Control of Breathing During Sleep and Anesthesia

Control of Breathing During Sleep and Anesthesia Edited by W. A. Karczewski P. Grieb and Joanna Kulesza Polish Academy of Sciences Warsaw, Poland

and G. Bonsignore University of Palermo Palermo, Italy

Springer Science+ Business Media, LLC

Library of Congress Cataloging in Publication Data

International Symposium on Control of Breathing during Sleep and Anesthesia (1987: Warsaw, Poland)

Control of breathing during sleep and anesthesia / edited by Witold A. Karczewski ... [et al.]

p. cm. "Proceedings of the International Symposium on Control of Breathing during Sleep

and Anesthesia, held September 10-12, 1987, in Warsaw, Poland" - T.p. verso. Indudes bibliographies and index. ISBN 978-1-4757-9852-4 ISBN 978-1-4757-9850-0 (eBook) DOI 10.1007/978-1-4757-9850-0 1. Sleep apnea syndromes-Congresses. 2. Respiration-Regulation-Congresses. 3.

Anesthesia-Congresses. 4. Lungs-Diseases, Obstructive-Congresses. 1. Karczewski, Witold A. II. Title. [DNLM: 1. Anesthesia-congresses. 2. Respiration-drug effectscongresses. 3. Respiratory System-physiology-congresses. 4. Sleep-drug effectscongresses. 5. Sleep-physiology-congresses. 6. Sleep Apnea Syndromes-congresses. WF 102 1612c 1987] RC737.5.157 1987 616.2-dcI9 DNLM/DLC for Library of Congress

Proceedings of the International Symposium on Control of Breathing during Sleep and Anesthesia, held September 10-12, 1987, in Warsaw, Poland

© 1988 Springer Science+Business Media New York Originally published by Plenurn Press, New York in 1988 Softcover reprint of the hardcover 1 st edition 1988 AII rights reserved

88-19689 CIP

No part of this book may be reproduced, stored in a retrieval system, or transmitted in any form or by any means, electronic, mechanical, photocopying, microfilming, recording, or otherwise, without written permis sion from the Publisher

PREFACE

Contrary to the popular belief, "Le sommeil n'est plus milieu s(lr" (J. Cocteau, cf. Cl. Gaultier, Pathologie respiratoire du sommeil, La Presse Medicale, 16, 561-563, 1987), and anesthesia is even less safe. Sudden Infant Death Syndrome, Obstructive Sleep Apnea, Ondine's Curse and various respiratory complications of general anesthesia are not so rare; as a matter of fact they happen much too frequently.

The idea of organizing another symposium dealing with breathing in sleep and anesthesia has been discussed almost immediately after we said "good bye" to the Organizers of the excellent Paris meeting "The Regulation of Respiration during Sleep and Anesthesia" (R.S. Fitzgerald, H. Gautier, S. Lahiri eds., Advances in Experimental Medicine and Biology, vol. 99, Plenum, New York 1978).

Taking into account the impressive amount of data that have emerged during the last few years, we have decided that we shall meet and discuss them; we hoped also that the publication of the scientific material might be useful for everybody interested in the physiology and pathophysiology of breathing, anesthesia and sleep. So we met in Warsaw under the auspices of the European Society for Clinical Respiratory Physiology and the Polish Academy of Sciences, we discussed vividly many fascinating papers presented by our Colleagues from Europe and America and Plenum Press has published the proceedings. I hope that the final result will satisfy the reader.

Witold A. Karczewski

v

CONTENTS

Opening Remarks . . W.A. Karczewski

Sleep as a Physiological Phenomenon .........•.... J. Nar~bski

Heavy Snorers Disease • . . . . . . . . . . E. Lugaresi, F. Cirignotta, G. Coccagna, P. Montagna, M. Zucconi

Hypoxia During Sleep. . . . . ......... . G. Bonsignore, 0. Marrone, V. Bellia, F. Cibella

The Upper Airway ~luscles: Their Role in Sleep-Related Respiratory Dysrhythmias ........... .

N.S. Cherniack, D.W. Hudgel

Effectiveness and Side-Effects of Nasal Continuous Positive Pressure Therapy in 66 Patients with Sleep Apnea . . . . . . . . . . . . . . . . .

H. Becker, U. Koehler, J.H. Peter, M. Steinberg, P.von Wichert

The Influence of Obesity on Disordered Breathing in Patients with Obstructive Sleep Apnea Syndrome /OSAS/ .................... .

A. Brzecka

Chronic Obstructive Pulmonary Disorders (COPD) and Sleep . . . ........... .

A. Gianotti, P. Moscatelli, N. Franconieri

Polysomnographic Findings in Patients with Chronic Obstructive Pulmonary Disease (COPD) .....

E. Gozlikirmizi, N. Yildirim, H. Kaynak, S. Madazioglu, H. Denktas, F. Yenel

High and Fluctuating Muscle Nerve Sympathetic Activity in the Sleep Apnea Syndrome: A Pathogenetic Mechanism in the Development of Hypertension ?

J. Hedner, J. Sellgren, H. Ejnell, G. Wallin

Cephalometry for Evaluation of Geometry of the Upper Airway ................... .

