4.10 which show typical records from animals following pulmonary denervation of this kind. It can be seen in those tracings that central inspiratory activity continues despite the maintained lung inflation - that is the procedure abolishes the Hering-Breuer inflation reflex.

After pulmonary denervation, brief stimuli to the arterial baroreceptors or chemoreceptors evoked a prompt bradycardia when delivered during air flow into the lungs, as well as during maintained inflations or during air flow out of the lungs (see figs 4.9, 4.10). That is, the inhibition of the reflexes which could be· caused during air flow into the lungs of animals with intact vagi could no longer be observed when the pulmonary vagi were cut.

Figs 4.9 and 4.10 illustrate all the phenomena associated with respiratory modulation of reflex-effectiveness and bear careful scrutiny as they illustrate and summarise the effects described in this chapter. The activity of central inspiratory neurones in the absence of lung expansion can block the reflexes and the movement of the lungs into an expanded state in the absence of central inspiratory activity can also block them. The reflexes are not blocked during maintained static inflation, nor during deflation. The effects of lung expansion are not 92 YAGI INTACT Heart rate /m'tn

Phrenic discharge

Intro-tracheal pressure mm Hg :[ ID= .

C Sinus pressure 200~ JI / / f I , mm Hg sot Wl,-JLJW),"\~

LUNGS DENERVATED Heart rate /min

Phrenic discharge

Intra- tracheal pressure mm Hg

C'Sinus pressure mm Hg

Fig 4.9. Dog, anaesthetised with chloralose and paralysed with D-tubocurarine. Upper and lower panels show heart rate, integrated phrenic activity, intratracheal pressure and· carotid sinus blood pressure, recorded during periods of temporary cessation of artificial ventilation. The effects of baroreceptor stimuli (pulses of intracarotid pressure) on heart rate are seen. When the vagi are intact (upper panel) baroreceptor stimuli evoke bradycardia except when delivered during periods of inspiration (first of the stimuli shown) or while the lungs are expanding in response to a rise in intratracheal pressure (third stimulus). After denervation of the lungs (lower panel), the stimuli remain ineffective during periods of central inspiratory activity (first stimulus), but now evoke a large bradycardia when delivered while the lungs are expanding in response to a rise in intratracheal pressure. 93 YAGI INTACT Heat rote /min J Phrenic discharge • Intra-tracheal pressure 10 sec mm Hg J r l C Sinus pressure mm Hg

Heart rote /min

Phrenic discharge

Intro· tracheal pressure nm Hg 1 C Sinus pressure mm Hg

Fig 4.10. Dog, anaesthetised with chloralose and paralysed with D-tubocurarine. Upper and lower panels show heart rate, integrated phrenic activity, intracheal pressure and carotid sinus blood pressure, recorded during periods of temporary cessation of artificial ventilation. The effects of chemo­ receptor stimuli (intracarotid injections of C02-saline at markers) on heart rate are seen. When the vagi are intact (upper panel), chemoreceptor stimuli evoke bradycardia except when delivered during periods of inspiration (second of the stimuli shown) or while the lungs are expanding in response to a rise in intratracheal pressure (third stimulus). After denervation of the lungs (lower panel), the stimuli remain ineffective during periods of central inspiratory activity (second stimulus) but now evoke a large bradycardia when delivered while the lungs are expanding in response to a rise in intratracheal pressure. 94 seen when the lungs are denervated.

The abolition of the inhibitory effects of lung inflation by pulmonary denervation is an important control. It shows that the receptors responsible for the inhibition lie in the lungs, and eliminates receptors elsewhere as possible contributors to the effect. For example, changes in blood gas tensions or in arterial pressure occurring secondarily to the lung inflations, and acting in some way through chemoreceptors or baroreceptors outside the lungs, are ruled out as causes of inhibition by this finding.

There are two major possibilities for the intrapulmonary receptor types which might be responsible for the inhibition. These are the slowly adapting pulmonary stretch receptors and the rapidly adapting 'lung irritant' receptors. Pulmonary stretch receptors fire on inflation, adapt slowly to a tonic, sustained level and then are silent during deflation. Activity arising from these receptors could well explain the effects seen.

However, the rapidly adapting receptors of Knowlton &

Larrabee (1946), or 'lung irritant receptors', must also be considered. These fire on inflation and deflation, especially when these are performed rapidly, but have

little or no discharge during sustained inflations. Again

the discharge properties might be suitable for the effects

described. Clearly the responses to deflation of the two 95 receptor types provide means of distinguishing them.

In all experiments in which inflating ramps were tested, the peak inflation pressure was maintained until chemoreceptor and baroreceptor reflex responsiveness was fully restored, and then the tracheal cannula was opened to the air and the lungs rapidly deflated. Chemoreceptor and baroreceptor stimuli given during these rapid deflations usually evoked a prompt and large bradycardia (see figs 4.9 and 4.10). Commonly, however, deflation evoked an inspiratory effort and the stimuli, which then fell within the inspiratory phase of breathing, were then ineffective. The failure of such deflations to inhibit cardiodepressor reflexes argues against the rapidly adapting 'lung irritant receptors' being the receptors responsible for the effects of lung inflations, described above. However, the deflations used were relatively small, and were achieved through passive recoil of the respiratory system, so the test may not have been conclusive.

The effects of large deflations were tested therefore, in four dogs. A pneumothorax was formed on each side of the chest so that each lung could be seen. A narrow, cuffed4 endotracheal tube was then tied into the trachea and advanced until its tip lay in the right main bronchus. With the cuff deflated both lungs were ventilated through this tube. When the cuff was inflated, only the right lung opened through the tube and the volume of the left lung remained unchanged. That this was so could be confirmed 96 by direct inspection of each lung. The left lung could then be inflated to 10 - 20 nmHg and maintained at this pressure while the right lung was inflated and deflated independently. Chemoreceptor and baroreceptor stimuli delivered during a fast inflating ramp of the right lung alone, or of both lungs together (10 - 20 mm Hg in 1 - 2 seconds) evoked no reflex bradycardia. However, the stimuli regained their effectiveness about 5 - 15 seconds after the ramp of pressure. The right lung was then rapidly deflated by suction applied to the endotracheal tube, so that it collapsed from its grossly inflated state to close to its residual volume within about 1 second. This deflation did not, however, evoke an inspiratory effort presumably because of a Hering-Breuer inflation reflex originating in the still-inflated left lung. Respiratory efforts recommenced when the left lung was allowed to deflate. In all four animals an indication that reflexes originating in the right lung were intact was obtained on evoking a Hering-Breuer inflation reflex by inflating the right lung while the left lung remained collapsed.

Chemoreceptor and baroreceptor stimuli were delivered by an experimenter watching the right lung and were timed to occur as the lung was close to complete collapse. They evoked their usual prompt and marked reductions in heart rate (see fig 4.11).

These more strenuous deflations, by failing to inhibit 97

Heat rate /rrin

JJ\/' JA 6 6

ntra-tracheal pres5'Xe '~ nm Hg 1 l ~ 1GL C Sinus • • • pre55Ul'e nm Hg 1 ~~~~ 10 sec

Fig 4.11. Dog, anaesthetised with chloralose and paralysed

with D-tubocurarine. Heart rate, integrated

phrenic activity, intratracheal pressure

(delivered to right lung only) and carotid

sinus B.P, recorded during two periods of

temporary cessation of artificial ventilation,

are shown. The effects of baroreceptor stimuli

(pulses of intracarotid pressure : left panel)

and chemoreceptor stimuli (intracarotid injec­

tions of Co2-saline at markers: right panel) on heart rate can be seen. In each panel the left

lung remains hyperinflated _from the start of the

record to the marker (open triangle) under the

phrenic record. Baroreceptor and chemoreceptor

stimuli evoke prompt and large bradycardia before,

but not dur~ng, a pressure rise inflating the

right lung. Both stimuli remain effective during

a rapid deflation of the right lung. Inflation

and deflation of the right lung, unaccompanied

by stimuli, cause little change in heart rate

(intratracheal pressure changes at far right). 98 cardiodepressor reflexes in the way inflations can, provide strong evidence against the 'lung irritant receptors' as being the intrapulmonary receptors involved.

3. DISCUSSION (CHAPTER 4)

Koepchen, et al (1961) and Haymet & Mccloskey (1975) have shown that during inspiration, baroreceptor and chemoreceptor effects on the heart rate are blocked. In experiments on paralysed animals, Davidson, et al (1976) showed that the activity of neural inspiratory mechanisms as indicated; by phrenic nerve activity, were alone sufficient to block these reflexes. The work described in this chapter confirms the work of Koepchen, et al (1961),

Haymet & Mccloskey (1975) and Davidson, et al (1976) in demonstrating that the presence of central inspiratory activity without movement of the lungs is sufficient to block these< reflexes. In addition, the work has been extended here as it has been shown. also that lung expansion without central inspiratory activity also blocks both reflexes. In the intact animal both mechanisms would act together. This would assure the effectiveness of the block.

The inhibition of both reflexes caused by expansion of the lungs is attributable to the excitation of receptors in the lungs because it is abolished by denervation of the 99 lungs. The sensory elements involved are sensitive to the rate of expansion of the lung, and the effectiveness of the inhibition which they cause is dependent on this rate of inflation. The inhibitory effect occurs when air flows into, but not our of, the lungs. It becomes gradually less potent with time as the lungs are held in the inflated state. In the experiments described here, the experimental animals were paralysed and ventilated on oxygen prior to the periods of stoppage of the pump during which observations were made. The levels of alveolar or arterial carbon dioxide tension were not monitored in these animals, and it is conceivable that they may have been elevated in order to maintain central inspiratory drive. The faiJure of maintained pulmonary inflation to block vagally mediated changes in heart rate is unlikely, however, to be due to this possible coincidence of raised oxygen and carbon dioxide tensions. Haymet & Mccloskey (1975) and Davidson, et al (1976) reported similar effects in animals spontaneously breathing air.

A brief report by Lopes & Palmer (197~) suggested that the bradycardia evoked by electrical stimulation of the carotid sinus nerve could be attenuated by central inspiratory drive activity or by sustained inflation of the lungs. These authors gave no indication of the magnitude of the pressures used to achieve sustained inflations, however, and it is likely (cf., Anrep, et al., 1936a) that these considerably exceeded the pressures used here (for example, their fig le shows a lung inflation that caused sufficient 100 obstruction of venous return to reduce arterial pressure by approximately 60 mm Hg within 10 seconds).

The identity of the intrapulmonary receptor type or types responsible for the inhibition of both cardiodepressor reflexes is not revealed by these experiments. If a single receptor type were responsible, it should possess the following properties to account for the findings: a sensitivity to the rate of lung inflation but no response to deflation, and a significant degree of adaptation during maintained inflation. The pulmonary stretch receptors come closest to meeting these requirements. They discharge in response to inflation but not deflation, are sensitive to the rate of lung inflation, and adapt slowly at a fixed lung volume (e.g., Paintal, 1973; Widdicombe, 1974); The property which is difficult to fit with the findings described here, is the adaptation to a maintained inflation. The degree of adaptation of pulmonary stretch receptors is not great. It can be seen, for example in figs 4.3, 4.4, 4.9 and 4.10 that maintained lung inflations sufficient to cause a Hering-Breuer apnoea (as judged from the phrenic electroneurogram) did not block the cardiodepressor reflexes. This was also the experience of

Haymet & Mccloskey (1975) and Davidson, et al (1976). Because the Hering-Breuer apnoea was maintained it can be concluded that appreciable activity of the pulmonary stretch receptors was continuing. However, the inhibition of the cardiodepressor reflexes did not continue. This observation suggests that the discharges of the pulmonary 101 stretch receptors are not, in an unmodified form, responsible for inhibiting the baroreceptor and chemoreceptor reflexes on heart rate. However, central neurones which relay the phasic but not the tonic elements of stretch receptor inputs, i.e., which differentiate, mathematically, the stretch receptor input with respect to time, could provide the basis for the inhibition of cardiodepressor reflexes. In this respect the R~ neurones (Baumgarten & Kanzow, 1958) found in the vicinity of the nucleus of the tractus solitarius, are of interest. These neurones discharge in phase with phrenic motoneurones and are stimulated also by lung inflation. Thus they are active in circumstances in which the mechanisms causing vagal bradycardia are refractory, and thus may be associated with this phenomenon. While many investigators report no marked adaptation of the response of Rfo neurones to sustained lung inflation, such adaptation is apparently frequently seen (Berger & Mitchell, 1977; K.M Spyer and R.-M McAllen, personal conununication) so that this aspect of their behaviour also correlates well with the conditions associated with vagal refractoriness.

Another intrapulmonary receptor type possibly concerned in the effects described is the 'fast-adapting' receptor

(Knowlton & Larrabee, 1946), the so-called 'lung-irritant' receptor. Lung-irritant receptors discharge in a roughly rate-sensitive way to lung inflations, particularly to large inflations and adapt rapidly so that they have little or no activity during maintained inflations (Paintal, 1973; 102

Widdicombe, 1974; Sampson & Vidruk, 1975). These properties could enable these receptors to be responsible for these findings. However, the lung-irritant receptors are also sensitive to deflations of the lung, and it has been seen in figs 4.9, 4.10 and 4.11, that both cardiodepressor reflexes can be readily evoked during lung deflations. It is true, however, that the lung-irritant receptors are not particularly sensitive to deflations from moderate expansions to functional residual capacity, as tested in some of the present experiments (figs 4.9 and 4.10)

(Widdicombe, 1974; Sampson & Vidruk, 1975). It might, therefore, be claimed that the lung irritant receptors could still account for the findings. Experiments of the type shown in fig 4.11 are therefore, most important. In these experiments, very large deflations (from large expansions to volumes close to residual volume) were tested. Such large deflations are known to excite lung irritant receptors powerfully (Widdicombe, 1974; Sampson

& Vidruk, 1975). They did not, however, inhibit the baroreceptor and chemoreceptor reflexes on heart rate. It therefore appears unlikely that lung irritant receptors are responsible for the effects described nere. It could still be argued that the lung irritant receptors are responsible, but only during inflations, because their effects during deflations are in some way modified by other information about the direction of lung movement. One cannot rule out this possibility altogether, but note that in experiments of the type shown in fig 4.11, the pulmonary stretch receptors of the non-moving lung were 103

signalling tothecentral nervous system that the lungs were inflated, and were doing so sufficiently strongly to prevent an inspiratory effort when the moving lung collapsed.

It is unlikely, therefore, that lung irritant receptors

are 'gated' by pulmonary stretch receptor activity in

such a way as to enable them to be responsible for the

inhibition of cardiodepressor reflexes described here. 104

CHAPTER 5

RESPIRATORY MODULATION OF OCULOCARDIAC AND NASOPHARYNGEAL REFLEXES 105

1. OCULOCARDIAC REFLEX

The application of pressure to the eyes was tested in nineteen anaesthetised dogs and the typical response was bradycardia. In some dogs it was considerable, in others it was slight, but in all there was an enhancement of respiratory sinus arrhythmia accompanying the bradycardia. In about two thirds of the dogs tested, there was also some reduction in the rate and depth of breathing during the application of pressure.

In order to test the effectiveness of the oculocardiac reflex during periods in which an animal was not breathing, Hering-Breuer inflation reflexes were evoked by occluding the trachea at the end of normal inspirations. When the lungs were thus held inflated the animal made no attempt to breathe for many seconds, during which time the effectiveness of the reflex could be tested. Typically, there was a slight increase in heart rate during a control trial of lung inflation alone. This w_as probably attributable to the slight fall in blood pressure which occurred during the period of positive intrathoracic pressure. When pressure was applied to the eyes during Hering-Breuer inflation apnoea, however, the bradycardia that was evoked was large, and was always larger than the bradycardia evoked by similar pressure applied during normal breathing (see fig 5.1). If both eyeball pressure and lung inflation were maintained for long enough, the 106

Pulse -.::r:.. Interval sec 20sec L_J

Tidal vol ml

Fig 5.1. Dog, anaesthetised with chloralose. Records show pulse interval and tidal volume (measured

by bag-in-box method, inspiration upwards).

Marker bars at bottom of figure show periods

during which digital pressure was applied to

the eyes. On two occasions the tidal volume

record shows inspiratory apnoea, achieved by

occluding the expiratory line. Inspiratory

apnoea or ocular pressure alone evoke little

change in heart rate. When ocular pressure

is applied during a period of inspiratory

apnoea, there is marked b~adycardia; note

that this bradycardia is transiently reduced

when the inspiratory apnoea is interrupted

by a breath. 107 animal would ultimately make an inspiratory effort. Whether such an inspiratory effort was permitted to draw air into the lungs (fig 5.1) or not (figs 5.2, 5.3), it was accompanied by a sudden and marked reduction in the bradycardia which then returned as the inspiratory effort concluded. Responses of this type are shown in fig 5.1. Similar responses were obtained in all dogs in which oculocardiac reflexes could be reliably evoked (fifteen of nineteen dogs).

Recordings from single preganglionic cardiac vagal fibres were made in some dogs, while digital pressure was applied to the eyes. Digital pressure to the eyes evoked a marked increase in discharge of the vagal fibre. This continued while pressure was maintained. During an inspiratory effort, vagal activity was markedly reduced, or abolished, despite continuing application of pressure to the eyes (see fig 5.2).

In seven dogs more elaborate experiments were performed in order to elucidate the respiratory mechanisms responsible for reducing oculocardiac bradycardia whenever an inspiration occurred (e.g., figs 5.1, 5.2). These animals were paralysed with D-tuboct,1rarine · (1 - 2 mg/kg) and art if ic:ially ventilated on oxygen with a Starling 'Ideal" pump. From time to time the pump was stopped and experiments on the oculocardiac reflex were performed. Respiratory activity in the paralysed animals was indicated by the integrated phrenic electroneurogram. When pressure 108

CVE impulses

digital pressure to eyes

I ' •1nsp I t 2s

Fig 5.2. Dog, anaesthetised with chloralose. Records of

activity in a single cardiac vagal efferent

fibre during application of pressure to the

eyes. Digital pressure to both eyes evokes

an increase in discharge of the vagal fibre.

This discharge is inhibited during inspiration,

and is most pronounced when the inspiratory

effort concludes. VAGI INTACT 109

Heart rate /mn

Phrenic dscharge

Intra-tracheal pressure nvn Hg l _ _,f7'-_

LUNGS DENERVATED

Heart rate /m'ln J flt,renic disc.harge t

Intra-tracheal pressure nm Hg :[ h LI BP nvn Hg

Fig 5.3. Dog, anaesthetised with chloralose, paralysed with D-tubocurarine. Records of heart rate, integrated phrenic neural activity, intratracheal pressure and arterial B.P, obtained during periods of temporary cessation of artificial ventilation, are shown. Ocular pressure was applied between the arrows. In the upper panel the response to ocular pressure is seen : there is a pronounced bradycardia which is reduced on each occasion that inspiratory (phrenic) activity occurs. The bradycardia is also reduced as the lungs are inflated, but returns while the inflation is maintained. In the lower panel the response to ocular pressure is seen after surgical denervation of the lungs; bradycardia is again evoked but reduced on each occasion that inspiratory activity occurs. The bradycardia is no longer reduced by lung inflation, nor does inflation inhibit inspiration as it did before pulmonary denervation. 110 was applied to the eyes in these animals, oculocardiac reflex bradycardia occurred. With each inspiratory effort indicated by the phrenic electroneurogram, the bradycardia was reduced, but it returned when the effort concluded. When the lungs were inflated quickly (within 1 - 2 seconds) to a pressure of 4 - 8 mm Hg by the experimenter blowing into a tube connected to the tracheal cannula, this again reduced the oculocardiac reflex bradycardia. The bradycardia returned within a few seconds of the air flow into the lung even when the lung inflation was maintained. Such inflations were associated with a silencing of the activity in the phrenic nerve, presumably because of the Hering­ Breuer inflation reflex. In all seven dogs the lungs were then surgically denervated by cutting the left vagus in the neck, and the right pulmonary vagus through a thoracic incision at the level of the fourth rib (see Methods). This procedure denervated the lungs, but maintained cardiac vagal innervation on the right side. Repeating the initial experiment after pulmonary denervation showed that oculocardiac reflex bradycardia still occurred and was still reduced with each inspiratory effort, but was no longer reduced· by lung inflation. A complete set of findings from a typcial experiment is shown in fig 5.3.

This oculocardiac bradycardia was mediated mainly by the vagus nerves. This was shown in two ways. First, by recording directly from single cardiac vagal fibres, it was shown that digital pressure to the eyes evoked a marked increase in vagal discharge (fig 5. 2). In addition, 111 in six dogs in which heart rate was recorded, the marked bradycardia evoked by digital pressure to the eyes was all but abolished following vagotomy or administration of atropine (1 - 2 mg/kg). In all six, however, slight bradycardia was still evoked by eyeball pressure, and this was attributable to withdrawal of sympathetic neural tone, as it was no longer evoked after administration of propranolol ( 1 mg/kg). In three of the six animals in which bradycardia due to sympathetic withdrawal could be demonstrated, there was a greater response elicited when eyeball pressure was applied during the expiratory phase of breathing than when it was applied during inspiration, This effect is shown in fig 5.4 and is similar to respiratory modulation of sympathetic responses to chemoreceptor and baroreceptor stimuli described earlier (see chapter 3, figs 3.3, 3.5).

2. NASOPHARYNGEAL STIMULATION

Experiments similar to those on the oculocardiac reflex described above were performed in ten dogs. The passage of.cold water through the nasopharynx was associated with a pronounced bradycardia which was reduced with each inspiratory effort, and enhanced in periods of Hering­ Breuer inflation apnoea.

Seven of the dogs were paralysed with D-tubocurarine, artificially ventilated on oxygen, and then tested with 112

Air flow insp up rnnm1 17711777711111 Heart -e e- T rate 200[ /min 150 BP mm Hg 1:f

20 sec

Fig 5.4. Dog, anaesthetised with chloralose; bilateral

vagotomy. Records show tracheal air flow

(inspiration upwards), heart rate and arterial

blood pressure. Marker bars below the air flow

record show periods in which pressure was

applied to both eyes. In the panel at the

left the bradycardia in response to sustained

ocular pressure is seen. In the panel at the

right the bradycardia evoked by ocular pressure

applied only during the expiratory phase of breathing e is compared with that evoked by similar pressure applied only during inspiration L 113

nasopharyngeal stimulation in periods during which the

pump was temporarily stopped. In these animals nasopharyngeal

stimulation evoked a prompt reflex bradycardia which was·

reduced on each occasion that the phrenic electroneurogram

indicated inspiratory activity. The bradycardia was also

reduced by the inflation of the lungs (to 4 - 8 mm Hg within

1 - 2 seconds), al though the bradycardia returned despite maintained static inflation of the lungs (see fig 5.5).

Inflation of the lungs in this way was associated with

silencing of the phrenic electroneurogram, presumably

because of a Hering-Breuer inflation apnoea. Denervation

of the lungs abolished the effects of lung inflation.

Recordings of cardiac vagal activity were not made during

nasopharyngeal stimulation, as they were with the

oculocardiac reflex, to support these findings. This was

because of mechanical and electrical interference

introduced by the stimulation procedure. These made such

recordings too difficult to achieve here.

3. DISCUSSION (CHAPTER 5)

(a) Oculocardiac reflex. The oculocardiac

reflex was first described by Aschner (1908) and Dagnini

(1908) and is evoked by pressure on the eyeball or traction

on the extrinsic muscles of the eyes. It has its afferent

limb in the trigeminal nerve and efferent limbs in the

vagus and (as shown here) sympathetic nerves. The reflex 114

Heart rate /mn Phrenic discharge t t ~tra-tracheal pressure 80( ____10_sec _ __. nm Hg h~----

BP , -~,~~1,\\\\\\(~\l\f/1~~, nm Hg 0

LUNGS OENERVATEO Heart rate /rnn Phrenic 3 dischage L/L/L Intra-tracheal pressure nm Hg J__ t10_sec __ ~L.JL_

BP 150[, nvn Hg 50

Fig 5.5. Dog, anaesthetised with chloralose, paralysed with D-tubocurarine. Records of heart rate, integrated phrenic neural activity, intratracheal pressure and arterial blood pressure, obtained during periods of temporary cessation of artificial ventilation, are shown. Nasopharyngeal stimulation was applied between the arrows. In the upper panel the response to nasopharyngeal stimulation is seen there is a pronounced bradycardia which is reduced on each occasion that inspiratory (phrenic) activity occurs. The bradycardia is also reduced as the lungs are inflated but returns while the inflation is maintained. In the lower panel the response to nasopharyngeal stimulation is seen after surgical denervation of the lungs; bradycardia is again evoked but reduced on each occasion that inspiratory activity occurs. The bradycardia is no longer reduced by lung inflation. 115 is of obscure significance and was used here because it provides a mechanism for evoking bradycardia which is not apparently dependent upon arterial baroreceptor or chemoreceptor reflexes. Aserinsky & de Bias (1963) reported that oculocardiac reflex bradycardia is suppressed by artificial ventilation, a finding which would be attributable to the phasic activity of intrapulmonary receptors, because denervation of the lungs abolishes the inhibitory effect of lung inflation.

The results presented in this chapter show in addition, that, as with baroreceptor and chemoreceptor reflexes described above in chapter 4, central inspiratory activity alone can also inhibit the reflex bradycardia evoked by application of pressure to the eyes.

{b) Diving reflex. The reflexes elicited by nasopharyngeal stimulation in the dog are similar to the diving response seen in natural diving animals and man

(Angell, James and Daly, 1972b). It is known that diving bradycardia in the duck (Andersen, 1963) is reduced when a breath is taken, even if the inspired gas does not alter the blood gas tensions. The experiments described here on dogs suggest that the reduction is brought about by both central inspiratory and phasic pulmonary afferent mechanisms. Studies by aamford and Jones (1976) in the duck also lead to this conclusion. 116

CHAPTER 6

RESPIRATORY MODULATION OF CARDIODEPRESSOR REFLEXES IN NORMAL HUMAN SUBJECTS 117

Some of the reflexes demonstrable in dogs and described in chapters 3, 4 and 5 were tested also in normal human subjects. It was noted in the Introduction in chapter 1 that Daly & Angell-James (1975) have shown that well established diving bradycardia in man can be interrupted when the diving subject takes a breath through a tube while under water,even when the gas mixture inspired cannot

'improve' existing blood gas levels. This fits well with the findings presented here. The cardiac vagal excitation occurring duringthedive, and responsible for the bradycardia described, would be expected on the grounds of the findings in chapters 3, 4 and 5 to be interrupted by central inspiratory activity and by lung inflation in the circumstances of Daly & Angell-James' experiment.

Unfortunately, their experiments could not distinguish between the central and peripheral accompaniments of inspiration in responsibility for the interruption of vagal discharge.

In the experiments now described, two si~ple manoeuvres were tested in an atte~pt to excite central inspiratory activity with minimal accompanying lung inflation. These manoeuvres were, first, the taking of what is termed here a 'false breath' and second, swallowing. A subject can take a 'false breath' simply by making a voluntary inspiratory effort against a closed glottis - this is presumed to activate central inspiratory mechanisms with minimal lung expansion. The second manoeuvre, swallowing, 118 is known to be associated with brief, intense bursts of activity in medullary 'inspiratory' neurones (Hukuhara & Okada, 1957; Sumi, 1963). Swallowing during normal, quiet breathing causes transient tachycardia (fig 6.1 : see discussion at the end of this chapter).

Other experiments described in this chapter concern the cardiovascular ~ccompaniments of breath holding in normoxic and in hypoxic conditions.

1. OCULOCARDIAC REFLEX

Oculocardiac reflex bradycardia was evoked on many occasions in each of eight normal volunteer subjects by application of pressure to the eyes. The bradycardia was easier to demonstrate and was more marked during breath holding. This finding is consistent with the observations in the dog, reported above (e.g., see fig 5.1). No consistent difference was found in the magnitude of the bradycardia evoked when breath-holding was performed in the inspiratory or expiratory position.

The oculocardiac reflex bradycardia was reduced whenever a subject took a breath, despite the continued application of pressure to the eyes. The bradycardia was also reduced if the subject made a 'false breath', or if the subject swallowed. Fig 6.2 shows results from a typical subject. 119

ecg 11111111111111111111111111111~11111111m1111111111m11111111111111111111~11111111w111111111m1111111111m1111111~~1111111111111t11111m111111mtm1

90 =/mn sor • • • • • •

20 sec

Fig 6.1. Effects of swallowing on heart rate.

Records show e.c.g and heart rate of

a normal resting human subject. The

subject swallows on six occasions,

marked by filled triangles. 120

Heart rate 80[ /min 40 ecg

10 sec Heart rate 80[ /min 40 ecg tttttttttt tt t Hitt t t ~ tt tt tt~~tH 4

Heart rate 80[ /min 40 ecg 'tttlttttl lttt ttt II tt t t~ ttt tt I ~f fltt

Fig 6. 2. Records show the heart rate and e .. c. g from

a normal human subject during three periods

of voluntary apnoea in which bilateral

ocular pressure was applied (marker bars).

The upper panel shows a control response.

The middle panel shows the effect of the

subject taking a 'false breath' (see text)

at the filled triangle. The lower panel

shows the effect of the subject swallowing

at the filled triangle. 121

2. DIVING RESPONSE

Eight normal volunteer subjects participated in many 'dives' in which they immersed the nose and mouth in a bowl of cold water. Each dive was associated with a reflex bradycardia, and no consistent differences--were found in the responses to dives in which the breath was held in inspiration and those in which the breath was held in expiration. Most subjects normally swallow at the moment of immersing the face in water and find it most difficult to prevent themselves from doing this (Ebbecke, 1943). Diving bradycardia crune on more slowly in dives which commenced with a swallow and was reduced whenever a subject took a 'false breath', or swallowed. Fig 6.3 shows results from one subject.

3. BREATH HOLDING IN NORMOXIC AND HYPOXIC CONDITIONS

Although oreath-hold diving reliably causes bradycardia in normal subjects, simply. holding the breath with the face in air during normal breathing causes little or no consistent bradycardia. This is shown for exrunple, in fig 6.4 (see also Schneider, 1930; Eckberg, 1976; Baskerville

Eckberg & Thompson, 1979). In the present series of experiments, subjects were given hypoxic (8 -12% 02 in N2 ) gas mixtures to breathe. Typically, breathing was stimulated and heart rate·rose. When, however, a hypoxic subject was asked to hold his breath, pronounced bradycardia 122

Heart rate 80[ /min 40 ecg

Heart 10 sec rate 80[ /min 40 ecg

Heart rate 80[ /min 40 ecg

Fig 6.3. Records show the heart rate and e.c.g from a

normal human subject during three periods of

voluntary apnoea in which th~ face and mouth

were immersed in water (marker bars). The

upper panel shows a control response. The

middle panel shows the effect of the subject

taking a 'false breath' at the filled triangle.

The lower panel shows the effect of the

subject swallowing at the filled triangle. 123

H. Rate /rrin 37 e.c.g. -~~,1~11~!~,!~llll~llll!!!~ll\l!!!lill1M11-.11~m1111~~1m~~!M~1,1~m1!IIMll~ll!ll!l~!ll11ffl-!IJPl~M~-,111m11111~~!!W.ll~-~•1111m11~11-

20 Resp. Rate /rrin

---AIR--+.------8%02 i, N2------+l•AIR•I

Fig 6.4. Records from a normal human subject showing

heart rate (beat-by-beat), electrocardiogram

(e.c.g), and respiratory rate. The left side

of the figure shows heart rate when the

subject holds his breath, having first

breathed room air. Heart rate shows little

change. The middle portion shows the effect

on heart rate when the subject breathes a

gas mixture of 8% o2 in N2 . Heart rate and respiratory rate gradually increase throughout

this period. The right side shows the effect

on heart rate when the subject, now hypoxic,

voluntarily holds his breath. There is now a

marked bradycardia, which is relieved as soon as breathing restarts. 124

~eveloped. All these phenomena are shown in fig 6.4.

Similar findings were reported by Gross, Whipp, Davidson,

Koyal & Wasserman (1976).

4. DISCUSSION (CHAPTER 6)

The reflexes elicited by diving in man and by nasopharyngeal stimulation in the dog are similar (Angell

James & Daly, 1972a). It is known that diving bradycardia in man (Daly & Angell James, 1975) and in the duck

(Andersen, 1963) is reduced when a breath is taken, even if the inspired gas does not alter the blood gas tensions.

It appears that the reduction is brought about by both central inspiratory and phasic pulmonary afferent mechanisms. Recent studies by Bamford & Jones (1976) in the duck lead also to this conclusion. In the episodes of diving and oculocardiac reflex bradycardia in which 'false breaths' were taken-(figs 6.2 and 6.3), both mechanisms could have been operating, although the pulmonary distortion was presumably much less than would have occurred during a normal breath.

Swallowing was a manoeuvre which was used in an attempt to activate central respiratory neurones without causing an accompanying distortion of the lungs. It is known that many medullary neurones which discharge in association with inspiration also discharge at the commencement of a

swallow (Hukuhara & Okada, 1956; Sumi, 1963). The same 125 neurones discharge when a stimulus known to induce swallowing is presented in a paralysed animal (Sumi, 1963). These inspiratory neuronal discharges may be associated with the very slight inspiratory effort ('schluckatmung') which is a characteristic early event in swallowing (Doty, 1968). It has also been shown that swallowing inhibits cardiac vagal efferent discharge (Okada, et al., 1961), so the coincidence of central inspiratory activity and vagal inhibition is again prominent.

The association of tachycardia with swallowing during quiet breathing was observed by Miller & Sherrington (1916) and has been demonstrated here (fig 6.1). In addition it was shown here that swallowing reduces oculocardiac or diving reflex bradycardia.

Attempts were also made here in many experiments on three domestic ducks, to look at the effects of swallowing on diving bradycardia. in these diving animals, but they could not be induced to swallow while diving.

The experiments on breath holding in normoxic and hypoxic conditions fit well with the mechanisms of vagal inhibition discussed to this point. The bradycardia that is so prominent during breath holding in hypoxia develops, presumably, because the inhibitory influences of inspiration - both central and peripheral - cease when breathing is voluntarily halted. Powerful chemoreceptor-cardiodepressor reflexes can then be observed, as in fig 6.4. These findings 126 parallel those of Daly & Scott (1958) in the anaesthetised dog - chemoreceptor stimulation by hypoxia evoked increases in breathing and tachycardia (like the mid-portion fig 6.4) except when ventilation was restrained, when bradycardia developed (as in the end-portion of fig 6.4).

An important consideration emerging from these simple experiments on conscious human subjects is that it is possible to demonstrate qualitatively similar mechanisms in man to those studied here in greater detail in anaesthetised dogs. 127

CHAPTER 7

ANALYSIS OF MECHANISMS OF INSPIRATORY INHIBITION OF VAGAL ACTIVITY 128

Earlier chapters have shown that lung inflation alone, when central inspiratory activity is absent, or central inspiratory activity alone, whan the lungs re~ain at resting volume, can inhibit the baroreceptor or chemoreceptor reflex. Other cardiodepressor reflexes, such as the diving reflex or the reflex responses to nasopharyngeal stimulation or pressure on the eyes, are similarly affected.

Because a single volley of vagal efferent action potentials can slow the heart for up to several seconds (Brown & Eccles, 1934), transient interruptions of vagal efferent discharge cannot be fully reflected in changes in heart rate. An interruption in vagal discharge associated with inspiration may leave heart rate relatively little affected, especially when that inspiratory event is brief. The observations reported up to this point on heart rate, therefore, while showing qualitatively the vagal inhibition described, cannot accurately reflect the timing and extent of changes in vagal discharge.

In this chapter the inhibitory effects of central inspiratory discharge and lung inflation on efferent vagal discharge are more fully documented, and particular attention is paid to differences between the types of inhibition they cause. Already it has been noted in the recordings of vagal efferent activity presented in previous chapters that while lung inflation and central 129

inspiratory activity are comparably effective in inhibiting

reflexly evoked increments in vagal firing, lung inflation

has markedly less effect upon tonic vagal discharge (e.g.,

see figs 4.1, 4.2, 4.5, 4.6). This matter is taken up

again here.

This chapter deals particularly with the effects of brief,

selective baroreceptor stimuli. This is because the time­

course of baroreceptor stimulation by brief intracarotid

pressure pulses is precisely known - it lasts for the

duration of the rise in pressure (Haymet & Mccloskey, 1975;

see also fig 7.4). Brief, selective chemoreceptor stimuli

(i.e., the injections of CO2 - saline already described)

were less suitable here because the precise time-course of

each individual stimulus is not known.

1. RESPIRATORY EFFECTS ON VAGAL TONE

The separate contributions to inhibition of

vagal tone from central inspiratory ac~ivity and from

lung inflation were conveniently studied in fourteen

paralysed dogs. Observations were made in periods during

which 1 artificial ventilation was temporarily halted:

throughout these periods (usually, 1 minute) the animals

remained well oxygenated as they had been ventilated on

pure oxygen before the pump was stopped. Central

inspiratory activity was reflected in phrenic neural

discharge. Inflation of the lungs could be achieved 130

independently of this activity by the experimenter simply blowing into the tracheal tube.

Lung inflation and central inspiratory activity did not

inhibit cardiac vagal motoneurones equally effectively.

When a moderate degree of resting vagal tone was present

each burst of central inspiratory activity completely

inhibited it, whereas inflation of the lungs with pressures within the physiological range (up to 15mmHg) had little

or no effect on it. Inflations just beyond these (to

approximately 20mmHg) frequently decreased tonic vagal

discharge without suppressing it entirely. These effects

are illustrated in fig 7.1.