A. Kukwa, B. de Berry-Borowiecki, R.M.I. Blanks, I. Fleszar, A. Komorowska, M. Ryba

1

3

5

19

29

39

43

47

49

55

61

vii

The Clinical Relevance of Very Severe Pure Diaphragm Weakness ............... .

C. Laroche, N. Carroll, A. Mier, C. Brophy, M. Green

Cardiac Involvement in Obstructive Sleep Apnea Syndrome (OSAS) - Case Report ...... .

M. MigdaJ, K. Kubicka, W. Kawalec, L. OrYowski, M. Zubrzycka, P.S. Gutkowski

RespiratoryPatterning and Arterial Oxygenation During Sleep in Laryngectomised Patients . . . . . .

W. Oldfield, L. Sawicka, M.S. Meah, W.N. Gardner

Vagal Reactivity During Sleep: Study of the Oculocardiac and the Hering-Breuer Reflex in Preterms ................. .

J. Ramet, J.P. Praud, A.M. D'Allest, A. Carofils, M. Dehan, Cl. Gaultier

Resumption of Ventilation Sleep Apneas is not

at the End of Obstructive Determined by Diaphramatic

Fatigue ..... . S. Sanci, F. Cibella, R. Modica, S. Romano,

0. Marrone, G. Cuttitta, V. Belli.a

The Effect of Chronic Pulmonary Denervation on the Pattern of Breathing During Sleep in Man . .

S.A. Shea, R.L. Horner, E. McKenzie, N.R. Banner, M.H. Yacoub, A. Guz

Conchoplasty in the Treatment of the Obstructive Sleep Apnea Syndrome . . . . . . . . . . . . . . . . .

H. Skarzynski, W. Jeglinski, A. Kt~wa, G. Opolski, M. Ryba, z. Szlenk, P. Radzimowski, M. Lisicka

The Effects of Adeno- and Tonsillectomy in Children with Sleep Apnea Syndrome ........... .

H. Skarzynski, A. Kukwa, G. Opolski, R. Krauze, W. Jeglinski, K. SYomka, A. Kalotka-Bratek

Surgery Therapy for Obstructive Sleep Apnea - Present and Future .................. .

B. de Berry-Borowiecki

Anesthesia and Central Nervous System B.K. Siesjo

Respiration and Anesthesia . . . . . . . . . . . . . . . . . . B. Kaminski

Control of Breathing by Neuropeptides ..... M.P. Morin-Surun, J. Champagnat, A. Foutz, M. Denavit-Saubie

Neurotransmission and Neuromodulation Involved in the Control of Respiration . . . . . . . . . . . . .

P. Grieb

The Control of Breathing Movements in the Fetus G.S. Dawes

viii

65

67

73

77

83

89

95

97

99

105

107

111

119

133

Pathophysiology of Sudden Infant Death Syndrome ....... . Cl. Gaultier

Depression of Excitatory Amino Acid Neurotransmitters in Brain During Carbon Dioxide Narcosis, Anesthesia and Hypoxia: Glutamic Acid and Aspartic Acid .

R.E. Dutton, P.J. Feustel, E.H. Dutton, A. Szema, V.E. Shih, P.M. Renzi, G.D. Renzi

Respiratory Actions of Cholecystokinin and its Interaction with Opioids at the Brainstem Level .....

M.A. Hurle, M.M. Dierssen, M.P. Morin-Surun, J. Fl6rez

Influence of the Neurohormones: Hwnan Corticotropin Releasing Factor (hCRF), Thyrotropin Releasing Hormone (TRH), Adrenocorticotrophic Hormone (ACTH) on Ventilation in Humans . . . . . . . . . . . . .

M. Nink, I. Huber, U. Krause, H. Lehnert

Tube Breathing under General Anesthesia ....... . V. Smejkal, F. Palecek, R. Havelka, D. Miloschewsky

Effects of Enflurane on the Ventilatory Response to Increased Carbon Dioxide and Metabolic Rate in Dogs ................ · · ·

D.S. Ward, R. Ginsburg, I.H. Abdul-Rasool, K. Aqleh

Effects of Changes in Inspiration Volume and Flow Rate on Respiratory Activity ........... .

M.D. Altose, A.F. Connors Jr., A.F. DiMarco

Excitatory Effects of Electrical and Chemical Stimulation of the Botzinger Complex on Expiratory Activity in the Cat ..

F. Bongianni, G. Fontana, T. Pantaleo

Trigeminal Nerve, Breathing and Sleep Apnea ...... . H. Gromysz, A. Kukwa, U. Jernajczyk, W.A. Karczewski

Lateralized Response of the Hypoglossal, Facial and Phrenic Nerves to Lung Inflation . . . . . . .

W.A. Janczewski

Laryngeal Contribution to Respiratory Pattern in Nlesthetized Rabbits ........ .

B. Kamosinska, M. Szereda-Przestaszewska

Propriospinal Inspiratory Neurons in the Upper Cervical Spinal Cord of the Rabbit: Location and Efferent Spinal Projections . . . . . . . . . . . . . . . .

L. Kubin, J.R. Romaniuk

Spectral Analysis of Breathing Pattern in Man During Exercise . . . . . . . . . . . . . . . . . .

J. Siegelova, S. Feitova

Disturbed Patency of the Upper Airway and Its Consequences . . . . . . . . . . . . . . .