When vagal tone was increased as, for example, by inflating

an intra-aortic balloon so as to raise arterial pressure

and stimulate arterial baroreceptors, each burst of central

inspiratory activity remained associated with complete, or

(less frequently) almost complete, inhibition of vagal

activity. At higher levels of vagal tone, however, lung

inflations became more effective inhibitors of it. This

was so for all levels of inflation.(see fig 7.2). The

effects of lung inflation were demonstrable in the absence

of secondary changes in arterial pressure. They were

abolished by cutting the left vagus nerve (see also chapter

4, and Gandevia, et al., 1978a - the right had previously

been cut prior to recording from it). In no animal,

however, did lung inflation completely abolish cardiac

vagal activity in the way that was typical of the action 131

Phrenic discharge ~ - CVE impulses

Intra-tracheal pressure 2:[ - ~ I\ mm Hg ' 5s

Fig 7.1. Dog, anaesthetised with chloralose, paralysed

with pancuronium. Records of integrated phrenic

nerve activity, cardiac vagal efferent impulses

recorded from a single fibre dissected from

the cervical vagus, and intratracheal pressure

are shown. Cardiac vagal efferent activity is

completely inhibited during activity in central

inspiratory centres (indicated by phrenic nerve

activity), while lungs remain uninflated. Lung

inflation to 10 mm Hg ( during the expiratory phase

of central inspiratory cycling) has little

effect on cardiac vagal efferent activity. Lung

inflation to 20 mm Hg has a slight inhibitory

effect on cardiac vagal efferent activity. 132

C.V.E. activity

Intra -tracheal pressure 20f10 mm Hg 0

C.V.E. activity

Intra-tracheal pressure 20f10 mm Hg 0

Fig 7.2. Dog, anaesthetised with chloralose, paralysed

with pancuronium. Records show few fibre

cardiac efferent activity in a filament dissected

from the cervical vagus and intratracheal

pressure, at two levels of vagal tone. When

vagal tone is low (top panels) lung inflation

of either 10 mm Hg or 20 mm Hg. given during the

expiratory phase of central respiratory cycling

have little or no inhibitory effect on cardiac

vagal efferent activity. When vagal tone is

high (bottom panels) lung inflations of both

10 mm Hg and 20 mm Hg inhibit vagal activity :

note vagal activity is not completely inhibited

even when the lungs are inflated to 20 mm Hg. 133 of the central inspiratory drive.

2. PROPERTIES OF THE VAGAL EXCITATORY PATHWAY

Pulses of pressure of 50 - 200 ms duration delivered into the carotid sinus selectively stimulate afferent baroreceptor fibres in the carotid sinus nerve for the duration of the pressure pulse (Haymet & Mccloskey,

1975). The cardiac vagal responses to such stimuli are bursts of activity which outlast the stimulus pulses (see figs 7.3, 7.4). This phenomenon was seen in all animals in the present study and formed the basis of the further experiments described below.

(a) Interactions with central inspiratory

activity. In paralysed dogs, during intervals when artificial ventilation was temporarily halted, intracarotid pressure pulses were delivered in varying temporal relationship to central inspiratory activity, monitored as phrenic neural discharge .. The essential features of the vagal responses seen in all animals are shown in fig 7.5.

When baroreceptor stimuli were delivered during periods of phrenic silence, typical bursts of vagal discharge were evoked, and decayed to control levels within a period that varied from fibre to fibre, between 1 and 3 seconds.

However, if phrenic discharge commenced soon after a vagal 134

CVE I ' . 20[10 ' 'I' impulses/s , . : j ; . . ' I , o _Jlfvf11r lf 11111,p__j/ rJ (Irr _J

Phrenic discharge C.Sinus BP 200[_! mm Hg so IL ____

5s

Fig 7.3. Dog, anaesthetised with chloralose, paralysed

with pancuronium, respiratory pump halted

temporarily. Records of cardiac vagal efferent

activity (counter reset every 1 sec), 'integrated'

phrenic nerve activity, and arterial blood

pressure measured in the carotid sinus, are

shown. Pulses of pressure delivered to the

carotid sinus during phrenic silence and

while the lungs are motionless evoke reflex

increases in vagal activity which only gradually

return to resting levels (in~ 3 seconds). 135

CY.E. activity

200 C.Sinus BP mm Hg 50

l ls

Fig 7.4. Dog, anaesthetised with chloralose. Records of

cardiac vagal efferent activity (few fibre

preparation, top panel) and arterial blood

pressure measured in the carotid sinus, are

shown. A pulse of pressure delivered to the

carotid sinus during the expiratory phase of

breathing, evokes a reflex increase in vagal

activity which only gradually returns to

resting levels (in-2 seconds). The pulse of

pressure to the carotid sinus elevates blood

pressure for only - 350 ms. CVE 136 impu1se/s

Phrenic disch.

C.Sinus 200[ BP mm Hg 50

CVE impulses/s

Phrenic disch.

C.Sinus BP mm Hg

Ss Fig 7.5. Dog, anaesthetised with chloralose, paralysed with pancuroniurn. Records of single cardiac vagal efferent activity (counter reset every 500ms), 'integrated' phrenic nerve activity, and arterial blood pressure measured in the carotid sinus, are shown. Panel 'a' shows a pulst of pressure delivered into the carotid sinus (baroreceptor stimulus) during phrenic silence and while the lungs are motionless : it evokes a reflex increase in vagal discharge which returns to resting level in 3 seconds. Panel 'b' shows a similar baro­ receptor stimulus given just before central inspiratory activity begins, and while the lungs remain uninflated : it evokes a reflex increase in vagal discharge but the 'tail' of this response is inhibited from the moment phrenic discharge starts. Panel 'c' shows a baroreceptor stimulus given during central inspiratory centre activity, while the lungs remain uninflated : it fails to evoke the immediate reflex increase in discharge seen in control conditions (panel 'a') but as soon as central inspiratory neural activity ceases, the 'tail' of reflex response is seen as an increase 137 response had begun all vagal activity was inhibited for the duration of that phrenic discharge (fig 7.5b). That is, if the baroreceptor stimulus was given just before the phrenic discharge began, the vagal responses started but then the typical decaying burst of vagal activity together with tonic vagal activity,were inhibited through­ out the period of phrenic discharge. If the baroreceptor stimulus was given instead during a period of phrenic discharge, little or no immediate vagal response was evoked. Nevertheless, whenever a stimulus was given late enough in the course of phrenic discharge for the usual duration of decaying vagal after-discharge to overlap into the period beyond the end of phrenic discharge, that increment of vagal response appeared as soon as phrenic discharge stopped (fig 7.5c). In every circumstance, therefore, the typical vagal response with its decaying after-discharge appeared in full except for that portion of it which coincided with central inspiratory activity. It was as if the stimuli always set in train similar central processes which were excitatory to cardiac vagal motoneurones, but that central inspiratory activity inhibited the ultimate vagal responses to these processes.

(b) Interactions with effects from lung inflation. In the same paralysed dogs, again during intervals when artificial ventilation was temporarily halted, intracarotid pressure pulses were delivered in varying temporal relationship to lung inflations with pressures of 5 - 15 mm Hg. These lung inflations were 138 achieved by the experimenter simply blowing into a tube connected to the trachea. All intracarotid pulses and lung inflations in this part of the study were given during periods of phrenic silence. The essential features of the responses seen in all animals are shown in fig 7.6.

The same prolonged bursts of cardiac vagal activity were observed as above. When baroreceptor stimuli were delivered immediately before (0. 5 - 1 seconds) commencement of lung inflations vagal responses started, but the typical decaying activity which usually followed was cut short at the moment the lungs inflated (fig 7.6b). Only the increment of discharge expected in response to the baroreceptor stimulus was inhibited - the pre-existing level of tonic discharge was little affected. This contrasts with the effects of central inspiratory activity which, when similarly related in time to a baroreceptor stimulus, inhibited both the expected vagal response and the pre-existing tonic discharge (fig 7.5b, see also fig

4.1).

If a baroreceptor stimulus was given instead during a brief lung inflation, little of no vagal response was evoked either immediately, or after the end of the period of inflation (fig 7.6c, see also fi~ 4.5). Tonic vagal activity was little affected through these manoeuvres. Again the effects of lung inflation were in contrast to those of central inspiratory activity. As described above, central inspiratory activity inhibited both vagal tone and 139 CVE 20[ C /~ impulses/s 10 I 0~ ,y~--j~~'L Phrenic discharge ____/\.._ ___ _ Intro - trocheol P 10[ mm Hg o ------200[ (1 control C.Sinus BP _____jl mm Hg 50 \.,_,...,w_.~-.. _,__...,..,

C d CVE impulses/s 1~[ Phrenic discharge /\ Intro·- tracheal p mm Hg 1g[ (\

C.Sinus BP 200[ mm Hg 50

1---1 2s Fig 7.6. Dog, anaesthetised with chloralose, paralysed with pancuronium, respiratory pump temporarily halted. Records of single cardiac vagal efferent activity (counter reset every 1 sec), 'integrated' phrenic nerve activity, (indicating activity in central inspiratory centres), intratracheal pressure, and arterial blood pressure measured in the carotid sinus, are shown. Panel 'a' shows the reflex increase in discharge evoked by a pulse of pressure to the carotid sinus (baroreceptor stimulus) given during phrenic silence and while the lungs are motionless: note the typically prolonged discharge. Panel 'b' shows the effect of a similar stimulus given during phrenic silence and just before lung inflation. The reflex increase in discharge is evoked, but the 'tail' of after-discharge is then inhibited although resting vagal tone remains relatively unaffected. Panel 'c' shows the response to a similar stimulus given during lung inflation while the phrenic nerve is silent. No reflex increase in vagal discharge is evoked and resting vagal discharge is relatively unaffected by lung inflation. A stimulus given just after the lungs are deflated (panel 'd') also fails to evoke a reflex increase in vagal discharge. 140 the increment in vagal firing expected in response to a stimulus, but both tonic discharge and the residual response emerged unaffected at the end of the central activity (fig 7.5c).

When baroreceptor stimuli were given soon after the end

(~l second) of brief periods of lung inflation, vagal responses were again not evoked, although vagal tone was little affected (fig 7.6d). The responses were similar, therefore, to those seen when stimuli were applied during a brief period of inflation (fig 7.6c).

3. DISCUSSION (CHAPTER 7)

Striking differences between the central and intrapulmonary afferent sources of inhibition have been documented here. Tonic vagal discharge is inhibited most powerfully and consistently by central inspiratory activity. Lung inflation has relatively little effect, except when vagal tone.is high. This may explain why

Jewett (1964) and McAllen & Spyer (1978a) found little alteration in vagal discharge in response to lung inflation while others, studying vagal tone through its effects on heart rate, attributed quite powerful effects to lung inflation (e.g., Anrep, et al., 1936b; Aserinsky & de Bias,

1963; Daly, 1972; Gandevia, et al., 1978a,b). In general those studying heart rate seem to have been working with a background of strong vagal tone, and with larger inflations. 141

An important analytical device in the present study was the use of the typically prolonged vagal responses to brief stimuli. These prolongations of respo_nse a:v.e mainly due to the properties of the central relays between baroreceptor inputs and vagal motoneurones: they cannot be attributed solely to a temporal spread of arrival in the central nervous system of inputs along afferent fibres with different conduction velocities. Recording in the nucleus of the solitary tract from cells excited by electrical stimuli applied to the carotid sinus nerve, Trzebski, et al (1975) found the spread of latencies to be from 3 to 21 ms. Even allowing a conduction distance of 5 cm from carotid sinus to medulla in the dog, a spread of only

-50 ms could be accounted for between afferents conducting at 1 m/s and 100 m/s, but the cardiac vagal response outlasts the stimulus which evokes it by 500 ms or more.

Intracarotid pressure pulses were chosen here as it is known that these selectively activate arterial baroreceptors for the duration of the rise in pressure (Haymet & Mccloskey, 1975). Selective, briefly-acting stimuli for arterial chemoreceptors have also been described (e.g., Black &

Torance, 1971; Haymet & Mccloskey, 1975) but these were less suitable for the present experiments because, while they have a rapid onset, the precise duration of an individual stimulus cannot be determined. The fast onset of vagal firing accompanying a rise in intracarotid pressure but slow decline after a fall, was noted previously by Katona, Poitras, Barnett & Terry (1970). 142

Humphrey (1967), Biscoe & Sampson (1970b)and Trzebski, et al (1975) made intramedullary recordings of responses to carotid sinus nerve stimulation and to pulses of intra­ carotid pressure, and described cells which discharged promptly and maintained their activity beyond the duration of the stimuli. Such cells may be involved in the effects described here.

In attempting to account for the differences between the inhibitory effects of central inspiratory activity and of lung inflation the following findings must be accounted for. Central inspiratory activity inhibits both tonic and specifically-evoked vagal discharge, but leaves unaffected the residua of 'expected' responses to stimuli delivered while itis in progress. Lung inflations more effectively inhibit vagal responses to brief stimuli than tonic discharge, and this inhibition extends to involve also the residua of 'expected' responses to stimuli given during a period of inflation;

All vagal activity is inhibited for the duration of a central inspiratory burst, although the prolonged central excitatory state induced by a brief stimulus seems unaffected except in its ability to evoke a vagal response.

It is as if the central inspiratory drive inhibits only the outflow of vagal impulses from the CNS, without interfering with the processes which normally evoke those impulses. Such an action would be consistent with the central inspiratory drive acting towards the end of the 143 baroreceptor-vagal pathway and, indeed, there is no reason from the present study to suppose that it is not acting on the vagal motoneurones themselves, as suggested by McAllen

& Spyer (1978b). Additional effects at other points in the baroreceptor-vagal pathway cannot be entirely excluded.

The inhibitory effects of receptors excited by lung inflation are more complicated. Lung inflation seems to have its main actions on phasic inputs, leaving responses to sustained inputs relatively unaffected. Also, it appears to have its inhibitory effect earlier in the vagal excitatory pathway (although not directly on the baroreceptor afferent terminals; Jordan & Spyer, 1979) and perhaps at more than one site. A simple model would be that lung inflations inhibit the access of brief baroreceptor stimuli to those parts of the central pathways which are responsible for 'prolonging' a brief input,

while central inspiratory activity inhibits vagal motoneurones directly. This could explain why vagal responses to brief stimuli are blocked by simultaneous

lung inflations, while vagal tone remains_ relatively unaffected. A diagram of this simple model is shown in

fig 7.7.

However, a lung inflation which occurs soon after a brief stimulus has established a brisk vagal response, inhibits

the 'tail' of vagal after-discharge usually evoked by

such a stimulus (fig 7.6c). In these circumstances, there

can be no doubt that the stimulus has gained access to the 144

Insp.

PSR X

2 Bar~recepto r input V

Fig 7.7. Diagrammatic representation of proposed sites of inhibition by pulmonary stretch receptors (PSR) and by central inspiratory neurones (Insp) Inhibition is represented by filled circles. The simple baroreceptor afferent to vagal efferent pathway includes reverberating circuit to represent a mechanism by which vagal tone is maintained: other mechanisms could achieve the same result. Pulmonary stretch receptors are shown to inhibit baroreceptor input early in the reflex pathway, i.e., before tone-sustaining mechanism, while inspiratory centres inhibit vagal efferent activity relatively late in the pathway, i.e., beyond the tone-sustaining mechanism. This model shows how pulmonary stretch receptors can inhibit reflex inputs leaving resting vagal activity unaffected, while central inspiratory activity can inhibit both reflex inputs and resting vagal activity. 145 central, 'prolonging' processes which usually excite vagal responses: nevertheless, the outflow from those processes through the vagus is inhibited. Clearly in this respect the situation is more complex than the simple model above would suggest.

Figure 7.8 is a model of the baroreceptor-vagal pathway based on the results of this study. The model shows the central inspiratory drive inhibiting vagal motoneurones directly: although the present results do not indicate that this is the precise location of this action they do indicate, as discussed above, that the effect is exerted late in the reflex pathway. The central reflex pathway is shown as comprising two distinct elements, which might be thought of as neuronal pools. One element ('1' in fig 7.8), whose inputs and outputs are unaffected by afferent discharges set up during lung inflations, only weakly and slowly transmits brief, intense baroreceptor stimuli. This element transmits sustained afferent inputs, and its output only slowly falls when those inputs stop. This first element can be thought of as generating and sustaining vagal tone, and its properties might derive from those of its individual synapses, or possibly from reverberating circuits within it. The second element ('2' in fig 7.8), whose inputs and outputs are inhibited by afferent discharges from intrapulmonary receptors, transmits brief baroreceptor stimuli strongly and rapidly. Also, its rapid response decays slowly on removal of a baroreceptor stimulus, so that this element can be regarded as responsible for the 146

Central lnsp Drive

11\ lntra-pulm receptors signalling rate of lung inflation

Baroreceptor Vagal output (? and other) to heart inputs

Fig 7.8. More complex model to account for findings

presented in this chapter. For discussion,

see text. 147 typical declining after-discharge of vagal units following a brief baroreceptor input. Again, these properties would reflect behaviour of synapses or circuits within the element. Because lung inflation exerts its inhibitory effects on this element of rapid transmission, the preservation of tonic vagal discharge during inhibition of stimulus-evoked discharge can be explained. The present study has shown, however, that lung inflation can inhibit tonic vagal discharge to some extent, and particularly when vagal tone is high. This property can be accounted for if the '·tone­ sustaining' element receives an input from the more rapidly transmitting element - such an input is shown as 'b' in fig 7.8. (Indeed, both inputs, 'a' and 'b' to the tone generating element need not be proposed, 'b' alone would suffice).

It was noted that the vagal responses to baroreceptor stimuli delivered immediately after the conclusion of a brief lung inflation were inhibited in the same way as stimuli given during a lung inflation (see fig 7.6d). However, as the state of lung inflation was gauged here from records of intra-tracheal pressure, it cannot be concluded with confidence that lung volume had returned to its control value as soon as intratracheal pressure had returned to zero. Thus, the inhibition of vagal response may have been caused by continuing discharges in intra-pulmonary volume receptors. Alternatively, the inhibitory inpµts from intra-pulmonary receptors may themselves be prolonged in their transmission through central pathways, so that 148 the inhibition they cause outlasts the duration of the sensory input. If this were so, an appropriately prolonged inhibitory input acting only at the output (and not also at the. input) of element '2 1 in the model described above, could account for the findings described here. 149

CHAPTER 8

COMPARISON OF VAGAL RESPONSES TO ELECTRICAL STIMULATION OF THE CAROTID SINUS NERVE AND TO BRIEF BARORECEPTOR STIMULI 150

Electrical stimulation of the carotid sinus nerve is a convenient method of engaging baroreceptor and chemoreceptor afferents. In this chapter, the effects of such stimulation on cardiac vagal motoneurones are described, then compared with the effects of brief, selective baroreceptor stimuli.

1. RESPONSES EVOKED BY SINGLE ELECTRICAL STIMULI TO THE CAROTID SINUS NERVE

(a) Variability of response latencies. Electrical stimulation of the central end of the carotid sinus nerve has been used by many workers (for example, Iriuchijima &

Kumada, 1963, 1964; Humphrey, 1967; Biscoe & Sampson,

1970 a, b; Kunze, 1972; McAllen & Spyer, 1978 a, b) to identify central baroreceptor or chemoreceptor pathways or the efferent limb of these pathways, such as preganglionic cardiac vagal fibres. An action potential is evoked in a cardiac vagal efferent fibre in response to stimulation of the carotid sinus nerve and this has been used as a way of identifying cardiac vagal efferent fibres in the cervical vagus (see Iriuchijima & Kumada, 1963). The latency of this response has been reported to vary from fibre to fibre, from 26-120ms (Iriuchijima & Kumada, 1964; Kunze, 1972). Even for a single vagal fibre, the response evoked by electrical stimulation of the carotid sinus nerve varies by up to 20 ms : this was noted by Iriuchij ima & Kumada (1964) and Kunze (1972) but they did not attempt to account 151

for it.

Latencies of cardiac vagal responses (recorded at mid­ cervical level - see Methods, chapter 2) to electrical stimulation of the carotid sinus nerve were recorded here

in forty-two single efferent fibres in thirty dogs. Mean response latencies lay between 10 and 130ms. A histogram showing the distribution of these latencies for forty-one of the vagal fibres is shown in fig 8.1. The fibre with the shortest response latency, lOms, is not included in the histogram as it was in other ways atypical (see below).

In this relatively small sample of cardiac vagal fibres there was no evidence of the two distinct groups of

latencies observed by McAllen & Spyer 0.978b) in their recordings from the cell bodies of cardiac vagal motoneurones: however, as they noted, such grouping may be less evident when recording from efferent axons in the neck rather than

from motoneurones in the medulla, as the extra conduction

distance involved in the former may cause blurring of any

separation.

(b) Respiratory modulation of variability of

response latencies. In all fibres tested,

latencies of carotid sinus nerve stimulation were shorter

when stimuli were given only in the expiratory phase of

breathing (that is, during expiratory air flow or the

expiratory pause: see fig 8.2 top panel, and fig 8.3).

When stimuli were given only during inspiration usually

no response was evoked. This confirms previous work 152

16 Mean latency of responses in cardiac vagal efferent 14 fibres after stimuli to carotid sinus nerve 12

10 Number of 8 fibres 6

4

2

50 60 70 80 90 100 110 120 130 Latency (ms)

Fig 8.1. Mean latency of responses in single

cardiac vagal efferent fibres after

electrical stimuli to the carotid

sinus nerve. 153

Exp'n

lnsp'n

0 50 100

Latency ms

Fig 8.2. Dog, anaesthetised with chloralose. Records of a single vagal efferent fibre dissected from the cervical vagus in response to single electrical stimuli to the carotid sinus nerve. Both panels show several sweeps of the oscilloscope super­ imposed. The top panel shows the latency of the response in a vagal fibre when the stimuli were delivered only in the expiratory phase of the respiratory cycle (16 sweeps). The bottom panel shows the latency of the response in the vagal fibre when stimuli were given only during the inspiratory phase of the respiratory cycle. Responses are not usually evoked during inspiration: this bottom panel has 72 sweeps of the oscilloscope superimposed. Vagal responses evoked by stimuli given during inspiration are usually at a longer latency than those to stimuli given during the expiratory phase. Note there is still some variation in the latency of the response in both expiration and inspiration. lnsp'n Exp'n 154

Tracheal air flow

100 , . . ·. Latency CSN?Vagus ms 50

0 5

Time s

Fig 8.3. Dog, anaesthetised with chloralose. Records obtained by triggering the sweep of a storage oscilloscope from the commencement of inspiratory pressure change in the trachea. 120 sweeps are superimposed from-10 minutes of recording. Throughout the recording period electrical stimuli applied to the carotid sinus nerve were triggered from the e.c.g. Each stimulus reset a ramp generator which provided vertical deflection as an input to a channel of the oscillo­ scope: this trace was brightened with each occurrence of a vagal action potential. Thus, a display of spots was obtained in which single vagal action potentials could be seen throughout many successive respiratory cycles: where the stimuli to the carotid sinus nerve regularly evoked vagal responses, an aggregation of spots is positioned according to the latency of those responses. The figure shows a silence of spontaneous vagal activity in inspiration, a longer latency of vagal responses to stimuli given during inspiration and fewer responses to stimuli given during inspiration than to those given in expiration. Note that some variability of response latency remains at any point in the respiratory cycle for these stimuli given at a constant point in the cardiac cycle. 155

(Koepchen, Wagner & Lux, 1961; Iriuchijima & Kumada, 1964;

Gandevia, McCloskey & Potter, 1978a; previous chapters here).

When an occasional response was evoked during inspiration it was always at a longer latency than responses evoked during the expiratory phase (see fig 8.2 bottom panel, and fig 8.3). Most fibres studied showed a variation in latency through the respiratory cycle of about 10 ms (figs

8.2 and 8.3). This agrees well with the variation of response latencies to continuous stimulation of the carotid sinus nerve reported but not accounted for by

Iriuchijima & Kumada (1964) and Kunze (1972). Three fibres showed a variation between inspiration and expiration of only 3 - 4 ms.

Even when stimuli were delivered to the carotid sinus nerve in only one phase of the respiratory cycle, considerable variability in the latency of vagal response remained. This is clear, for example, in fig 8.2. It seemed possible that the cardiac cycle, possibly acting through arterial baroreceptor inputs, might impose a further cyclical change in the excitability of the reflex pathways involved here, and so of the latencies of

responses. To test this possibility, stimuli to the carotid sinus nerve were triggered from the e.c.g.

When response latencies to such stimuli delivered at

corresponding points in the respiratory cycle were

compared, both cardiac.and respiratory influences could be

kept constant. In all five fibres tested in this way,

elimination of the possibility of a cyclical cardiac 156 influence did not further reduce the variability of latencies (see fig 8.3).

One vagal fibre, mentioned above, responded to stimulation of the carotid sinus nerve with a mean latency of only lOms. While the ability to excite this nerve fibre through electrical stimulation of the carotid sinus nerve satisfied one of the criteria adopted for identification of vagal efferent fibres as cardiac (see Methods), its other properties raised doubts on this matter. The fibre had no tonic discharge, and could not be made to discharge on raising arterial blood pressure by inflation of an intra-aortic balloon. Of interest, however, was that the latency of its response to carotid sinus nerve stimulation varied from 9 ms for stimuli given during expiration to llms for stimuli given during inspiration (see fig 8.4).

In this respect it was like the cardiac vagal efferents described above, although it was unlike them in that all inspiratory stimuli evoked responses.

(c) Inhibition of vagal discharge following

responses elicited by single electrical stimuli.

The discharge patterns of twenty-six of the single cardiac vagal fibres described above were studied in the periods between stimuli at , lHz to the carotid sinus nerve. In this part of the study post-stimulus histograms were constructed for the distribution of vagal activity in the periods after each stimulus. These revealed, for all

fibres, a period of inhibition of vagal discharge 157

Exp'n l

lnsp'n

0 10 20

Latency ms

Fig 8.4. Dog, anaesthetised with chloralose. Records of a

single vagal efferent dissected from the cervical

vagus. Responses to single electrical stimuli

to the carotid sinus nerve are shown. The top

panel shows the latencies of the responses when

the stimuli were given only during the

expiratory phase of the respiratory cycle. The

bottom panel shows the latencies of the

responses when the stimuli were given only

during the inspiratory phase of the respiratory

cycle. Six sweeps of the oscilloscope are

superimposed for both traces. Stimuli given

during inspiration always evoked a response -

although at a longer latency. See text for

discussion. 158 commencing after the first evoked action potential and lasting for 100 - 150 ms. In twenty-three of the fibres this inhibition was complete, and no action potentials were recorded in this period (see fig 8.5). In the remaining fibres vagal spikes occured in this period, but less frequently than normal (fig 8.6). In several vagal units it appeared that the period of inhibition was followed by post-inhibitory facilitation (e.g., fig 8.6).

It seemed possible that the 'inhibition' of tonic vagal discharge following an electrically-evoked spike might simply be a natural pause in firing imposed by a particular fibre's effective refractory period. That is, the observed fibre might have been capable of a discharge frequency of no greater than, say, 10 Hz, so that having been made to fire once by electrical stimulation it could not fire again for another 100 ms. This was not so.

Figure 8.7, for example, shows the distribution of vagal firing following electrical stimulation of the sinus nerve

(and the post-excitatory inhibition is marked) and, in the same fibre, following naturally occurring spikes (and the natural refractory period is relatively brief). It was commonly found that cardiac vagal motoneurones were capable of natural discharge frequencies of 25 - 100 Hz - that is, with effective refractory periods of no more than 10 - 40 ms. These figures are in agreement with direct measurements of refractory periods made by McAllen & Spyer (1976).

As noted in the Methodssection above, the carotid sinus 159

CVE impulses

16 Impulses per bin 0 .1' Stirn 200ms

Fig 8.5. Dog, anaesthetised with chloralose. Records of

responses in a single cardiac vagal efferent

fibre to carotid sinus nerve stimulation.

Bottom trace shows a histogram accumulated from

superimposed traces of vagal firing. The

response to carotid sinus nerve stimulation

is evoked at a latency of about 70 ms.

Following this excitatory response there is a

period of complete inhibition of '"V 100 ms.

This can be seen in the top trace showing no

spikes in this period, and is seen also in the

histogram. 160 6l t Impulses per bin 6l 6l

Stirn

50ms intervals

Fig 8.6. Dogs, anaesthetised with chloralose. Histograms

accumulated from activity in three single cardiac

vagal fibres from three dogs. Responses to

carotid sinus nerve stimulation (marked 'stim')

are evoked at a latency of about 50 ms for

each fibre. This is £ollowed by an inhibition

(incomplete) lasting about l00ms. In the

bottom two panels, the inhibition seems to have

been followed by a post-inhibitory facilitation. 161

C.Sinus nerve stim Natural activity

CVE impulses

Impulses 32[ per bin 0

L..-..-1 L....--..1 lOOms 40ms

Fig 8.7. Dog, anaesthetised with chloralose. Records of activity in a single vagal efferent fibre in response to carotid sinus nerve stimulation (left panel: stimuli at marker) and natural firing frequency (right panel). Bottom trace in each panel shows a histogram accumulated from super­ imposed traces of vagal firing. The left panel shows the evoked response of a single vagal efferent fibre in response to an electrical stimulus to the carotid sinus nerve. A response is evoked in the cardiac vagal efferent fibre at a latency of about 50 ms. Following this excitatory response there is a period of inhibitbn. This is a true inhibition and is not due to the refractory state of the fibre after the excitatory response. This is shown by the records in the right panel of the behaviour of the same fibre during its natural discharge. In the right panel the oscilloscope sweep and histogram were triggered by every seventh naturally-occurring vagal spike. After a naturally occurring action potential the vagal fibre can discharge again as soon as l0ms, but after an action potential evoked by stimulation of the carotid sinus nerve it does not discharge again for more than 100 ms. 162 nerve was usually left intact for electrical stimulation. This preserved the natural baroreceptor and chemoreceptor inputs through the nerve and helped to maintain cardiac vagal tone. It is known, however, that stimulation of fsensory nerves can impair their normal receptor function (e.g., Goodman, 1973). It seemed possible, therefore, that shocks applied to the sinus nerve might briefly halt natural baroreceptor or chemoreceptor inputs, thereby removing that source of cardiac vagal tone, and so cause an apparent inhibition of the vagus following each shock. This was not the cause of the post-excitatory inhibition described above. When the carotid sinus nerve was crushed between the sinus and the stimulating electrodes so that it could no longer transmit naturally evoked baroreceptor or chemoreceptor impulses, post-excitatory inhibition of vagal fibres remained_prominent. Indeed, in the conditions of lowered vagal tone prevailing after this procedure, the inhibitory period often seemed prolonged (see fig 8.8).

It seemed possible also that stimulus spread from carotid sinus nerve directly to other nerves in the vicinity, even to the vagus itself, may have been responsible for the post-excitatory vagal inhibition. To see whether such stimulus spread was involved, the central end of the carotid sinus nerve was crushed in four further dogs after first obtaining the response described above, that is, excitatory response followed by inhibition. When this was done and the carotid sinus nerve was again stimulated, no excitation or inhibition was seen. This is shown in fig 8.9. 163

C. S. Nerve intact 32

0 Impulses per bin C.S.Nerve crushed 32

0 lOOms

Fig 8.8. Dog, anaesthetised with chloralose. Histograms accumulated from activity in a single cardiac vagal efferent fibre. Top panel shows excitatory response evoked by carotid sinus nerve stimulation at a latency of 65 ms followed by inhibition of -100 ms. In the bottom panel the carotid sinus nerve has been crushed between the carotid sinus and the stimulating electrodes. The response to carotid sinus nerve stimulation is evoked at the same latency as in the top panel (-65 ms). Now the post-excitatory inhibition is possibly more marked. This is presumably because tonic baro­ receptor inputs have been removed and resting vagal discharge is decreased. For this reason it is now more difficult to see tonic discharge, and so more sweeps of the oscilloscope have been superimposed in the bottom trace. 164 C.S.Nerve intact lOV 11J s tbsll • C.S.Nerve crushed Impulses per bin 11 • 11

• lOOms

Fig 8.9. Dog, anaesthetised with chloralose. Histograms accumulated from activity in a single cardiac vagal efferent fibre. Top panel shows excitatory response evoked by carotid sinus nerve stimulation (10 V) at a latency of about 55 ms, followed by complete inhibition of about 70 ms. The bottom two panels show effects of carotid sinus nerve stimulation after the carotid sinus nerve had been crushed between the stimulating electrodes and the central nervous system. Now, no excitatory responses were evoked and no inhibition was seen - only spontaneous activity in the vagal fibre continued. The middle panel shows this result when the carotid sinus nerve was stimulated at 10 V. In the bottom panel, the carotid sinus nerve is stimulated at 40 V. These recordings indicate that the excitatory response and the inhibitory pause were not due to stimulus spread, but to an effect of carotid sinus nerve stimulation. 165

2. RESPONSES EVOKED BY MULTIPLE ELECTRICAL STIMULI TO THE CAROTID SINUS NERVE

(a) Vagal respb~ses to pairs of electrical

stimuli. The inhibitory effects of electrical stimulation of the carotid sinus nerve on vagal discharge were shown also by the failure of the second of a pair of such stimuli to evoke the usual vagal response. Typically, a second vagal response was never evoked if the second stimulus of a pair fell within 40 - 80 ms of the first. A reduced frequency of occurrence of the second response occurred when the second stimulus was delayed up to a further 50 - 200 ms. Only for delays between pairs of stimuli exceeding this total duration (i.e., 100-280ms) were equal responses evoked by both stimuli.

This phenomenon is illustrated in the records shown in

fig 8.10.

(b) Vagal responses to trains of electrical

stimuli. For trains of electrical stimuli

applied to the carotid sinus nerve, the inhibitory effects

of successive stimuli on cardiac vagal responses became

less marked. This was so for stimulus frequencies below

a critical level, which varied from fibre to fibre from 30 to

100 Hz (see below). Trains of 0.1 - 0.6 seconds duration were

used. As described above, the second stimulus was less

effective than the first: typically, however, later stimuli

became successively more effective, and ultimately evoked 166 a 3:[ ... Ja., .J...J.1,, .... I .... 51•

b 3j I ...... AA ~ S1S2

Impulses C 3:r per bin .... h J. I I I I••• .... A A S1 S2

d 3:r I I II ll I .l.i 1a1 1.111 I A A S1 S2

e 3:r •• JJ. I I I I...... • I A A a S1 S2

lOOms Fig 8.10. Dog, anaesthetised with chloralose. Histograms accumulated from activity in a vagal filament containing two active cardiac fibres in response to electrical stimuli given to the carotid sinus nerve. Panel 'a' shows the response (at l'V72 and 97 ms) evoked by single stimuli (S1) delivered to the carotid sinus nerve at lHz. Panels 'b' to 'e' show res­ ponses to pairs of identical stimuli at various intervals apart. In each case the responses to the first stimulus of the pair (S1) were as for the single stimuli given alone. When the second stimulus of the pair (S2) fell within approx 75ms of the first, it evoked no further response (e.g., panel 'b'). When the second stimulus was given more than 75 ms after the first, responses were evoked, but less frequently than with a single stimulus (panels 'c' and 'd' ""100 and 132 ms, respec­ tively). For the fibres shown, an interval of 150 - 200 ms was required between stimuli for the responses to each to be similar (e.g., panel 'e'). 167 vagal responses as effectively as the first stimulus.

Depression of tonic discharge succeeded each individual vagal response, including the response to the last stimulus in the train. At the conclusion of the stimulus train, vagal discharge rose, then remained above control levels for 1 - 3 seconds. These phenomena are all illustrated in the records shown in fig 8.11.

At higher frequencies of stimulation above the critical level, which varied from 30 to 100 Hz depending on the individual vagal motoneurone studied - the responses departed from the descript,ion just given. The initial stimulus of the high-frequency train evoked the usual vagal response, but responses to later stimuli in the train were infrequent or absent. An example is shown in fig 8.11. At the conclusion of such high frequency trains of stimuli vagal discharge was elevated above control

levels for 1 - 3 seconds, as with the lower frequency trains.

That is, with high frequency bursts of stimulation of the carotid sinus nerve cardiac vagal motoneurones were stimulated, but the increased activity evoked occurred mainly after the train of stimuli.

3. RESPONSES TO BRIEF BARORECEPTOR STIMULI

(a) Single stimuli. Brief pressure pulses (50-

l00ms) were produced within the carotid sinus by rapid

retrograde injections of 2 - 5 ml of saline through the 168

CVE impulses 32 15Hz Impulses [ per bin 0

L.....J L-.J 200ms 200ms

Fig 8.11. Dog, anaesthetised with chloralose. Records of cardiac vagal efferent activity recorded in a single vagal efferent fibre dissected from the cervical vagus nerve (top trace), in response to stimulation of the carotid sinus nerve at two frequencies. Bottom trace shows histograms accumulated from many such cardiac vagal efferent responses. Left hand panel shows vagal response to stimulation of the carotid sinus nerve at 15 Hz for 300 ms. Note that the second stimulus of the train fails to evoke a response in the vagal fibre. Note also the final evoked response (to the sixth stimulus) is followed by an inhibitory period. Following this inhibitory period there is a further excitatory response before the vagal activity returns to resting levels. The right hand panel shows vagal responses to stimulation of the carotid sinus nerve at 100 Hz for 400 ms. Only the first and the last stimulus now consistently evoke a response in the vagal fibre. The response evoked by the last stimulus is also followed by an inhibitory period and then by a further excitation before the fibre returns to a resting level of activity. 169 cannula in the external carotid artery, after first clamping the connnon carotid artery below the carotid sinus. Such pulses of pressure stimulate intensely and selectively the baroreceptor fibres within the carotid sinus nerve for the duration of the elevation of pressure (Haymet & Mccloskey,

1975; and see fig 8.13, panel B).

Responses to these brief baroreceptor stimuli were studied in twelve cardiac vagal efferent fibres in this part of the work. Typically, the intra-carotid pressure pulses evoked increases in vagal discharge lasting 1 - 3 seconds - that is, well beyond the duration of the baroreceptor volleys. This phenomenon was analysed in detail in the preceeding chapter. The particular feature that is of interest here was the absence of any period of vagal inhibition following the evoked response. This was clear from post-stimulus histograms of vagal firing triggered from the intra-carotid pulse and continued for 1 - 5 seconds thereafter (bin-widths 3. 9 ms - 19. 5 ms). One such histogram is shown in fig 8 .12.

(b) Pairs of baroreceptor stimuli. Vagal responses evoked by two similar pressure pulses delivered into the carotid sinus within 1 - 4 seconds of each other were smaller for the second of the stimuli: this was so for all eight vagal motoneurones tested in this way. Each stimulus of the pair evoked increases in vagal discharge lasting 1 - 3 seconds, without evidence of any inhibitory phase in the response. In addition, simultaneous recordings of heart rate showed that the heart rate did not fall to 170

CVE impulses

C.Sinus BP 250[ ~ mm Hg 75 ..______16 Impulses per bin 0

L,__J 500ms

Fig 8.12. Dog, anaesthetised with chloralose. Records show activity of a single cardiac vagal efferent fibre, arterial blood pressure in the carotid sinus and a histogram of the activity of the vagal fibre. The records were obtained by triggering the oscilloscope trace and commencing a histogram sweep when a pulse of pressure was applied within the carotid sinus. The top trace shows the vagal responses recorded in response to such a stimulus in a single sweep of the oscilloscope. The middle trace shows superimposed records.of pressure pulses within the carotid sinus. The bottom trace shows a histogram of unitary vagal responses following these repeated pulses: that is, the histogram (bin width, 19. 5 ms) is accumulated from many sweeps of which the upper trace is one example. The vagal excitation evoked by a brief intracarotid pressure pulse outlasts the duration of that pulse. No vagal inhibition is seen following the excitatory response. 171

as low a level after the second stimulus as after the first.