H. Skar~ynski, W. Jeglinski, A. Kukwa, G. Opolski, K. SJomka, ~. Ryba, R. Krauze

141

149

155

159

163

167

173

177

183

187

193

197

203

207

ix

The Vagus Nerve and the Control ot Breatning During Postnatal Development ............ . 209

D. Marlot

Neural Mechanisms That Lead to Apnea . . . . . 217 A.I. Pack, L.R. Kline, J.C. Hendricks, M.F. Cola

Comparison of the Importance of Mechanisms Taking Part in Self-Regulation of Breathing ........ . 227

V.D. Glebovski

Closing Remarks .. 239 W.A. Karczewski

PARTICIPANTS 241

INDEX 243

X

OPENING REMARKS


Department of Neurophysiology, Polish Academy of Sciences Medical Research Centre, Warsaw, Poland

In an ancient text written in China about 3000 years ago and called "The Yellow Emperor's Canon of Internal Medicine" we can encounter the following dialogue:

"The Yellow Emperor said: 'If one is afflicted with abnormal respiration, one cannot sleep and one's breathing has noise; ( ... )What organs in the body cause this? I desire to hear about the etiology.'

Ch'i Po answered: 'Those who cannot sleep and breathe with noise have disorders in the Yang Ming. Usually the three Yang of the foot travel downward. In these cases they travel upward, hence the noise ... '

The Yellow Emperor said: 'Excellent'."

(cf. Kao, F.F., 1975, Respiratory Research in the People's Republic of China, Geographic Health Studies, John E. Fogarty Int. Center, U.S. DREW Publication, Washington D.C., pp.31-32).

One can very easily imagine a similar dialogue being held today; obviously, we wouldn't use the notion of Yang Ming to explain the mechanisms of breathing disturbances during sleep, and since we are usually very critical of other people's work nobody would say "excellent" listening to his or her colleague's interpretation. Otherwise, however, I am not sure that we would be prepared to give more precise answers to the Yellow Emperor's question.

!'lore or less the same applies to the effects of various anaesthetics, sedatives and narcotics on the respiratory pattern generator; respiratory neurophysiologists would be grateful if somebody would be kind enough to explain for example why the depressant effect of one group of drugs (e.g. volatile anaesthetics) consists in a rapid and shallow breathing, whereas another drug - or group of drugs - will slow breathing with less effect on tidal volume (cf. Pavlin, E.G., and T.F. Hornbein, 1985, Anesthesia and the control of ventilation, in: "Handbook of Physiology - The Respiratory System II", Chapter 25, Amer. Physiol. Soc., Bethesda, MD.). The problem is quite serious since anaesthesia is an inherent part of our present paradigm; moreover, we are sometimes trying to use anaesthetics or narcotics as experimental tools; perhaps we shouldn't.

There ar·e many more questions and I hope that this Symposium will give us answers to at least some of them. I am sure also that the papers

and discussions will focus not only on symptoms, diagnosis and treatment, but also on physiological mechanisms.

Finally, I am entitled by Professor J. K. Kostrzewski, the President of the Polish Academy of Sciences to welcome you on his behalf. Being presently on a service trip to China he was not able to attend this meeting personally, but, before leaving, wished us a successful conference. Professor M. J. Mossakowski, the Scientific Secretary of the Medical Section of the P.A.S. is with us today and I know that - as a neuroscientist -he is joining us in our scientific expectations.

Last, but not least: I am sure that I will express the feelings of all members of the local organizing committee by saying that we are most happy to have you all here in Warsaw. You certainly know that it is never easy to organize a really good scientific meeting; it is particularly difficult after the magnificent June Congress of the SEPCR in Antwerp, which many of us have attended. Since we have our own very special problems, the conditions may be a bit spartan, and improvisation - Polish specialite de la maison- may be sometimes too obvious. I hope, however, that if we all try very hard we shall have a good and fruitful conference. The presence of so many experts in the field is a sufficient guarantee of the final success.

2

SLEEP AS A PHYSIOLOGICAL PHENOMENON

ABSTRACT*

Juliusz Nanbski

Department of Physiology, Medical School Bydgoszcz, Poland

Sleep appeared at almost the same stage of evolution as endothermy. It occupies more than one fourth of humans', mammals', and birds' life. Its crucial symptoms are unconsciousness and immobility. Therefore sleep is a curious phenomenon because it is inseparably related to the reachest behavionll possibility of highly developed brain. "If sleep does not serve an absolutely vital function it is the biggest mistake the evolutionary process has ever made" (A. Rechtschaffen, 1971). Unfortunately this ftmction up to date is obscure.

EEG is the basis of sleep diagnosis. From the physiological standpoint EEG is only an epiphenomenon of brain processes during sleep. EEG enables the documentation of sleep onset, sleep termination, and the consecutive ultradi<m outcome of sleep dichotomy: NREM-REM cycles. The computer spectral EEG analysis enables the qualitative estimation of the intensity of NREM sleep process jointly with sleep profile expressed as a hypnogram.

Sleep is the particular state only of the brain. The rest of the body during sleep is certainly in a special state during sleep, but it differs only slightly- from the habitual relaxed wakefulness. Sleep may be n<'Hneo the unavoidable "tax" payed to the nature for the top position of the brain on the evolutionary ladder.

The full MS has not been received

3

HEAVY SNORERS DISEASE

E. Lugaresi, F. Cirignotta, G. Coccagna, P. Montagna, and M. Zucconi

Institute of Neurology, University of Bologna Bologna, Italy

Two syndromes characterized by chronic alveolar hypoventilation (CAR) in the absence of pleuro-pulmonal or musculo-skeletal alterations were identified in the Fifties. One, called cardio-respiratory syndrome of obesity or Pickwickian syndrome, was found in severely obese patients; the other, primary, idiopathic or essential CAR, seen in non obese patients was attributed to a reduced excitability of the breathing centers. In the Sixties, polysomnography documented the presence of obstructive apneas during sleep in Pickwickian cases (and also in many non obese patients, until then wrongly classified as primary CAR), and the beneficial effects of tracheostomy demonstrated that obstructive apneas were the cause of the syndrome, called by us Hypersomnia with Periodic Apneas, but better known as Obstructive Sleep Apnea Syndrome (OSAS). Heavy Snorers Disease (HSD), however, is a term which better emphasizes the concept that snoring and obstructive apneas represent just the end-points of the same process.

CLINICAL ASPECTS OF OSAS

Intermittent snoring (when noisy breathing acts alternate with apneas lasting many seconds) and daytime somnolence are the cardinal symptoms of OSAS.

Behavioral disturbances, cognitive impairment, automatic behavior during wakefulness and sleep, headache and confusion on morning awakening, impotence, frequency of micturition and nocturnal enuresis are also found in OSAS. Anoxic convulsive attacks during sleep, induced by very prolonged apneas, were recorded in 3 patients of ours. In advanced cases, breathing becames noisy also during wakefulness and a respiratory flutter is seen on spirometry, possibly indicating increased resistance to air flow during breathing.

CAR is seen only in a small number of patients with OSAS (about 10% of our series). Pulmonary arterial hypertension, cardiomegaly and polycythemia are all consequences of chronic hypoxia. Systemic arterial hypertension is frequently found in OSAS (36% of our cases) irrespective of CAR.

5

Ol

AW

AK

E

Oro-Re

ap.~

cm~O

lntr

a-O

eso

ph

.. .J~

pre

ss.

-:1

Th

or.

Res

p. ~

a

CE

NT

RA

L

~rlH

.l.P...!

!i:...

({Y-~'-·"----i~

b

MIX

ED

O

BS

TR

UC

TIV

E

~~/Mn[

' J1J

W\

~

IN

C"'!'\

0 ...

.--,,f

1111

11,1

111·

"1 :1 ~i-~ff(ffif~~

~~~III~JLWroiVI/vY\IVIJIIIAMMI

V~II'1

t1M.II

Fig

. 1

. R

ecor

ding

s o

f o

ral

resp

irati

on

, in

tra-

oes

op

hag

al

pre

ssu

re a

nd t

ho

racic

re

spir

ati

on

du

rin

g w

akef

uln

ess

and

cen

tral,

m

ixed

an

d o

bst

ruct

ive

apn

eas.

POLYSOMNOGRAPHIC FINDINGS OF OSAS

Apneas may be obstructive (85%), mixed (12-13%) or central (2-3%) (Fig. 1). Central apneas occur during drowsiness and REM sleep and are short (10-20 s). Mixed and obstructive apneas may last 180 s or more. The most prolonged apneas occur during REM sleep and may be so prolonged as to provoke an anoxic attack. When apneas become persistent, they may take up 70% of total sleep time, and even 80% of REM sleep. In mild cases, apneas occur almost only in light (St. I, II) and especially REM sleep, particularly with the patient lying supine. When lying on his side or during deep sleep, the patient may only snore heavily. Apneas have relevant effects:

On sleep: the end of every apnea is accompanied by EEG, vegetative and motor signs of arousal; a true behavioral arousal is exceptional. These repeated arousals, however, disrupt the cyclic pattern of sleep: deep sleep disappears, REM stages become fragmented and may arise directly upon wakefulness.

On blood gases: when the obstructive apneas become persistent, hypoxia and hypercapnia appear and worsen, the longer the apneas. Hypoxia and hypercapnia thus are worst during REM sleep, when apneas are most protracted and breathing least efficient. Sa02 patterns differ, however, according to the severity of the disease. Patients with normal ventilation during wakefulness show only phasic 02 desaturation linked to the apneas. When hypoventilation is evident also during wakefulness, these phasic 02 desaturations are associated with tonic decrements of Sa02 during every episode of REM sleep. Even simple snoring may bring about persistent hypoxia in cases with severe hypoventilation during wake.

On hemodynamics: heart rate slows down during the apneas, and quickens when breathing is resumed. Variations in heart rate may reach 70 beats/minute (Lugaresi et al., 1978). Some patients show sinus, IIdegree-AV blocks and atrial tachycardia (Tilkian et al., 1978). Sudden death during sleep has been attributed to protracted asystolia.

Pulmonary arterial pressure increases progressively during each apnea, to values even higher than 100 mmHg, and decreases when breathing is resumed, never reaching, however, the basal levels. The increase is especially correlated with the degre of hypoxia (Coccagna et al., 1972a).

Systemic arterial pressure oscillates widely, decreasing at the beginning and progressively reaching its highest values at the end of every apnea. It also increases during sleep, reaching the highest levels during REM sleep. It then shows a tendency to stay at elevated levels also during wakefulness. That systemic hypertension is linked to the apneas is shown by the fact that it disappears after tracheostomy (Coccagna et al., 1972b) and after continuous positive airway pressure (CPAP) (Sullivan et al., 1985), which completely abolish the apneas.

NATURAL HISTORY OF OSAS

Patients have all been heavy snorers for a long time (20 years as a mean in our cases) before developing the intermittent snoring and daytime somnolence of OSAS (Fig. 2). Polysomography also suggests a link between snoring and OSAS, since: snoring is an inspiratory noise due to narrowing of the upper airways; heavy and habitual snorers display isolated or clusters of obstructive apneas during light and REM sleep; heavy snoring induces similar, though milder, ventilatory and hemodynamic effects to

7

CD

35

'

I ~OBSERVATION

1 SO

MN

OL

EN

CE

30

BM

I

25

20

20

3

0

40

5

0

ag

e

Fig

. 2.

S

no

rin

g p

rece

des

in

term

itte

nt

sno

rin

g a

nd d

ayti

me

som

nole

nce

by 1

5-2

0 y

ears

(1

30

pati

en

ts).

B

ars

rep

rese

nt

SD;

BMI

=b

od

y m

ass

ind

ex.

OSAS. Thus we regard trivial snoring and the severest OSAS just the endpoints of the same disorder: sleep-related stenosis of the upper airways. In the English literature, onset of OSAS is customarily r elated to the number of apneas (5, or 10 per hour in more conservative estimates) recorded during sleep. We do not rely, however, on these laboratory data. In our opinion, the transition from a preclinical to a true disorder state is marked by the appearance of daytime somnolence. Worsening somnolence also represents the feature most indicative of progression of the disease. The appearance of chronic respiratory insufficiency and its consequences (cardiomegaly, polycythemia, etc . ) marks instead the full development of the disease. These considerations led us to propose the staging of the clinical course of HSD into 4 phases (Lugaresi et al., 1983): stage 0, preclinical, characterized by heavy and continuous snoring and isolated or clusters of apneas only during light or REM sleep, especially with the patient lying supine; stage I, or i nitial, with slight daytime somnolence and persistent apneas during light and REM sleep in the supine position; stage II, overt, when somnolence becomes more marked and apneas persist throughout sleep, independent of the patient's position; and stage III, or complicated, when somnolence becomes severe, hypoventilation is found also during

% 100]

Sa02 :J AWAKE N-REM REM

•·o.~j-·--~~ %

St. I

•·o.'~j-'" ~···· · %

100

90