This occurred despite the heart rate being usually slower

than its control level (as a result of the first stimulus)

at the time of delivery of the second. Figure 8.13

illustrates this phenomenon for one of the vagal units.

A complication in this experiment arises because the first

baroreceptor stimulus can sometimes lead to a reflex fall

in blood pressure. When this happens the second stimulus

occurs at a time when natural baroreceptor input would be

reduced, and so when the baroreceptor contribution to

vagal tone would be reduced. By choosing arterial pressure

pulses of appropriate size, vagal discharge could be

increased without significantly altering arterial pressure.

This was done in five dogs. In those five animals, blood

pressure was measured in the carotid sinus on the

experimental side and also, as a monitor of systemic blood

pressure, on the unstimulated and unoccluded contralateral

side (e.g., see fig 8.13). Under these conditions, the

second baroreceptor stimulus of a pair still evoked

smaller increases in vagal discharges than the first.

This effect cannot be attributed to reductions in arterial

pressure.

It is argued in the Discussion section which follows that

the reduced effectiveness of the second of a pair of

arterial pulses is due to processes occurring within the

central nervous system. The alternative view, that the

second pulse evoked a smaller volley of afferent 172

.!. ":."'.: J--- I Min. HR ] after~ min-1 t

Fig 8.13. Dog, anaesthetised with chloralose. Records in panel A are of efferent activity in a single cardiac vagal fibre, arterial blood pressure measured in the ipsilateral carotid sinus, a histogram of accumulated vagal action potentials, systemic blood pressure measured in the contra­ lateral unoccluded carotid sinus, and heart rate. Records were obtained by superimposing 20 sweeps of an oscilloscope trace triggered from the first of pairs of pulses of pressure delivered into the ipsilateral carotid sinus 2.5 s apart. Systemic blood pressure (mean ± S. E. M) was measured in the contralateral carotid sinus at the time of delivery of each pulse of pressure. Heart rate was measured as the minimum (mean± S.E.M) rate following each pulse of pressure. The first of the pair of pressure pulses evoked an increase in vagal discharge and fall in heart rate: the second of the pair of pulses evoked less vagal discharge and did not elicit as great a fall in heart rate. Systemic arterial pressure was not significantly different at the moment of delivery of each pulse of the pairs. Panel B shows the responses to two pulses of pressure given 2. 5 s apart into the carotid sinus. The activity recorded in the baroreceptor fibre (top trace) and the histogram accumulated from this trade (middle trace) show that the baroreceptor fibre responds almost identically to each pulse of the pair. There is no inhibition of the second response. 173 baroreceptor impulses, was examined in direct recordings of arterial baroreceptor fibres dissected from the carotid sinus nerves in five dogs. These recordings gave no evidence of reduction in baroreceptor responses to the second of a pair of pµlses (see fig 8.13).

4. DISCUSSION (CHAPTER 8)

The results described here have confirmed that the latencies of response evoked in single cardiac vagal efferent fibres by electrical stimuli applied to the carotid sinus nerve vary between fibres and, for any one fibre, can vary by 5 -15 ms (Iriuchijima & Kumada, 1964;

Kunze, 1972). Also, the variability of this latency for individual vagal units depends in part upon the timing of stimuli within the respiratory cycle. A similar respiratory variation of latency was apparent in the published data of McAllen & Spyer (1978b) on recordings made from cardiac vagal motoneurones in the nucleus ambiguus of cats (see their fig 6), but was not commented upon there. Respiratory changes in central reflex transmission can be held responsible for much of the previously unaccounted for variability of individual response latencies. This phasic influence on latencies is another manifestation of the respiratory modulation which operates on excitatory mechanisms for cardiac vagal motoneurones (e.g., Koepchen, et al., 1961; see also chapter 4). Some variability of latency remains, however, 174 in both expiration and inspiration, as is clear from figs 8.2 and 8.3. This was shown here not to be attributable to changes associated with the cardiac cycle because, when stimuli were triggered from the electrocardiogram, there was no further reduction in the variability of the latency. Other factors that might influence response latencies were not studied in these experiments.

While the latencies of electrically evoked responses in cardiac vagal efferent fibres were examined in some detail here, and by other workers (Iriuchijima & Kumada, 1964;

Kunze, 1972; McAllen & Spyer, 1978b), little information has previously been available about the behaviour of vagal efferent nerves following their initial responses to these stimuli. Trzebski, et al (1975) recorded in the nucleus of the solitary tract fr0II1 eighty-four cells which were excited by electrical stimuli applied to the carotid sinus nerve, and reported post-excitatory depression lasting "up to several hundred milliseconds" in twenty-three of these. Whether or not these cells were relays to cardiac vagal motoneurones, however, was not determined.

Vagal efferent activity well beyond the time of the initial evoked response was looked at here, and a period of marked depression of spontaneous discharge lasting for 100-150ms was demonstrated. This depression was not due to long effective refractory periods of the vagal motoneurones leading to prolonged pauses in spontaneous activity following each evoked response (fig 8.7), nor was it due to inhibition 175 of receptors by 'backfiring' (fig 8.8) or stimulus spread to other nerves (see fig 8.9), but represented a true inhibition of discharge caused by the electrical stimulation.

The cause of the post-excitatory inhibition and any possible functional significance it might have are not clear. Vagal inhibition was not seen following brief, selective stimulation of carotid baroreceptors using intracarotid pulses of pressure, although it was still apparent after the last pulse in a train of electrical stimuli applied to the carotid sinus nerve. Possibly the synchronous volley of action potentials produced in the carotid sinus nerve by electrical stimulation, but presumably not by functional stimulation of the baroreceptors, favoured the demonstration of vagal inhibition. If so, the inhibition need not have been an artefact of electrical stimulation it might usually be present, but obscured by excitatory effects exerted simultaneously by other inputs. Certainly, bursts of afferent impulses, whether produced electrically or with functional stimuli, do evoke vagal excitation which can outlast the stimuli by some seconds (for example, see figs 8.11, 8.12 and chapter 7). Excitation of this kind is not usually apparent after a single electrically­ induced afferent volley, but appears to build up as a result of successive stimuli. Thus, while a second electrical stimulus in a pair delivered to the sinus nerve is greatly reduced in its effectiveness, successive further stimuli are reduced progressively less (fig 8.11). This excitatory or facilitatory influence of trains of stimuli 176

builds up slowly and, when the train of stimuli ceases,

declines slowly. At frequencies of afferent input higher

than a certain level (here from 30 - 100 Hz, depending on

the individual vagal motoneurone), vagal motoneurones fail

to follow the input frequency despite any excitatory

influences which may have been developing. Indeed, such

excitatory effects do seem to be induced because vagal

discharge is stimulated as soon as the high frequency

afferent volley stops. This was shown in fig 8.11. The

failure of vagal units to follow high frequencies of

afferent stimulation may, therefore, represent a conduction

block at some point in the reflex pathway.

:Another difference between electrical and functional

stimulation of afferents in the carotid sinus nerve would

be in the types of afferent fibre stimulated. The pulses

of pressure employed here selectively stimulate arterial

baroreceptors (Haymet & Mccloskey, 1975). However, both

baroreceptor and chemoreceptor fibres would be engaged by

electrical stimulation. Fidone & Sato (1969) showed that

both receptor types transmit through myelinated (A) and

unmyelinated (C) afferents, and claims that one or other

type can be selectively stimulated electrically are most

doubtful. While both chemoreceptor and baroreceptor

inputs evoke cardiac vagal responses (e.g., Daly & Scott,

1963; Davidson, et al., 1976) and so could presumably

contribute to initial excitatory effects, chemoreceptor

afferents are excitatory also to breathing (e.g., Black &

Torrance, 1972). Possibly, therefore, the excitation of 177

central respiratory neurones by chemoreceptor afferents

JI}igh.t, in turn, transiently inhibit cardiac vagal motoneurones : such inhibition could then appear after the

'primary' reflex excitation, and so account for the post- excitatory depression noted here. No consistent respiratory effects of single electrical stimuli given to the carotid sinus nerve were noted here, although breathing was recorded by relatively insensitive methods.

The effects of electrical stimulation of the sinus nerve

on efferent activity in cardiac vagal fibres are shown by

the present study to be limited in several ways. The

inhibitory phase which succeeds evoked vagal responses,

limits the effectiveness of single shocks and prevents

effects from the second shock in temporally close pairs.

Seller & Illert (1969) described limitations on

transmission in cells in the nucleus of the tractus solitarius

(the first synapse of the baroreceptor pathway) as the

frequency of afferent stimulation increased above lHz.

Possibly the post-excitatory inhibition described here was

important for these effects. Longer bursts of afferent

stimuli offset these inhibitory effects to some extent by

contributing excitatory effects of somewhat slower onset

and conclusion. These slower effects mean, however, that

vagal excitation is sustained beyond the end of a stimulus

train. At higher frequencies of electrical stimulation

vagal excitation can be severely limited during a train of

stimuli although elevated afterwards. This phenomenon may

lead to paradoxical responses in vagal units during 178 intermittent bursts of sinus nerve stimulation, and could be relevant to studies in which intermittent and continuous stimulation are compared (e.g., DougLas, Ritchie & Schaumann,

1956; Kendrick, Matson, Oberg & Wennergren, 1973; Sedin,

1976). All these considerations must limit the usefulness of carotid sinus nerve stimulation for some investigations.

The prolonged responses of cardiac vagal motoneurones to brief trains of electrical stimuli or pulses of intracarotid pressure cannot be attributed to a temporal spread of arrival in the central nervous system of inputs along afferent fibres of different conduction velocities.

Allowing a conduction distance of 5 cm from carotid sinus to medulla in the dog, a spread of only -50 ms could be accounted for between afferents conducting at lm/s and

1oom/s, but the cardiac vagal response outlasts the stimulus which evokes it by 500 ms or more.

The final element in the chain of excitations and inhibitions described here is the apparent inhibition of the vagal response to a pulse of arterial pressure by a similar preceding pulse. As noted in the results above, this inhibition does not seem to involve a reduction in the response of the baroreceptors themselves to the second of the pair of pulses, nor can it be attributed to changes in vagal excitability which are secondary to changes in systemic blood pressure. The well-recognised pulse modulation of efferent cardiac vagal fibres (e.g.,

Jewett, 1964; Katona, et al., 1970; McAllen & Spyer, 179

1978b) indicates that inhibitory influences of a train of pul~es cannot extinguish the vagal responses altogether.

Presumably an equilibrium between excitatory and inhibitory influences is reached. In any case, the pressure pulses used here were large and abrupt and, while useful experimental devices, were unlike stimuli which might occur physiologically. Th~ functional significance of responsesthat such pulses reliably evoke remains, therefore, to be determined. 180

CHAPTER 9

GENERAL DISCUSSION 181

1. PHENOMENA STUDIED

It has been known for many years that primary reflex bradycardia evoked by stimulation of arterial chemo­ receptors is inhibited by the increase in ventilation simultaneously evoked (see Bernthal, Greene and Revzin, 1951; Daly and Scott, 1958). That mechanisms associated with the inspiratory phase of breathing were responsible for inhibiting the reflex bradycardia was made clear by the experiments of Koepchen, et al (1961) and Haymet and Mccloskey (1975). Electrical stimuli delivered to the carotid sinus nerve (Koepchen, et al, 1961) or brief functional selective stimulation of the arterial baro­ receptors or chemoreceptors (Haymet and Mccloskey, 1975) were only effective in slowing the heart when given in ~he expiratory phase of breathing. Stimuli given in inspiration evoked little or no change in heart rate.

Koepchen, et al (1961) and Davidson, et al (1976) were able to show that this inspiratory inhibition of reflex bradycardia was due at least in part to central inspiratory activity. The work described in this thesis extends these findings.

Not only does central inspiratory activity inhibit reflexly evoked bradycardia but a phasically active mechanism based on the activation of intrapulmonary receptors by lung inflation can also inhibit reflexly evoked bradycardia. This was so for all the reflexes that slow the heart and 182 which were examined here (see fig 9.1). Inspiratory inhibition of cardiodepressor reflexes seems, therefore, to be a general phenomenon. One can predict that cardio­ depressor responses not studied here - such as vasovagal responses or reflexes based on ventricular or intrapulmonary 'J' receptors (see Introduction, chapter 1) will be inhibited similarly.

In the intact animal, of course, both the central and the pulmonary reflex mechanisms of inhibition would act together and so their effects would be doubly assured.

2. METHODS

The use of anaesthesia in experiments such as those described here has already been defended. Ultimately, that defence rests on the demonstrations (here, and by others) of essentially similar phenomena in conscious human subjects. Those same demonstrations also vindicated the use of dogs as experimental animals here, although it is clear from the work of others (e.g., McLeod and Scott, 1964; Daly, et al., 1978a) that qualitatively similar phenomena may appear in different species in quantitatively different ways.

A significant part of the present work involved observations on heart rate. As increases in cardiac vagal activity are reliably reflected in changes in heart rate, these 183

CHEMORECEPTOR BARORECEPTOR DIVING OCULOCAR0IAC STIMULATION STIMULATION REFLEX REFLEX

BLOCKED BY INSPIRATION A Central inspiratory Stimulation of activity lung receptors by inflation l

~ HEART RATE

Fig 9.1. Summary of the inhibitory effects of inspiration

on the reflexes studied. Central inspiratory

activity alone or lung inflation alone inhibits

the reflex bradycardia evoked by stimulation of

the baroreceptor reflex, chemoreceptor reflex,

oculocardiac reflex or the diving response. 184 observations have been valuable in analysing stimuli that excite the vagus.

The responses of heart rate to vagal discharge are markedly asymmetrical. Heart rate slows immediately the vagal discharge increases but only slowly returns to resting levels once vagal activity ceases. Such asymmetry is caused by mechanisms at intracardiac vagal terminals (e.g.,

Brown & Eccles, 1934; see chapter 1 for discussion). In addition, as has been shown here, the cardiac preganglionic vagal motoneurones behave asymmetrically to neural inputs for example, an increase in baroreceptor input excites a large and prompt increase in vagal discharge but a decrease in baroreceptor input is only slowly followed by a decline in vagal discharge. Such considerations limit the usefulness of heart rate responses in a study of vagal mechanisms. A sudden interruption of vagal firing will only slowly be followed by a rise in heart rate, and, if that interruption is relatively brief (as, for example, if it occurs only for the duration of a single inspiration), the rise in heart rate caused will be quite small. In general, increases in heart rate caused by a brief or intermittent withdrawal of vagal activity are likely to be poor reflections of neural changes of greater magnitude. Direct recordings from cardiac vagal efferents were, therefore, necessary for full documentation of vagally­ mediated reflexes.

Not only were the timing and magnitude of vagal effects 185 best seen in nerve recordings, as the arguments above lead one to expect, but also differences between central and reflex inhibitory effects on vagal firing were able to be demonstrated. The contrasting effects of lung inflation and central inspiratory drive on tonic and reflexly evoked vagal firing would not have been demonstrable without direct neural recordings.

The identification of cardiac vagal efferent fibres presents some difficulties. The work of Jewett (1963, 1964) and Iriuchijima and Kumada (1963, 1964).upon which the criteria for identification of such fibres was based, was discussed earlier (chapter 1, section 1.7a). Ultimately these criteria depend upon the excitation of cardiac vagal units by baroreceptor afferents and, perhaps (in the case of electrical stimulation of the sinus nerve), by chemo­ receptor afferents as well. Such criteria are widely used and accepted. Nevertheless, if cardiac vagal efferents exist that are not excited by such inputs, then they will not have been selected for study here. Nor will they have been studied in other work on vagal unitary behaviour. It is not possible to say from the present study whether such baroreceptor-independent units exist, nor, if they do, whether the influences of neural respiratory mechanisms on them are similar to those described here. Korner and colleagues (Korner, et al., 1973; see also Korner, 1979, for review) described 'baroreceptor-dependent' and 'baro­ receptor-independent' pathways influencing the baroreceptor heart rate reflex. It is possible that the baroreceptor 186 independent effects they described may well influence heart rate by pathways other than cardiac vagal motoneurones (e.~, sympathetic pathways), and so the selection of cardiac vagal motoneurones based on baroreceptor inputs are not necessarily invalidated by such demonstrations.

None of the vagal efferent units that were identified anatomically as cardiac was independent of baroreceptor inputs (Rijlant, 1936a, b; Green, 1959; Jewett, 1963, 1964;

Kunze, 1972). Davidson, et al (1976) found some cardiac vagal units that responded to chemoreceptor but not baroreceptor inputs, but in their experiments stimuli were given in only one carotid reflexogenic zone, and baroreceptor inputs from other sensory zones on to these units activated by chemoreceptors were not ruled out. In recordings from the medullary cell bodies of cardiac vagal neurones, no units have been identified by 'back-firing' from cardiac branches, that do not also receive baroreceptor inputs

(see Spyer, 1979, for review). Thus, while one must maintain some reservations about identification of cardiac vagal units, the methods used here seem as reliable as any.

Fibre splitting is a technique used here to isolate single

fibre or few-fibre activity. The procedure favours selection of large axons (with large action potentials)

over smaller axons. Unmyelinated axons are most difficult

to select, small myelinated fibres (studied here) are next

hardest and large myelinated axons are least difficult.

Chapter 1 reviewed the evidence that vagal cardio-inhibitory 187 fibres were of the small myelinated type, but also outlined anatomical evidence that there is a considerable efferent cardiac innervation with unmyelinated fibres. It is most unlikely that any recordings were made here from unmyelinated fibres, and certainly any active units that might have been picked up would not have been studied if they had not met the selection criteria discussed above. Clearly, unmyelinated vagal efferent fibres deserve special study. Possibly the best approach might be through recording compound activity in the so-called 'antidromic collision technique' (Douglas and Ritchie, 1957), rather than by attempting unitary recordings.

3. MECHANISMS

The original work of Koepchen and his colleagues (1961) was outlined in chapter 1, section 1.6a, and illustrated in figs 1.3 and 1.4. In brief, they showed (i) respiratory modulation of the heart rate responses to baroreceptor and carotid sinus nerve stimuli; (ii) that this modulation persists in paralysed animals (showing the adequacy of purely central events in imposing it); (iii) that in unparalysed animals a central excitatory state set up by a stimulus given in one expiratory phase can be augmented by a stimulus given later in a succeeding expiratory phase, and (iv) that pairs of stimuli can only summate in their effects on central vagal excitation when both stimuli fall in the expiratory phase. Findings listed 188 in (iii) and (iv) above show, respectively, that occurrence of an inspiration with its accompanying inhibitory effects cannot extinguish the central excitatory state set up by a stimulus given in expiration, and that stimuli given in inspiration are not simply prevented from evoking an output from the vagal motoneurone but are prevented also from adding to the central excitatory state. These conclusions are stressed here because they fit so readily with the simple idea of two inhibitory mechanisms, one (central) inhibiting the output from a central excitatory state, and another (intrapulmonary receptor-based) inhibiting access of excitatory inputs to that central excitatory state. That was the basis of the simple model proposed earlier here in fig 7.7, and clearly the results of Koepchen and his colleagues fit with it. However, Koepchen, et al., (1961) did not see things in quite this way, and their analysis went only so far as suggesting the alternative models illustrated in fig 9.2.

Koepchen and his co-workers discussed only central sources of inhibition, and did not propose any action for afferent receptor-based inhibition.

The simple model of fig 7.7 proposes that intrapulmonary receptor-based inhibition prevents access of excitatory inputs to a central excitatory state. The central excitatory state can be envisaged as a vagal tone-sustaining mechanism. Central inspiratory activity inhibits the emergence of activity from the vagal motoneurones, but 189

Central rhythm

MOOEL A. MODEL B MOOEL A.+B

Fig 9.2. Models suggested by Koepchen, et al (1961

redrawn) to explain their findings on respiratory modulation of the baroreceptor

- cardiodepressor reflex. See text for discussion. 190 does not affect the access of excitatory inputs to the tone-sustaining mechanism. The model does not propose precise locations for the two points of inhibition, except that one is early in the reflex pathway (although not necessarily at the first synapse) and the other is late (although not necessarily at the vagal cell body itself).

Seller and Illert (1969) recorded potentials evoked in neurones of the nucleus of the tractus solitarius by carotid sinus nerve stimulation, and found no differences in amplitudes or latencies of responses evoked when stimuli were delivered in different respiratory phases. This fitted well with the later observation by Jordan and Spyer (1979) that, at least for arterial chemoreceptor or baroreceptor inputs, the inhibition caused by central inspiratory activity is not exerted presynaptically on the terminals of afferents from the carotid sinus nerve, In experiments on cats, McAllen and Spyer (1978 b : see also Spyer, 1979; Spyer and McAllen, 1980) showed that pulse­ synchronous activity evoked through arterial baroreceptor inputs could be recorded extracellularly from the cell bodies of cardiac vagal motoneurones throughout the respiratory cycle. Such activity was usually demonstrable during the central inspiratory phase only after direct iontophoretic application of excitant amino acids to the vagal cell bodies. Spyer and McAllen (1980) argued that because they were able to "demonstrate qualitatively identical inputs from the carotid sinus baroreceptors and 191

the sinus nerve during both inspiration and expiration ...

that this observation excludes a respiratory 'gating' of reflex inputs at an earlier stage in the reflex pathway . . " . Certainly their analysis does show that there is no

complete block of access of baroreceptor inputs to vagal motoneurones during inspiration, and permits McAllen and

Spyer's suggestion that central inspiratory activity exerts

its vagal inhibitory action at the vagal motoneurones

themselves. However, additional incomplete, but possibly

quite powerful, inhibition at earlier stages of the reflex

pathway is not at all excluded. Spyer and McAllen's (1980)

"qualitatively identical inputs" to the vagal motoneurones

evoked responses which, during inspiration, were greatly

reduced in amplitude, and also were delayed in latency to

peak firing by approximately 15 ms ( see their fig 6; see

also chapter 8 here). Inhibitory influences earlier in

the reflex pathway could well have been responsible for

these effects. It has been argued here that inhibitory

effects from intrapulmonary receptors are exerted at such

earlier stages in the reflex pathway from baroreceptors to

vagal motoneurones. In any case, McAllen and Spyer's

(1978a) experiments were unsuited to demonstrating effects

from intrapulmonary receptors: their observations were

made in open-chested animals in which one lobe of the lung

had been removed, and which were receiving a shallow artificial ventilation.

Only the findings reported in this thesis required the

refinement of the simple model of fig 7.7, to the more 192

complex one of fig 7.8. The reported observations of other workers can be accommodated by both models.

4 SIGNIFICANCE

It is difficult to say with certainty what is

the primary physiological significance of inspiratory

inhibition of the reflex effects on heart rate, It is

clearly important because it is based on two independent mechanisms that can function alone, but which usually work

together.

The effects described in this thesis can be best understood

in terms of oxygen conservation and delivery. When chemo­

receptors are stimulated during hypoxia it is appropriate

that the increased demand for oxygen that is met by

increased ventilation should be made quickly available to

hypoxic tissues. This could be assisted by increasing the

heart rate. Vasodilatation of peripheral vessels, also a

reflex accompaniment of increased breathing (Daly and Scott,

1963) further facilitates oxygen distribution. The more

thorough interruption of vagal activity made possible by

increased breathing is shown diagrammatically on the right

hand side of fig 9.3. In this way, tachycardia is seen in

states of increased ventilation, whether evoked by hypoxia,

muscular exercise or a variety of other circumstances.

Again the responses are appropriate for the handling of 193

CONSTANT VAGAL EXCITATION +

VARIABLE BREATHING

X •••• ...... X······················· X v v ---A..J\Jl v l l l Sinus arrhythmia +H.R. t H.R.

Fig 9.3. Diagrammatic representation of the

interaction between breathing and

vagal activity at three different

levels of breathing. See text for

discussion. 194 oxygen when breathing is descreased or halted as in the diving response. Now the heart can slow more effectively and bradycardia enables the body to conserve the oxygen present. Selective vasoconstriction also occurs in these circumstances (e.g., Daly, 1972) and helps conserve the available oxygen while allowing essential organs, especially heart and brain, to function. This interaction between breathing and heart rate is shown diagrammatically in the middle of fig 9.3. Bradycardia occurring in the hypoxic foetus may also be a manifestation of this effect.

Sinus arrhythmia, the waxing and waning of heart rate with breathing, can also be understood in terms of this periodic interruption of vagal activity by inspiration. Any steady excitatory input will be converted to an oscillating output in this way. The participation of chemoreceptor stimulation in the expiratory bradycardia seen in sinus arrhythmia has been demonstrated (Tafil and Trzebski, 1980). Similarly an exaggerated sinus arrhythmia can be seen in animals during perfusion of the carotid bodies with hypoxic blood

(Levy, De Geest and Zieske, 1966). As with chemoreceptor stimulation, stimulation of carotid baroreceptors can also evoke an exaggerated sinus arrhythmia (see Schweitzer,

1937; Przybyszewski and Trzebski, 1980). Diagrammatic representation of the generation of sinus arrhythmia during quiet breathing is shown on the left side of fig 9.3. As noted in the Introduction in chapter 1, the additional participation of oscillating inputs may augment sinus arrhythmia set up by the mechanisms just described. 195

5 APPLICATIONS

Pressure to the eyes (oculocardiac reflex), carotid sinus massage (baroreceptor reflex), or application of water to the face (diving response) all evoke reflex bradycardia, and so are used by the clinician in attempts to terminate bouts of supraventricular tachycardia. The results presented in this thesis suggest that, as inspiration inhibits vagal activity, these manoeuvres would be more effective if carried out while the patient holds his breath.

On the other hand, in cases where severe bradycardia may become life-threatening as during nasopharyngeal or ophthalmic surgery (see Katz and Bigger, 1970), the observations presented here on the oculocardiac and diving reflexes (chapters 5 and 6) permit one to advise anaesthetists or surgeons to increase the frequency and amplitude of artificial ventilation to relieve such bradycardia. Vasovagal faints could also be prevented by voluntarily increasing breathing. Moreover, as swallowing is known to be associated with intense bursts of firing in inspiratory neurones, and is accompanied by tachycardia (see fig 6.1), the glass of water given to someone who feels faint has a sound physiological basis. 196

6 FURTHER EXPERIMENTS

The models proposed in figs 7.7 and 7,8 and discussed above can be examined and refined by recording either in the nucleus of the tractus solitarius (the first synapse in the baroreceptor and chemoreceptor reflex pathway) or in the nucleus ambiguus (the last synapse in that pathway), and the relative effects of lung inflation and central inspiratory activity can be examined individually. The effects of lung inflation and central inspiratory activity have not, together, been examined th0roughly and compared at these two synapses.

In the introduction, other reflexes which slow the heart, such as those based on ventricular receptors and J-receptors were reviewed. They were not studied here but it can be predicted that these reflexes also would be inhibited by inspiration, i.e., lung inflation alone, or central inspiratory activity alone. This prediction needs to be tested experimentally.

From the experiments described in chapter 4 (see especially fig 4.11) activity of pulmonary stretch receptors was suggested as being responsible for the inhibition seen during lung inflation. This suggestion was made by excluding other principal intrapulmonary receptor~types as responsible. It should follow that electrical stimuli delivered at the hilum of the lung to pulmonary vagal branches, so as to engage only low threshold fibres (i.e., 197

pulmonary stretch receptor afferents ), shouJ!.d reproduce

the inhibitory effects of lung inflation. This should be

tested.

In the discussion of methods earlier in this chapter, it was noted that the function of unmyelinated vagal efferents

remains obscure. In the same section the matter of

convergence of inputs on to cardiac vagal motoneurones was

also discussed. Both topics deserve further study.

Finally, the observations reported in chapter 8 (see fig

8.13) on adaptation of the vagal responses to baroreceptor

stimuli, clearly deserve more extensive study. 198

REFERENCES

ABRAHAMS, V.C., HILTON, S.M and ZBROZYNA, A (1960) Active muscle vasodilation produced by stimulation of the brain stem : its significance in the defence reaction. J.Physiol. 154 : 49 - 513.

AGOSTINI, E., CHINNOCK, J.E., DALY, M.de B. and MURRAY, J.G (1957) Functional and histological studies of the vagus nerve and its branches to the heart, lungs and abdominal viscera in the cat. J.Physiol. 135 : 182 - 205.

ALEXANDER, J.P (1975). Reflex disturbances of cardiac rhythm during ophthalmic surgery. Brit.J. Ophthal. 59 : 518 - 524

ANDERSEN, H.T (1963). Factors determining the circulatory adjustment to diving. I. Water immersion. Acta physiol. scand. 58 : 173 - 185.

ANDERSEN, H.T (1966). Physiological adaptations in diving vertebrates. Physiol. Rev. 46 : 212 - 243

ANGELL JAMES, J.E and DALY, M. de B (1969). Cardio­ vascular responses in apnoeic asphyxia: role of arterial chemoreceptors and the modification of their effects by a pulmonary vagal inflation reflex. J.Physiol. 201: 87 - 104. -

ANGELL JAMES, J.E and DALY, M. de B. (1970). Comparison of the reflex vasomotor responses to separate and combined stimulation of the carotid sinus and aortic arch baroreceptors by pulsatile and non-pulsatile pressures in the dog. J.Physiol. 209 : 257 - 293.

ANGELL JAMES, J.E and DALY, M. de B. (1972 a). Some mechanisms involved in the cardiovascular adaptations to diving. Symposia of the Society for Experimental Biology no. 26. The Effects of Pressure on Organisms. pp. 313 - 341.

ANGELL JAMES, J.E and DALY, M. de B. (1972b). Reflex respiratory and cardiovascular effects of stimulation of receptors in the nose of the dog. J.Physiol. ~: 673 - 696.

ANGELL JAMES, J. E and DALY, M de B. ( 1973). The interact ion of reflexes elicited by stimulation of carotid body chemorecep:tors and receptors in the nasal mucosa affecting respiration and pulse interval in the dog. J.Physiol. 229 133 - 149. 199

ANGELL JAMES, J.E and DALY, M. de B. (1978). The effects of artificial lung inflation on reflexly induced bradycardia associated with apnoea in the dog, J.Physiol. 274 : 349 - 366.

ANGELL JAMES, J.E, DALY, M. de B., and ELSNER, R (1978). Arterial baroreceptor reflexes in the seal and their modification during experimental dives. Am. J.Physiol .. 234 : H730 - 739.

ANREP, G.V., PASCUAL, W. and ROSSLER, R (1936a). Respirator variations of the heart rate. I. The reflex mechanism of the respiratory arrhythmia. Proc. R. Soc. B 119 : 191 - 217.

ANREP, G.V., PASCUAL, Wand ROSSLER, R (1936b). Respiratory variations of the heart rate. II. The central mechanism of the respiratory arrhythmia and the interrelations between the central and reflex mechanisms. Proc. R. Soc. B 119 : 218 - 230.

APT, L., ISENBERG, Sand GAFFNEY, W.L (1973). The oculocardiac reflex in strabismus surgery. Am.J.Ophthalmol. 76 : 533 - 536.

ARMSTRONG, G.G.Jr, PORTER, H.Jr, and LANGSTON, J.B (1961). Alteration of carotid occlusion response by anesthesia. Am.J.Physiol. 201 : 897 - 900.

ASCHNER, B (1908). Uber einen bisher noch nicht beschriebenen reflex vom auge auf kreislauf und atmung. Wein, Klin. Wschr. 21 : 1529 -1540.

ASERINSKY, E and De BIAS, D.A (1963). Suppression of oculocardiac reflex by artificial respiration. Arch. Ophthalmol. (chicago) 69 : 484 - 491.

BAINBRIDGE, F.A (1920). The relation between respiration and the pulse rate. J .Physiol. 54 : 192 - 202.

BAMFORD, O.W and JONES, D.R (1976). Respiratory and cardiovascular interactions in ducks : the effect of lung denervation on the initiation of and recovery from some cardiovascular responses to submergence. J.Physiol. 259 : 575 - 596. -

BARCROFT, Hand EDHOLM, O.G (1945). On the vasodilatation in human skeletal muscle during post­ haemorrhagic fainting. J .Physiol. 104 : 161 - 175

BASKERVILLE, A.L., ECKBERG, D.L and THOMPSON, M.A (1979). Arterial pressure and pulse interval responses to repetitive carotid baroreceptor stimuli in man. J.Physiol. 297 : 61 - 71. 200

BAUMGARTEN, R. von and KANZOW, E (1958). The interaction of two type~ of jnspiratory neurones in the region of the tractus solitarius of the cat. Arch. ital. Biol. 96 ? 361 ... 373.

BERGER, A.J and MITCHELL, R.A (1977). Studies of medullary neurons participating in lung inflation reflexes in the cat. Proceedings of the International Union of Physiological Sciences. Vol. 13 Abstract no. 188.

BERNTHAL, T (1938). Chemo-reflex control of vascular reactions through the carotid body. Am.J. Physiol. 121 : 1 - 20.

BEZOLD, A.von and HIRT, L (1867). Uber die physiologischen Wirkungen des essigsauren veratrins. Unters. Physiol. Lab. Wurzburg., 1 : 75 - 156, cited by Sleight, Pin Cardiac Receptors Ch.II. ed. R. Hainsworth, C. Kidd and R.J Linden, 1976. Cambridge University Press, Cambridge.

BERNTHAL, T. , GREENE, W. Jr and REVZIN, A. M ( 1951). Role of carotid chemoreceptors in hypoxic cardiac acceleration. Proc. Soc. Exp. Biol and Med. 76 : 121 - 124.

BISCOE, T.J and SAMPSON, S.R (1970a). Responses of cells in the brain stem of the cat to stimulation of the sinus, glossopharyngeal, aortic and superior laryngeal nerves. J.Physiol. 209 359 - 373. -

BISCOE, T.J and SAMPSON, S.R (1970b). An analysis of the inhibition of phrenic motoneurones which occurs on stimulation of some cranial nerve afferents. J.Physiol. 209 : 375 - 393.

BLACK, A.M.S., McCLOSKEY, D.I and TORRANCE, R.W (1966). The responses of peripheral chemoreceptors to sudden changes of hypercapnic and hypoxic stimuli. J.Physiol. 185 : 67 - 68P.

BLACK, A.M.S., McCLOSKEY, D.I and TORRANCE, R.W (1971). The responses of carotid body chemoreceptors in the cat to sudden changes of hypercapnic and hypoxic stimuli. Resp.Physiol. 13 : 36 - 49. -

BLACK, A.M.S.and TORRANCE, R.W (1967). Chemoreceptor effects in the respiratory cycle. J.Physiol. 191 : 114 -115 P.

BLACK, A.M.S and TORRANCE, R.W (1971). Respiratory oscillations in chemoreceptor discharge in the control of breathing. Resp.Physiol. 13 221 - 237. 201

BLAIR, D.A., GLOVER, W.E., GREENFIELD, A.D.M and RODDIE, I.C (1969). Excitation of cholinergic vasodilator nerves to human skeletal muscles during emotional stress. J.Physiol. 148 : 633 - 647.

BRISTOW, J.D., PRYS-ROBERTS, C., FISHER, A., PICKERING, T.G and SLEIGHT, P (1969). Effects of anaesthesia on baroreflex control of heart rate in men. Anaesthesiology, 31 : 422 - 428. BRONK, D.W (1933). The nervous mechanism of cardiovascular control. Harvey Lecture 29. pp. 245 - 262.

BRONK, D.W and STELLA, G (1935). The response to steady pressures of single end organs in the isolated carotid sinus. Am.J.Physiol. 110 : 708 - 714.

BROWN, G.L and ECCLES, J.C (1934). The action of a single vagal volley on the rhythm of the heart beat. J.Physiol. 82 : 211 - 241.

BROWN, R.V and HILTON, J.G (1956). The effectiveness of the baroreceptor reflexes under different anesthetics. J.Pharmacol.Exptl.Therap, 118 198 - 203. -

BURGEN, A.S.V and TERROUX, K.G (1953). On the negative inotropic effect in the cats auricle. J. Physiol. 120: 449 - 464.

CALARESU, F.R and PEARCE, J.W (1965). Electrical activity of efferent vagal fibres and dorsal nucleus of the vagus during reflex bradycardia in the cat. J.Physiol. 176 : 228 - 240. CARMODY, J.J and SCOTT, M.J (1974). Respiratory and cardiovascular responses to prolonged stimulation of the carotid body chemoreceptors in the cat. Aust.J.Expt.Biol and Med. 52 ·: 271 - 283.

CIRIELLO, J and CALARESU, F.R (1980). Autoradiographic study of ascending projections from cardio­ vascular sites in the nucleus tractus solitarii in the cat. Brain Res. 186 : 448 - 453.

COLERIDGE, H.M. 1 COLERIDGE, J.C.G and KIDD, C (1964). Cardiac receptors in the dog, with particular reference to two types of afferent ending in the ventricular wall. J.Physiol. 174 : 323 - 339.

COLERIDGE, H.M. 1 COLERIDGE, J.C.G and ROSENTHAL, F (1976). Prolonged inactivation of cortical pyramidal tract neurones in cats by distension of the carotid sinus. J.Physiol. 256 : 635 - 649. 202

COOTE, J.H., HILTON, S.M and PEREZ-GONZALEZ, J.F (1979) Inhibition of baroreceptor reflex on stimulation in brainstem defence centre. J. Physiol. · 288 : 549 - 560.

COTTLE, M.K (1964). Degeneration studies of primary afferent of IXth and Xth cranial nerves in the cat. J.Comp.Neurol. 122 : 329 - 345.

COX, R.H (1972 a). Influence of chloralose anesthesia on cardiovascular function in trained dogs. Am.J.Physiol. 223 : 660 - 667.

COX, R.H (1972 b). Influence of pentobarbital anesthesia on cardiovascular function in trained dogs. Am.J.Physiol. 223 : 651 - 659.

CRILL, W.E and REIS, D.J (1968). Distribution of carotid sinus and depressor nerves in cat brain stem. Am.J.Physiol. 214 : 269 - 276.

CHALMERS, J.P., KORNER, P.I and WHITE, S.W (1967). The relative roles of the aortic and carotid sinus nerves in the rabbit in the control of respiration and circulation during arterial hypoxia and hypercapnia. J. Physiol. 188 : 435 - 450. -

DAGNINI, G (1908). Analisi di alcune forme di allaritmia cardiaca. Bull. sci.med.Bologna 8 380 - 390.