~~~~ Sa02 80 St. m:

70

60

50 30min.

40

Fig. 3 . Sa02 patterns dur i ng wakefulness a nd sleep i n t he different stages of HSD (Lugaras i et al, 1985) .

9

0

NO

RM

AL

H

EA

VY

S

NO

RE

R

J

\)/

I) -1

0 c

m.H

20

-so

cm

.H2

0

(

Fig

. 4

. T

he

incr

ease

d n

egat

ive

end

oth

ora

cic

pre

ssu

re d

uri

ng

sn

ori

ng

in

duc

es

a do

wnw

ard

tracti

on

of

the

lary

ng

o-t

rach

eob

ron

chia

l tr

ee,

low

erin

g t

he h

yo

id b

on

e.

wakefulness, and persistent hypoxia occur during every REM episode (Fig. 3). The apnea indexes, in the 130 patients of ours so classified, vary from 7 in stage 0 to 33 in stage I, and 68 and 75, respectively, in stages II and III. The most severe patients in our staging (those with persistent hypoxia during sleep and hypoventilation during wakefulness) have earlier onset of snoring and OSAS compared to patients in the other 3 stages. Thus, patients who begin to snore early in life are more prone to develop a severe form of OSAS. Another interesting finding was that daytime somnolence, which we objectively measured by multiple sleep latency tests (MSLT), is mild in the first 3 stages of disease without relevant interstage differences, but instead severely and persistently increases in stage III. Somnolence was not correlated with the number of apneas and the apnea index, and therefore neither with the number of arousals or hypoxic events, in any of the first 3 stages. The severe and persistent somnolence in stage III is associated with hypoxia and constant hypoventilation, and persistent hypoxia and hypercapnia could be seen as the major factors responsible for drowsiness. We favour, however, another explanation, that severe and persistent somnolence and the tendency to hypoventilation during REM sleep and wakefulness are both effects of lowered reactivity of the respiratory and vigilance centers. Evolution of HSD is not, however, a linear phenomenon. Heavy and habitual snoring arising early in life predisposes to the severe forms of the disease. Marked somnolence appears only in complicated stages.

PATHOPHYSIOLOGY OF OSAS

Tonic contraction of the dilator oropharyngeal muscles, the genioglossus in particular, normally prevents the floor of the tongue from collapsing against the pharyngeal walls and narrowing or obstructing the upper airways. Relaxation of the oropharyngeal muscles and increased resistance in the supine position favour snoring by narrowing the oropharyngeal isthmus during sleep. The abnormal inspiratory efforts associated with snoring increase negative endothoracic pressure to values as high as 70-80 em H20, and induce a downward traction of the larynx. Lowering of the laryngo-tracheo-bronchial stuctures stretches the walls of the pharynx, narrows the oropharyngeal isthmus so that air velocity increases and, in accordance with the Venturi effect, there is a rise in negative airway pressure.

When such negative pressure exceeds the airway dilating force of the tongue and oropharyngeal muscles, an obstructive apnea ensues (Fig. 4). Exceptionally, obstruction may be brought about by collapsing of the epiglottis into the hypopharyngeal cavity. The upper airways reopen when chemical (C02, 02, pH) and oro-nasal and laryngo-tracheal mechanical stimuli provoke an arousal which effects the dilator muscles. 02, and C02 then quickly reaching basal levels, are no longer effective stimuli and this favours the next obstruction. Apneas are more protracted during REM sleep because of increased motor inhibition and decreased central reactivity to arousing stimuli. In the initial stages of HSD apneas occur only during light and REM sleep because the periodic and irregular breathing patterns typical of these stages induce phasic weakenings or blocks of the central respiration, thus favouring obstruction. CAH, however, is seen in only some patients with OSAS. Possibly, heavy and, long-lasting snoring modifies in the long run the musculo-skeletal structures of the upper airways, responsible for self-worsening of the syndrome. Alternatively, only some patients, those we might call the "bad fighters", by way of adapting to a persistent respiratory overload, undergo a downward regulation of the respiratory set-point, thus developing hypoventilation (and daytime somnolence too, if both breathing and vigilance systems are less sensitive) (Fig. 5).

11

Pulmonary hypertension is directly related to hypoxia. Its rare persistence in subjects with normal ventilation during wake may be simply due to durable changes - hardening and thickening of small arteries in the pulmonary circulation. We do not simplistically atribute systemic hypertension to peripheral vasoconstriction induced by hypoxia. Obstructive apneas provoke wide and repeated pressure changes whithin the thoracic walls which affect heart and baroceptor functions. Phases of vagal and sympathetic activation alternate, and sympathetic overactivity predominates during sleep. The sympathetic hypertonus could explain a persistent hypertension during sleep, and, when circulation finally adapts to the elevated pressure, the systemic hypertension also during wakefulness. Many other, still unknown, mechanisms are, however, also possible, for instance an abnormal incretion of natriuretic hormone.

AWAKE N-REM REM

1~0l .......