DALE, H. H ( 1963). Otto Loewi. Ergeb Physiol. 52 : 1 - 19.

DALY, M. de B. ( 1963). Reflex circulatory and respiratory responses to hypoxia in oxygen in the animal organism. Ed. F. Dickens and E. Neil. Pergamon Press, Oxford. pp. 267 - 276.

DALY, M. de B ( 1972). Interaction of cardiovascular reflexes. Sci.Basis of Med. Ann.Dev. Ch.XVII, 307 - 332.

DALY, M. deB and ANGELL JAMES, J.E (1975). Role of the arterial chemoreceptors in the control of the cardiovascular responses to breath-hold diving. In : The Peripheral Arterial Chemo­ receptors. Ed. M.J Purves, Cambridge University Press, Cambridge. ·

DALY. M. deB., ELSNER, Rand ANGELL JAMES, J.E (1977). Cardiorespiratory control by carotid chemo­ receptors during experimental dives in the seal. Am.J.Physiol. 232 : H508 -H516. DALY, M.deB and EVANS, D.H.L (1953). Functional and histological changes in the vagus nerve of the cat after degenerative section at various levels. J.Physiol. 120 : 579 - 595. 203

DALY, M. de B. , HAZZLEDINE, J. L. and HOWE, A (1965). Reflex respiratory and peripheral vascular responses to stimulation ot the isolated perfused aortic arch chemoreceptors of the dog. J.Physiol. 177 : 300 - 322. DALY, M. de B. , HAZZLEDINE, J. L. and UNGAR, A ( 1967). The reflex effects of alterations in lung volume on systemic vascular resistance in the dog. J.Physiol. 188 : 331 - 351. DALY, M. de B. , KORNER, P. I . , ANGELL-JAMES, J.E. and OLIVER, J. A . ( 1978 a) . Cardiovascular and respiratory effects of carotid body stimulation in the monkey. Clin. Exp. Pharmacol. & Physiol. 5 : 511 - 524. - DALY, M. de B. , KOR.'N'ER, P. I . , ANGELL-JAMES , J. E and OLIVER, J .,R ( 1978 b). Cardiovascular-respiratory reflex interactions between carotid bodies and upper-airways receptors in the monkey. Am. J. Physiol. 234 : H293 - H299. DALY, M. de B. and ROBINSON, B.H (1968). An analysis of the reflex systemic vasodilator response elicited by lung inflation in the dog. J.Physiol 195 : 387 - 406. DALY, M. de B. and SCOTT, M.J (1958). The effects of stimulation of the carotid body chemoreceptors on heart rate in the dog. J.Physiol. 144 148 - 166. DALY, M. de B. and SCOTT, M. J ( 1962) . An analysis of the primary cardiovascular reflex effects of stimulation of the carotid body chemoreceptors in the dog. J.Physiol. 162 : 555 - 573.

DALY, M. de B. and SCOTT, M. J ( 1963). The cardiovascular responses to stimulation of the carotid body chemoreceptors in the dog. J.Physiol. 165 179 - 197. DALY, M. de B. and UNGAR, A (1966). Comparison of the reflex responses elicited by stimulation of the separately perfused carotid and aortic body chemoreceptors in the dog. J.Physiol. 182 : 379 - 403. DAVIDSON, N.S., GOLDNER, S. and McCLOSKEY, D.I (1976). Respiratory modulation of baroreceptor and chemoreceptor reflexes affecting heart rate and cardiac vagal efferent nerve activity. J.Physiol. 259 : 523 - 530. DAVIES, R.O. and EDWARDS, M.W.Jr. (1973). Distribution of carotid body chemoreceptor afferents in the medulla of the cat. Brain Res. 64 : 451 - 454. 204

DAWES, G.S (1947). Studies on veratrum alkaloids. VII Receptor areas in the coronary arteries and elsewhere as revealed by the use of veratridine J. Pharm. exp. Ther. '89 : 325 - 342. DAWES, G.S., MOTT, J.C and WIDDICOMBE, J.G (1951), Respiratory and cardiovascular reflexes from the heart and lungs. J. Physiol. 115 : 258 - 291. -

DEACOCK, A.R. de C. and OXER, H.F (1962). The prevention of reflex bradycardia during ophthalmic surgery. Brit. J. Anaes. 34 : 451 - 457. DESHPANDE, S.S. and DEVANANDAN, M.S (1970). Reflex inhibition of monosynaptic reflexes by stimulation of type J pulmonary endings. J.Physiol. 206 : 345 - 357. DONALD, K.W and CHRISTIE, R.V (1949). A new method of clinical spirometry. Clin. Sci. 8 : 21 ... 31. DONG, E. and REITZ, B.A (1970). Effect of timing of vagal stimulation on heart rate in the dog. Circ. Res. 27 : 635 - 646. DOTY, R.W (1968). Neural organisation of deglutition. In: Handbook of Physiology, section 6 : Alimintary canal, vol. IV, pp. 1861 - 1902. American Physiological Society, Washington D.C. DOUGLAS, W.W. and RITCHIE, J.M (1957). A technique for recording functional activity in specific groups of medullated and non-medullated fibres in whole nerve trunks. J.Physiol. 138 : 19 - 30. DOUGLAS, W.W., RITCHIE, J.M. and SCHAUMANN, W (1956). A study of the effect of the pattern of electrical stimulation of the aortic nerve on the reflex depressor responses. J.Physiol. 133 : 232 - 242. DOUGLAS, W.W. and SCHAUMAN, W (1956). A study of the depressor and pressor components of the cat's carotid sinus and aortic nerves using electrical stimuli of different intensities and frequencies. J. Physiol. 132 : 173 - 186. DOWNING, S.E., REMENSNYDER, J.P. and MITCHELL, J.H (1962). Cardiovascular responses to hypoxic stimulation of the carotid bodies. Circ.Res. 10 : 676 ... 68-5. 205

EAD, H. W., GREEN, J ,H. and NEIL, E (1952). A comparison o;f the e;tfects o;( pulsatile and non,..._pulsatile blood flow through the carotid sinus on the reflexogenic activity of the sinus baroreceptor~ in the cat. J.Physiol. 118 : 509 - 519.

EBBECKE, U (1943). Reflex gesetzmassigkeiten des menschlichen schluckreflexes bei seiner Auslbsung von der Gesichtshaut her. Pflngers Arch. ges. Physiol. 246 : 675 - 692.

ECKBERG, D.L (1976). Temporal response patterns of the human sinus node to the brief carotid baroreceptor stimuli. J.Physiol. 258 : 769 - 782.

ECKBERG, D.L. and ORSHAN, C.R (1977). Respiratory and b'aroreceptor reflex interactions in man. J. Clin. Invest. 59 : 780 - 785.

ECKBERG, D.L. and ORSHAN, C.R (1980). Central respiratory - baroreceptor reflex interaction in man. In Central Interaction between Respiratory and Cardiovascular Control Systems. ed. H.P Koepchen, S.M Hilton and A.Trzebski Springer-Verlag. 206 - 215.

EDIS, A.J. and SHEPHERD, J.T (1971). Selective denervation of aortic arch baroreceptors and chemoreceptors in dogs. J. Appl. Physiol. 30 ; 294 - 296. ELDRIDGE, F.L (1972). The importance of timing on the respiratory effects of intermittent carotid sinus nerve stimulation. J.Physiol. 205 : 527 - 548.

ELSNER, R., ANGELL JAMES, J.E. and DALY, M. de B (1977). Carotid body chemoreceptor reflexes and their interactions in the seal. Am.J.Physiol. 232 : H517 - H526.

ELSNER, R., FRANKLIN, D.L., VAN CITTERS, R.L. and KENNEY, D.W (1966). Cardiovascular defense against asphyxia. Science. 153 : 941 - 949.

ERNSTING, J. and PARRY, D.J (1957). Some observations on the effects of stimulating the stretch receptors in the carotid artery of man. J.Physiol. 137 : 45P.

FIOONE, S.J. and SATO, A (1969). A study of chemoreceptor and baroreceptor A and C fibres in the cat carotid nerve. J. Physiol. 205 : 527 - 548. FREYSCHUSS, U (1970). Cardiovascular adjustment to somatomotor activation. Acta physiol. scand. suppl. 342 .. : 1 - 63. 206

FUSSEY; I.F., KIDD, C. and WHITWM4, J. G (1967). Single unit activity associated with cardiovascular events in the·brainstem·of the dog, J.Physiol.· 191 57 - 58P. GABRIEL, M. and SELLER, H (1970). Interaction of baroreceptor afferents from carotid sinus and aorta at the nucleus tractus solitarii. Pflugers. Arch. 318 : 7 - 20. GANDEVIA, S.C., McCLOSKEY, D.1 and POTTER, E.K (1978 a). Inhibition of baroreceptor and chemoreceptor reflexes on heart rate by afferents from the lungs. J. Physiol. 276 : 369 - 381. GANDEVIA, S. C. , McCLOSKEY, D. I . and POTTER, E. K ( 1978 b) . Reflex bradycardia occurring in response to diving, nasopharyngeal stimulation and ocular pressure, and its modification by respiration and swallowing. J. Physiol. 276 : 383 - 394. GARCIA, M., JORDAN, D. and SPYER, K.M (1979). Identification of aortic nerve cell bodies in the nodose ganglion of the rabbit and their central projections. J .Physiol. 290 : 23 - 24P. GEBBER, G.L. and KLEVANS, L.R (1972). Central nervous system modulation of cardiovascular reflexes. Fed.Pree. 31 : 1245 - 1252. GEIS, G.S. and WURSTER, R.D (1980). Horseradish peroxidase localization of cardiac vagal preganglionic somata. Brain Res. 182 : 19 - 30. GOODEN, B.A., STONE, H.L. and YOUNG, S (1974). Cardiac responses to snout immersion in trained dogs J. Physiol. 242 : 405 - 414. GOODWIN, G.M., McCLOSKEY, D.I. and MITCHELL,J.H (1972). Cardiovascular and respiratory responses to changes in central command during isometric exercise at constant muscle tension. J.Physiol. 226 : 173 -190. GREEN, J.H (1959). Cardiac vagal efferent activity in the cat. J .Physiol. 149 : 47 - 49 P. GROSS, P.M., WHIPP, B.J., DAVIDSON, J.T., KOYAL, S.N. and WASSERMAN, K (1976). Role of the carotid bodies in the heart rate response to breath holding in man. J.Appl. Physiol. 41 : 336 - 340. - GUNN, C.G., SEVELIUS, G., PUIGGARI, M.J. and MYERS, F.K (1968). Vagal cardiomotor mechanisms in the hindbrain of the dog and cat. Am. J. Physiol. 214 : 258 - 262. 207

HAMLIN, R.L. and SMITH, C.R (1968). Effects of vagal stimulation on S•A and A-V nodes. Am. J. Physiol. · 215 : 560 - 568.

HAYMET, B.T. and McCLOSKEY, D.I (1974). Effects on heart rate of brief stimuli to arterial baroreceptors or chemoreceptors delivered at different phases of the respiratory cycle. Proceedings of the Australian Physiological and Pharmacological Society. 5 : 79 - 8oP.

HAYMET, B.T. and McCLOSKEY, D.I (1975). Baroreceptor and chemoreceptor influences on the heart rate during the respiratory cycle in the dog. J.Physiol. 245 : 699 - 712.

HEINBECKER, P (1930). The potential analysis of the turtle and cat sympathetic and vagus nerve trunks. Am. J. Physiol. 93 : 284 - 306.

HERING, H.E (1924). Der sinus caroticus an der unrsprungsstelle der carotis interna als ausgangsort eines hemmenden herzreflexes und eines depressorischen gef~ssreflexes. Muench. Med. Wochschr. 71 : 701 - 704. Cited in Reflexogenic Areas ofthe Cardiovascular System. ed. C.Heymans and E.Neil, 1958. J & A Churchill Ltd., London.

HEYMANS, J.F. and HEYMANS, C (1927). Surles modifications directes et sur la regulation reflexe de l'activite du centre respiratoire de la t~te isololee du chien. Arch. Intern. Pharmacodyn. 33 : 272 - 370. HEYMANS, C. and NEIL, E (1958). Reflexogenic Areas of the Cardiovascular System. J & A Churchill Ltd., London.

HILTON, S.M (1963). Inhibition of baroreceptor reflexes on hypothalamic stimulation. J.Physiol. 165 56 - 57P.

HILTON, S.M. and SPYER, K.M (1971). Participation of the anterior hypothalamus in the baroreceptor reflex. J.Physiol. 218 : 271 - 293.

HUKUHARA, T. and OKADA, H (1956). Effects of deglutition upon the spike discharges of neurones in the respiratory centre. Jap. J. Physiol. 6 : 162 - 166. -

HUMPHREY, D,R (1967). Neuronal activity in the medulla oblongata of cat evoked by stimulation of the carotid sinus nerve. In : Baroreceptors and Hypertension, pp 131 - 168, ed. P. Kezdi. Pergamon Press, Oxford. 208

HURSH, J.B (1939). Conduction velocity and diameter of nerve fibres. Am. J. Physiol. 127 : 131 - 139.

HUTTER, O.F. and TRAUTWEIN, W (1955). Vagal effects on the sinus venosus of the frogs heart. J.Physiol. 129 : 48P.

INGRAM, W.R. and DAWKINS, E.A (1945). The intramedullary course of afferent fibres of the vagus nerve in the cat. J. comp. neurol. 82 : 157 - 168.

IRIUCHIJIMA, J. and KUMADA, M (1963). Efferent cardiac vagal discharge of the dog in response to electrical stimulation of the sensory nerves. Jap. J. Physiol. 13 : 599 - 605.

IRIUCHIJIMA, J. and KUMADA, M (1964). Activity of single vagal fibres efferent to the heart. Jap. J. Physiol. 14 : 479 - 487.

IRVING, L (1939). Respiration in diving mammals. Physiol. Rev. 19: 112 - 134.

JARISCH. A. and ZOTTERMAN, Y (1949). Depressor reflexes from the heart. Acta physiol. scand. 16 : 31 - 51.

JEWETT, D.L (1963). Reflex activation of vagal efferent neurones. D.Phil. thesis, Linacre House, University of Oxford.

JEWETT, D.L (1964). Activity of single efferent fibres in the cervical vagus nerve of the dog, with special reference to possible cardioinhibitory fibres. J.Physiol. 175 321 - 357.

JOELS, N. and SAMUELOFF, M (1956). The activity of the medullary centres in diffusion respiration. J. Physiol. 133 : 360 - 372.

JOHNSON, P., ROBINSON, J.S. and SALISBURY, D (1972). The onset and control of breathing after birth. In : Foetal and Neonatal Physiology : Proceedings of the Sir Joseph Barcroft Centenary Symposium. ed. K.S Comline, K.W Cross, G.S Dawes and P.W Natanielsz. Cambridge University Press, Cambridge.

JONES, D.R. and PURVES, M.J (1970). The carotid body in the duck and the consequences of its denervation upon the cardiac response to immersion. J. Physiol. 211 : 279 - 294. JORDAN, D., KHALID, M.E.M., SCHNEIDERMAN, N. and SPYER, K.M (1979). Preganglionic vagal cardio­ motor neurones in the rabbit : location and properties. J.Physiol. 296 : 20-21P. 209

JORDAN, D. and SPYER, K.M (1977). Studies on the termination of sinus nerve afferents. Pflugers Arch. 369 : 65 - 73.

JORDAN, D. and SPYER, K.M (1978). The distribution and excitability of myelinated aortic nerve afferent terminals. Neuroscience Letters 8 113 - 117.

JORDAN, D. and SPYER, K.M (1979). Studies on the excitability of sinus nerve afferent terminals. J. Physiol. 297 : 123 - 134.

KALIA, M. and WELLES, R.V (1980). Brain stem projections of the aortic nerve in the cat : a study using tetramethyl benzidine as the substrate for horseradish peroxidase. Brain Res. 188 : 23 - 32.

KATONA, P.G., POITRAS, J.W., BARNETT, G.O. and TERRY, B.S. (1970). Cardiac vagal efferent activity and heart period in the carotid sinus reflex. Am. J. Physiol. 218 : 1030 - 1037.

KATZ, R.L. and BIGGER, J.T (1970). Cardiac arrhythmias during anesthesia and operation. Anesthesiology 33 : 193 - 214.

KENDRICK, J.E., MATSON, G.L., OBERG, B. and WENNERGREN, G (1973). The effect of stimulus pattern on the pressure response to electrical stimulation of the carotid sinus nerve of cats. Proc. Soc. Exp. Biol. Med. 144 : 412 - 416.

KERR, F.W.L (1962). Facial, vagal and glossopharyngeal nerves in the cat. Afferent connections. Arch. Neurol. 6 : 264 - 281.

KERR, F.W.L (1969). Preserved vagal visceromotor function following destruction of the dorsal motor nucleus. J.Physiol. 202 : 755 - 769.

KIRCHHEIM, H., GROSS, R. and BRANDSTETTER, K (1974). Effect of the respiratory cycle on the carotid sinus reflex in the unanaesthetised dog. Arch. ges. Physiol. 347 : R.46.

KNOWLTON, G.C. and LARRABEE, M.G (1946). A unitary analysis of pulmonary volume receptors. Am. J. Physiol. 147 : 106 - 114.

KOCH, E (1929). Z. Kreisl. Forsch. 22 : 220. Cited in Reflexogenic Areas of theCardiovascular System, ed. C. Heymans and E.Neil. J & A Churchill Ltd., London. 210

KOCH, E (1931). Die reflekforische selbsteuerung des kreislaufes, Steinkopff, Keipzig. Cited in Reflexogenic Areas of the Cardiovascular System, ed. C. Heymans and E. Neil. J & A Churchill Ltd., London.

KOELLE, G.B (1975). Neuromuscular blocking agents. In: The Pharmacological Basis of Therapeutics. 5th edition, ed. L.S Goodman and A.Gillman. Macmillan Publishing Co. Inc., N.Y.

KOEPCHEN, H.P., WAGNER, P.H. and LUX, H.D (1961). Uber die zusammenh~nge zwischen zentraler erregbarkeit, reflektorischem tonus und atemrhythmus bei der nerv~sen steuerung der herzfrequenz. Pflugers Arch. ges. Physiol. 273 : 443 - 465.

KOIZUMI' K. ' SELLER' H. ' KAUFMAN' A. and BROOKS' C. Mee (1971). Pattern of sympathetic discharges and their relation to baroreceptor and respiratory activities. Brain Res. 27 281 - 294.

KORDY, M.T., NEIL, E. and PALMER, J.F (1975). The influence of laryngeal afferent stimulation on cardiac vagal responses to carotid chemoreceptor excitation. J.Physiol. 247 24 - 25P. -

KORNER, P.I (1965). The role of the arterial chemo­ receptors and baroreceptors in the circulatory response to hypoxia of the rabbit. J.Physiol. 180 : 279 - 303.

KORNER, P.I (1970). Central nervous control of autonomic function - possible implications in the pathogenesis of hypertension. Circulation Res. 26/27 (suppl. 2) 159 - 168.

KORNER, P.I (1971). Integrative neural cardiovascular control. Physiol. Rev. 51 : 312 - 367.

KORNER, P.I (1979). Central nervous control of autonomic cardiovascular system. Handbook of Physiology - The Cardiovascular System I. Ch. 20, 691 .... 739. American Physiological Society, Bet.hesda.