% 100

90

~~~ Sa02 80

70

60

50 30 min.

40

Fig. 5. Transient 02 desaturations linked to the apneas; Sa02 values still return to baseline levels in the upper tracings. In a more advanced patient, a tonic fall in Sa02 is seen during REM sleep

(lower tracings).

ETIOLOGY OF OSAS

Several anatomic and functional factors, genetic or acquired, favour the development of snoring and OSAS. Familial predisposition might be based on congenitally restricted pharyngeal dimensions, abnormal extensibility of oro-pharyngeal tissues, instability of breathing during sleep or a plethoric constitution. Snoring and OSAS are favoured by

12

aging which increases the flaccidity of oropharyngeal muscles, and the instability of breathing. Anatomical factors in childhood and some adults are represented by marked adenoidal and tonsillar hypertrophy. The syndrome becomes manifest at around 40 years of age in heavy snorers who are overweight or obese, because obesity reduces the lumen of the oropharyngeal isthmus. Other factors comprise a deviated septum, hypertrophic turbinates, glottic web, macroglossia, micro- or retrognathia, laryngeal stenosis and pharyngeal stenosis after surgery for palatoschisis. On cephalometry and MRI, macroglosia and an elevated distance from tip of hyoid bone to the mandibular plane are common findings in full-blown OSAS. These musculo-skeletal changes could represent the effects rather than the causes of snoring and OSAS. The abnormal inspiratory efforts, in fact, provoke a strong downward traction of the laryngo-tracheo-bronchial tree, which is transmitted, through the hyoid bone, to the tongue and the jaw. The tongue enlarges, and the hyoid bone is lowered (Fig. 4). Such changes are hardly surprising. These same patients show a depression of the rib cage at the level of the insertion of the diaphragm, which is grossly hypertrophic at autopsy.

Functional factors are represented by irregular breathing, a delay in the contraction of the dilator pharyngeal muscles in relation to that of the diaphragm and an abnormally high compliance of the velum palatinum and the pharyngeal wall. A reclining position during sleep worsens muscle hypertonia because the tongue falls backwards. Sleep deprivation increases muscular hypotonia and delays the contraction of the dilator pharyngeal muscles.

Hormonal factors explain the higher incidence of snoring and OSAS in males (13:1 in our material). The lower female predisposition and the increase of OSAS after menopause have been attributed to the stimulant respiratory effects of progestational hormones, whereas in men testosterone has been implicated in the higher male predisposition. Hypothyroidism and acromegaly may contribute to snoring and apneas by both functional and musculo-skeletal alterations. Alcohol, benzodiazepines and other sedatives may cause sleep apnea in a snorer and worsen OSAS by depressing muscle tone and the respiratory centers.

EPIDEMIOLOGY OF SNORING AND OSAS

Habitual snorers represent 19% of an unselected population, 14% among males and 24% among females. Prevalence rises with age, but a decrement is seen after 60 years of age, especially among males. At 60 years of age, 60% of men and 40% of women snore almost every night (Lugaresi et al., 1980). Snoring is favoured by obesity: in the age range 30-59 years, 16% of thin, 32% of moderately obese and 45% of frankly obese people are habitual snores. Habitual snorers show systemic hypertension more often than non-snorers (Lugaresi et al., 1980; Mondini et al., 1983; Norton and Dunn, 1985; Koskenvuo et al., 1985). In a 3 year prospective study on 388 men aged 40-49 years, both heart disease and stroke were more frequent among snorers (Koskenvuo, 1987). Conversely, hypertensive patients show significantly more sleep apneas than controls, suggesting that sleep apnea could play a role in the development of essential hypertension (Kales et al., 1984; Fletcher et al., 1985; Williams et al., 1985). In our opinion, snoring certainly represents a risk factor for the heart and circulation. Whether it is snoring per se, or the increased apneas in heavy snorers which are responsible for this cardiocirculatory risk, cannot be ascertained since sleep apneas are found with increased incidence in males of advancing age, and the exact prevalence of sleep apneas in heavy snorers is unknown. In an at risk population (3100 males aged 30-69 years) 1.3%, mostly around 50-59 years

13

of age, had clinical and polysomnographic evidence of OSAS when an apnea index of 5 was taken as a criterion of disease (Gislason, 1987). Every night snoring was present in 8.2% of our sample of 3479 randomly chosen males aged 30-69 years. Prevalence of OSAS (for an apnea index of 10) was 3%. Snoring was commonest between 40 and 49 years (10.41), and OSAS between 50-59 years of age (4.7%). Both snoring and OSAS markedly diminish after 60 years of age. Obstructive apneas were found in all 32 every night snorers, aged 40-59 years, randomly chosen by us for polysomnography. In 12/32 the apnea index was less than 5, and in 19 less than 10. If we accept 5 or 10 apneas per hour as the normal limit, we should thus conclude that OSAS is found in 1/2 to 2/3 of every night snorers aged 40-69 years. The apnea index and related hypoxic events increased slowly and progressively in the first 22 patients with an apnea index up to 14, but exponentially in the other 10 patients in whom the index reached values from 29 to 80. Daytime somnolence so severe as to represent a relevant social handicap was, however, found only in the 2 patients with the highest apnea indexes, 60 and 80. Our epidemiological studies indicate that, over a critical limit of 20-30 apneas per hour, other self-aggravating factors operate and speed up progression of the respiratory disorder. These self-aggravating factors could be represented by a diminished central excitability induced by sleep fragmentation and/or by the hypoxic events of apneas, and the structural changes (macroglossia, low hyoid bone, etc.) provoked by the abnormal inspiratory efforts. If we accept daytime somnolence as the true marker of a disease state, the risk of transition from a preclinical to fullblown disease is indicated, in our material, by an apnea index of about 30. This is consistent with our proposal that only patients showing apneas confined to light and REM sleep should be classified as initial stages (St. I) of the disease.

TREATMENT OF SNORING AND OSAS

Drug treatments so far attempted (progesterone, protriptyline, strychnine, chloroimipramine, almitrine, doxepam, etc.) have all been discouraging, because of scarce benefit and the number of side-effects. Coffee and colas before going to sleep could help in some mild cases. Alcohol, benzodiazepines and other sedatives and analgesics, diuretics (thiazides, loop diuretics) that induce alkalosis, propranolol (which decreases ventilatory response to 02), steroids (which increase body weight) should all be avoided.