KORNER, P.I., LANGSFORD, G., STARR, D., OTHER, J.B., WARD, W. and WHITE, S.W (1968). The effects of chloralose-urethane and sodium pentobarbitone anaesthesia on the local and autonomic components of the circulatory response to arterial hypoxia. J. Physiol. 199 : 283 - 302. KORNER, P.I., SHAW, J., WEST, M.J. and OLIVER, J.R (1972). Central nervous system control of baro­ receptor reflexes in the rabbit. Circulation Res. 31 : 637 - 652. 211

KORNER, P.I., SHAW, J., WEST, M.J., OLIVER, J.R. and HILDER, R.G (1973). Integrative reflex control of heart rate in the rabbit during hypoxia and hyperventilation. Circulation Res. 33 : 63 - 73.

KORNER, P.I. and UTHER, J.B (1970). Stimulus-cardio­ respiratory effector response profile during arterial hypoxia in the unanaesthetized rabbit. Aust. J. Exptl. Biol. Med. Sci. 48 : 663 - 685. KORNER, P.I., UTHER, J.B. and WHITE, S.W (1968). Circulatory effects of chloralose-urethane and sodium pentobarbitone anaesthesia in the rabbit. J.Physiol. 199 : 253 - 265. KORNER, P.I., UTHER, J.B. and WHITE, S.W (1969). Central nervous integration of the circulatory and respiratory responses to arterial hypoxemia in the rabbit. Circulation Res. 24 : 757 - 776.

KORNER, P.I., WEST, M.J. and SHAW, J (1973). Central nervous resetting of baroreceptor reflexes Aust. J. Expt 1. Biol. Med. Sci. 51 : 53 - 64.

KUMADA, M., NOGAMI, K. and SAGAWA, K (1975). Modulation of carotid sinus baroreceptor reflex by sciatic nerve stimulation. Am. J. Physiol. 93 : 1535 - 1541.

KUNZE, D.L (1972). Reflex discharge patterns of cardiac vagal efferent fibres. J.Physiol. 222 1 - 15. LEVY, M.N., De GEEST, H. and ZIESKE, H (1966). Effects of respiratory centre activity on the heart. Circulation Res. 18 : 67 - 78. LEVY, M.N., IANO, T. and ZIESKE, H (1972). Effects of repetitive bursts of vagal activity on heart rate. Circulation Res. 30 : 186 -194.

LEVY, M.N. and MARTIN, P.J (1979). Neural control of the heart. Ch. 16, 581 - 620. In : Handbook of Physiology : The Cardiovascular System I - American Phsyiological Society, Bethesda. LINDEN, R.J (1973). Function of cardiac receptors. Circulation. 48 : 463 - 480.

LIPSKI, J., McALLEN, R.M. and SPYER, K.M (1975). The sinus nerve and baroreceptor input to the medulla of the cat. J. Physiol. 251 : 61 - 78. LIPSKI, J., McALLEN, R.M. and TRZEBSKI, A (1976). Carotid baroreceptor and chemoreceptor inputs onto single medullary neurones. Brain Res. 107 : 132 - 136. 212

LIPSKI, J. and TRZEBSKI, A (1975). Bulbo-spinal neurons activated by baro receptor afferents and their possible role in inhibition of preganglionic sympathetic neurones. Pflugers Arch. ges. Physiol. 356 : 181 - 192. LOEB, J.M. and VASSALLE, M (1978). An analysis of cholinergic involvement in postvagal tachycardia. Cardiology 63 : 270 - 279. LOEW!, 0 (1921). Uber humorale ubertragbarkeit der herznervenwirkung. Pflugers Arch. ges. Physiol. 189 : 239 - 242. LoPES, O.U. and PALMER, J.F (1976). Proposed respiratory 'gating' mechanism for cardiac slowing. Nature 264 : 454 - 456. LOPES, O.U. and PALMER, J.F (1977). Mechanism of hypothalamic control of cardiac component of sinus nerve reflex. Quart. J. Exp. Physiol. 63 : 231 - 254. LUMBERS, E.R., McCLOSKEY, D.I. and POTTER, E.K (1979). Inhibition by angiotensin II of baroreceptor­ evoked activity in cardiac vagal efferent nerves in the dog. J. Physiol. 294 : 69 - 80. MACLEOD, R.D.M. and SCOTT, M.J (1964). The heart rate responses to carotid body chemoreceptor stimulation in the cat. J.Physiol. 175 193 - 202. MATTHES, K. and EBELING, J (1948). Untersuchungen Uber die atemschwankungen des blutdrucks und der pulsfrequenz beim menschen. Pflugers Arch. ges. Physiol. 250 : 747 - 768. McALLEN, R.M., JORDAN, D. and SPYER, K.M (1980). Carotid baroreceptor input to the cat's brain - where is the first synapse? In : Central Interaction between Respiratory and Cardiovascular Control Systems, ed. H.P Koepchen, S.M Hilton, and A. Trzebski, Springer-Verlag, Berlin. 87 - 92. McALLEN, R.M. and SPYER, K.M (1972). 'Baroreceptor' neurones in the medulla of the cat. J.Physiol. 222 : 68 - 69 P. McALLEN, R.M. and SPYER, K.M (1976). The location of cardiac vagal preganglionic notoneurones in the medulla of the cat. J.Physiol. 258 187 - 204. McALLEN, R.M. and SPYER, K.M (1978 a). Two types of vagal preganglionic motoneurones projecting to the heart and lungs. J. Physiol. 282 : 353 - 364. 213

McALLEN, R.M. and SPYER, K.M (1978b). The baroreceptor input to cardiac vagal notoneurones. J.Physiol. '2'82:.,...... , 365.., 374, McCLOSKEY, D.I. and POTTER, E.K (1981). Excitation and inhibition of cardiac vagal motoneurones by electrical stimulation of the carotid sinus nerve. J.Physiol. accepted for publication Dec. 1980.

McRITCHIE, R.J. and WHITE, S.W (1974). Role of trigeminal, olfactory, carotid sinus and aortic nerves in the respiratory and circulatory response to nasal inhalation of cigarette smoke and other irritants in the rabbit. Aust. J. Exptl. Biol. Med. Sci. 52 127 - 140.

MELCHER, A (1976). Respiratory sinus arrhythmia in man effects of carotid sinus baroreceptor stimulation. Acta physiol. scand., supplementum 435.

MIDDLETON, S., MIDDLETON, H.H. and GRUNDFEST, H (1950). Spike potentials and cardiac effects of mammalian vagus nerve. Am. J. Physiol. 162 : 553 - 559.

MILNOR, W.R (1980). Cardiovascular physiology. Ch. 35. In:Medical Physiology, 14th edition, ed. V.B Mountcastle. The C.V Mosby Co., St. Louis.

MITCHELL, G.A.G. and WARWICK, R (1955). The dorsal vagal nucleus. Acta anat. 25 : 371 - 395.

MIURA, M. and REIS, D.J (1968). Electrophysiological evidence that carotid sinus nerve fibres terminate in the bulbar reticular formation. Brain Res. 9 : 394 - 397.

MIURA, M. and REIS, D.J (1969). Termination and secondary projections of carotid sinus nerve in the cat brain stem. Am. J. Physiol. 217 : 142 ~ 153.

MIURA, M. and REIS, D.J (1972). The role of the solitary and paramedian reticular nuclei in mediating cardiovascular reflex responses from carotid baro- and chemoreceptors. J.Physiol. 223 525 - 548.

MOREST, D.K (1967). Experimental study of the projections of the nucleus of the tractus solitarius and the area postrema in the cat. J. Comp. Neurol. 130 : 277 - 300.

MUERS, M.F. and SLEIGHT, P (1972). Action potentials from ventricular mechanoreceptors stimulated by occlusion of the coronary sinus in the dog. J. Physiol. 221 : 283 - 309. 214

MILLER, F.R. and SHERRINGTON, C.S (1916). Some observations on the buccopharyngeal stage of reflex deglutition in the cat. Quart. J.exp. Physiol. 9 : 147 - 186.

NOSAKA, S., YAMAMOTO, T. and YASUNAGA, K (1979). Localization of vagal cardioinhibitory preganglionic neurons within rat brain stem. J. Comp. Neurol. 186 : 79 - 92. OBERG, B. and THOREN, P (1972). Studies on left ventricular receptors, signalling in non­ medullated vagal afferents. Acta physiol. scand. 85 : 145 - 163.

OBERG, B. and WHITE, S (1970). The role of vagal cardiac nerves and arterial baroreceptors on the circulatory adjustments to hemorrhage in the cat. Act a physiol. scand. 80 : 395 - 403.

OKADA, H., OKAMOTO, K. and NISIDA, I (1961). The activity of the cardioregulatory and abdominal sympathetic nerves during swallowing. Jap. J. Physiol. 11 : 44 - 53.

PAINTAL, A.S (1954). The response of gastric stretch receptors and certain other abdominal and thoracic vagal receptors to some drugs. J.Physiol. 126 : 271 - 285.

PAINTAL, A.S (1973). Vagal sensory receptors and their reflex effects. Physiol. Rev. 53 : 159 - 227.

PELLETIER, C.L. and SHEPHERD, J.T (1973). Circulatory reflexes from mechanoreceptors in the cardio­ aortic area. Circulation Res. 33 : 131 -138.

PELLETIER, C.L. and SHEPHERD, J.T (1975). Effect of hypoxia on vascular responses to the carotid baroreflex. Am. J. Physiol. 228 : 331 - 336.

POTTER, E.K (1980). Excitation and inhibition of cardiac vagal motoneurones by stimulation of the carotid sinus nerve. J.Physiol. July 1980 meeting, Abstr~ct 59. POTTER, E.K (1981). Inspiratory inhibition of vagal responses to baroreceptor and chemoreceptor stimuli in the dog. J.Physiol. accepted for publication Dec. 1980.

POTTER, E.K. and McCLOSKEY, D.I (1979). Respiratory stimulation by angiotensin II. Respiration Physiology 36 : 367 - 373. PRZYBYSZEWSKI, A. and TRZEBSKI, A (1980). The role of arterial baroreceptors in respiratory modulation of heart rate. Proceedings of the International Union of Physiological Sciences, Vol. 14., Abstract no. 2847. 215

REID, J.V~O (1969). Cardiac pacemaker : Effects of regularly spaced nervous input. Am. Heart J. : ..-.-78 : 58 ,:-. 64 . RICHET, C (1899). De la resistance de carrards l l'asphyxie. J. Physiol. Path. g~n. 1 641 - 650. Cited by Andersen, H.T (1963) see above.

RICHTER, D.W., KECK, W. and SELLER, H (1970). The course of inhibition of sympathetic activity during various patterns of carotid sinus nerve stimulation. Pflugers Arch. 317 110 - 123. -

RIJLANT, P ( 1936 a). Le cont roleA extrinseque' de la . frequence du battement cardiaque. Comptes rendus de la societe de Biologie. 123 : 99 - 100.

RIJLANT, P (1936b). L'arythmie cardiaque respiratoire. Comptes rendu de la societe de Biologie. · 123 : 997 - 1001. -

SALISBURY, P.F., GALLETTI, P-M., LEWIN, R.J. and RIEBEN, P.A (1959). Stretch reflexes from the dog's lung to the systemic circulation. Circulation Res. 7 : 62 - 67. SALMOIRAGHI, G.C (1962). 'Cardiovascular' neurones in the brain stem of cat. J. Neurophysiol. 25 182 - 197. SAMAAN, A (1935). The antagonistic cardiac nerves and heart rate. J.Physiol. 83 : 332 - 340. SAMPSON, S.R. and BISCOE, T.J (1968). Electrical potentials evoked in the brain stem by stimulation of the sinus nerve. Brain Res. 9 : 398 - 402. SAMPSON, S.R. and VIDRUK, E.H (1975). Properties of 'irritant' receptors in canine lung. Respiration Physiol. 25 : 9 - 22.

SCHER, A.M. and YOUNG, A.C (1970). Reflex control of heart rate in the unanesthetized dog. Am. J.Physiol. 218 : 780 - 789. SCHNEIDER, E.C (1930). Observations on holding the breath. Am. J. Physiol. 94 : 464 - 470. SCHOLANDER, P.F., HAMMEL, H.T., Le MESSURIER, H., HEMMINGSEN, E. and GAREY, W (1962). Circulatory adjustments in pearl divers. J. appl. Physiol. 17 : 184 - 190. SCHWEITZER, A (1937). Die Irradiation Autonomer Reflexe. Karger : Basle. 216

SCOTT, J.C. and REED, E.A (1955). Comparison of unilateral with bilateral stimulation of the carotid sinus cardioinhibitory reflex. Am. J. Physiol. 181 : 27 - 29.

SEDIN, G (1976). Responses of the cardiovascular system to carotid sinus nerve stimulation. Upsala J. Med. Sci. 81 : 1 - 17.

SELLER, H. and ILLERT, M (1969). The localization of the first synapse in the carotid sinus baroreceptor reflex pathway and its alteration of the afferent input. Pflugers Arch. ges. Physiol. 306 : 1 - 19.

SELLER, H., LANGHORST, P., RICHTER, D. and KOEPCHEN, H.P (1968). Uber die abh~ngigkeit der pressore­ ceptorischen hemmung des sympathicus von der atemphase und ihre answirkung in der vasomotorik. Pflugers Arch. ges. Physiol. 302 : 300 - 314.

SLEIGHT, P (1964). A cardiovascular depressor reflex from the epicardium of the left ventricle in the dog. J.Physiol. 173 : 321 - 343.

SLEIGHT, P (1976). Possible physiological stimuli for ventricular receptors and their significance in man. In : Cardiac Receptors, ed. R. Hainsworth, C. Kidd and R.J Linden. Cambridge University Press, Cambridge. Ch. II, p.241.

SLEIGHT, P. and WIDDICOMBE, J.G (1965). Action potentials in fibres from receptors in the epicardium and myocardium of the dog's left ventricle. J.Physiol. 181 : 235 - 258.

SMITH, R.S. and PEARCE, J.W (1961). Microelectrode recordings from the region of the nucleus solitarius in the cat. Canad. J. Biochem. 39 : 933 - 939.

SMYTHE, H.S., SLEIGHT, P. and PICKERING, G.W (1969). A quantitative method of assessing baroreflex sensitivity. Circulation Res. 24 : 109 -121.

SPYER, K.M (1979). Baroreceptor control of vagal preganglionic activity. In : Integrative Functions of the Autonomic Nervous System, ed. C. McC. Brooks, K. Koizumi and A. Sato. Elsevier, North Holland Biomedical Press, Amsterdam. pp. 283 - 292. SPYER, K.M. and McALLEN, R.M (1980). Interaction of central and peripheral inputs onto vagal cardiomotor neurones. In : Central Interaction between Respiratory and Cardiovascular Control Systems, ed. H.P Koepchen, S.M Hilton and A. Trzebski. Springer-Verlag, Berlin. pp. 8 - 14. 217

SPYER, K.M. and WOLSTENCROFT, J.H (1971). Problems of the afferent input to the paramedian reticular nucleus and the central connections of the sinus nerve. Brain Res. 26 : 411 - 414.

STREATFEILD, K.A., DAVIDSON, N.S. and McCLOSKEY, D.I (1977). Muscular reflex and baroreflex influences on heart rate during isometric contractions. Cardiovascular Res. 11 87 - 93.

SUGIMOTO, T., ITOH, K., MIZUNO, N., NOMURA, S. and KONISHI, A (1979). The site of origin of cardiac preganglionic fibers of the vagus nerve : an HRP study in the cat. Neuroscience Letters 12 : 53 - 58.

SUMI, T (1963). The activity of brain-stem respiratory neurones and spinal motoneurones during swallowing. J. Neurophysiol. 26 : 466 - 477.

SZENTAGOTHAI, J (1952). The general visceral efferent column of the brain stem. Acta Morphologica (Budapest) 2 : 313 - 328.

TAFIL, M. and TRZEBSKI, A (1980). Contribution of the central respiratory drive and of the arterial chemoreceptor reflex to the mechanism of respiratory sinus arrhythmia in man. Pro­ ceedings of the International Union of Physiological Sciences. Vol. 14, Abstract no. 3357.

THAMES, M.D. and KONTOS, H.A (1970). Mechanisms of baroreceptor-induced changes in heart rate. Am. J. Physiol. 218 : 251 - 256.

THOMAS, M.R. and CALARESU, F.R (1974). Localization and function of medullary sites mediating vagal bradycardia in the cat. Am. J. Physiol. 226 : 1344 - 1349.

THOREN, P (1976). Reflex effects of left ventricular mechanoreceptors with afferent fibres in the vagal nerves. In : Cardiac Receptors, ed. R. Hainsworth, C. Kidd and R.J Linden. Cambridge University Press, Cambridge. Ch. 12, p.259.

THOREN, P (1979). Role of cardiac vagal C-fibres in cardiovascular control. Rev. Physiol. Biochem. Pharmacol. 86 : 1 - 94. THRON, H.L., BRECHMANN, W., WAGNER, J. and KELLER, K (1967). Quantitative untersuchungen uber die bedeutung der gef~ssdehnungsreceptoren im rahmen der kreislaufhomoiostase beim wachen menschen. Pflugers Arch. ges. Physiol. 293 : 68 - 99. 218

TRZEBSKI, A (1980). In : Central Interaction between Respiratory and Cardiovascular Control Systems ed~. H~P Koepchen, S.M Hilton and A. Trzebski, Springer-Verlag, Berlin. pp. 127, 195 and 212.

TRZEBSKI , A. , LIPSKI , J. , MAJCHERCZYK, S. , SZULCZYK, P. and CHRUSCIELEWSKI, L (1975). Central organization and interaction of the carotid baroreceptor and chemoreceptor sympathetic reflex. Brain Res. 87 : 227 - 237.

TRZEBSKI , A. , PETERSON, L. H. , ATTINGER, F. , JONES, A. and TEMPEST, R (1962). Unitary responses in the medulla oblongata related to carotid sinus receptor functions. Physiologist (Washington) 5 : 222.

WALLACH, J.H. and LOEWY, A.D (1980). Projection of aortic nerve afferents to the nucleus tractus solitarius in the rabbut. Fed. Proc. 39 : 842, abstract no. 3032.

WARNER, H.R. and COX, A (1962). Mathematical model of heart rate control by sympathetic and vagus efferent information. J. Appl. Physiol. 17 : 349 - 355.

WARNER, H.R. and RUSSELL, R.O (1969). Effect of combined sympathetic and vagal stimulation on heart rate in the dog. Circulation Res. 24 : 567 - 573.

WEBER, E.F.W. and WEBER, E.H (1845). Experimenta physiologica in theatro anatomico. Annali Universali di Medicina, Milano. 1845. 116 225 - 233. esp. Experimenta, quibus -­ probatur nervos vagos rotatione machinae galvano-magneticae irritatos, motum cordis retardare et adeo intercipere, pp. 227 - 228.

WIDDICOMBE, J.G (1974). Reflex control of breathing. In : MTP International Review of Science. Physiology Series One, Vol. 2 : Respiratory Physiology, ed. Widdicombe J.G. University Park Press, Baltimore, Maryland.

GOODMAN, N.W. (1973). Efferent control of arterial chemoreceptors mediated by glossopharyngeal fibres and artifacts introduced by stimulation techniques. J. Physiol. 230 : 295 - 311.