Weight reduction should be attempted first. It has been followed by a substantial improvement in 50% of our cases. Dramatic improvements may be obtained in very obese subjects, but neither the initial weight nor the magnitude of weight loss are predictive of the results . When the results are unsatisfactory, surgical procedures may be indicated: tonsillectomy in children with large tonsils, corrective interventions on the nose in cases with functional nasal insufficiency. Uvulo-palatopharyngo-plasty eliminates a redundant soft palate and a hypertrophic uvula, but must be reserved for patients with a normal hypopharyngeal cavity. Half of the patients may be so helped, but the remainder show little improvement or even worsen (Riley et al., 1985). Corrective interventions may be conducted in cases with retro-or micrognathia. In severe cases without clear anatomic causes, tracheostomy dramatically resolves even the most advanced stages of the disease. Non-invasive procedures proposed comprise tongue retaining and dental devices, which are useful, but poorly tolerated. CPAP, a device consisting of a compressor, that through a mask emits air at weakly positive pressure (5-15 em H20) in the upper airways, is more successful and also very useful for physio-pathogenic studies. Its shortcomings for long-term use are

14

the noise of the machine, intolerance of the mask, local irritation and aerophagia. Technical improvements have, however, allowed its application in a numerous series of patients (Sullivan et al., 1985; Sanders, 1984; McEvoy and Thronton, 1984; Wilhoit et al., 1984).

DIFFERENTIAL DIAGNOSIS OF HSD

We saw how some patients reported in the Fifties as primary, neurogenic, or central alveolar hypoventilation were really typical cases of OSAS arising in non-obese subjects. A true idiopathic chronic alveolar hypoventilation (ICAH) exists, however, as a nosological entity, clinically and polysomnographically distinct from HSD.

CLINICAL FEATURES OF ICAH

ICAH may be congenital or acquired, cryptogenic or secondary to developmental, vascular or inflammatory disorders. Episodes of transient respiratory failure arising or worsening during sleep are the main clinical features. The most common symptoms of sleep-related respiratory insufficiency are headache and drowsiness persisting for a while on awakening and always correlated with hypercapnia. On occasions, sleep-related respiratory insufficiency may be so severe as to cause persistent stupor or even coma. The episodes of lethargy and hypercapnic coma are usually provoked by concomitant respiratory infections (bronchopneumonia, tracheitis, simple rhinitis), administration of sedative drugs (eg. barbiturates), anesthesia, or rib cage fractures. A history of repeated admission to intensive care units in which the patient is given a few days of assisted ventilation and then discharged free of symptoms, is a frequent finding. Patients do not complain for continuous sleepiness or heavy, habitual or intermittent snoring. The latter features facilitate the clinical differentiation from HSD. Relatives often report that a patient has "blue lips" during sleep. Spirometric tests are usually normal. Slight hypercapnia, hypoxia or acidosis may be found on blood gas analysis; these are, however, rapidly brought to normal by voluntary hyperventilation. The cardiomegaly is a feature only in long-lasting cases.

POLYSOMNOGRAPHIC ASPECTS OF ICAH

During active wakefulness Paco2, Pao2 and pH remain within normal limits (Coccagna et al., 1984). A tendency to hypoventilation becomes evident during quiet wakefulness and in the relaxed state prior to falling asleep. With sleep onset, there is a rapid and progressive fall in alveolar ventilation, and Sa02 may drop to values of 40-60% and even lower in a few minutes. Alveolar hypoventilation is not related to the presence of apneas.

Central or obstructive apneas, as a rule, occur only sporadically throughout sleep, whereas single apneas provoke transient but severe worsenings of hypoxia (Coccagna et al., 1984). In particular instances the automatic control of breathing is affected to such a degree that mechanical ventilatory assistance or the use of phrenic nerve stimulators may be needed throughout sleep. Continuous Sa02 monitoring by ear oxymeter shows several different patterns. In fact, hypoxia may persist at low levels throughout sleep, but it may also return for quite long periods to values comparable to wakefulness. In most patients REM sleep is associated with worsening ventilation. However, occasional patients, especially children, may show the worst ventilatory performances during

15

NREM sleep (Fleming et al., 1980). In at least 3 out of 9 patients observed by us during repeated polysomnographic recordings, hypoxia was worse particularly during snoring. Systemic arterial pressure does not increase during sleep, despite the severity of hypoxia (Coccagna et al., 1984).

ETIOLOGY AND PATHOPHYSIOLOGY OF ICAH

In many cases of ICAH pathological data are either lacking or negative. In some autopsied cases changes have been found in the postero-lateral bulbar areas, where the primary centers for automatic ventilation are located (Barlow et al., 1980). Malformations of the neural crest (ganglioneuroma, multiple ganglioblastoma, Hirschsprung's disease) account for congenital cases (Barlow et al., 1980).

Acquired forms include cases of bulbar poliomyelitis, encephalitis, syringobulbia, brainstem tumors and latera-bulbar infarcts (Plum and Leigh, 1981). These pathological findings and the failure to respond to C02 suggest that the primary disturbance lies in a decreased excitability of the automatic breathing centers or a lesion of the descending bulbospinal pathways. Peripheral disorders, e.g. of the respiratory muscles, may be more frequent factors in ICAH than is commonly thought. In fact diaphragmatic paralysis has been reported (Newsom-Davis et al., 1975) as a cause of ICAH, and hypotonia and weakness of the diaphragm and intercostal muscles, rather than a primary central defect, could account for the ICAH observed in cases of Prader-Willi syndrome (Vela-Buena et al., 1984). More clinical experience and confirmation of pathological data are, however, needed in order to elucidate further the clinical and polygraphic aspects of ICAH and to delineate possible subtypes.

SO-CALLED CENTRAL APNEA SYNDROMES

We do not consider this syndrome an independent nosological entity. In our cases central apneas have always been a minor finding compared to obstructive apneas and do not induce significant 02 desaturation or pressure changes unless they occur in patients with ICAH or respiratory impairment of peripheral origin. In patients with HSD, central apneas may be frequent following tracheostomy, thus indicating a delayed resetting of the centers to the new breathing conditions. We have never observed a patient with clinical and polygraphic features indicative of central apnea syndrome (Guilleminault et al., 1973; White, 1985). One exceptional case presented periodic breathing and serial central apneas throughout sleep. However, these respiratory oscillations were accompanied by marked changes in heart rate and arterial pressure and the patient's history was typical of ICAH.

REFERENCES

Barlow, P.B., Bartlett, D., Hauri, P., Hellekson, C., Nattie, E.E., Remmers, J.E., and Schmidt-Nowara, W.N., 1980, Idiopathic hypoventilation syndrome: importance of preventing nocturnal hypoxemia and hypercapnia, Am. Rev. Resp. Dis., 121: 141.

Coccagna, G., Cirignotta, F, Zucconi, M., Gerardi, R., Medori, R., and Lugaresi, E., 1984, A polygraphic study of one case of primary alveolar hypoventilation (Ondine's Curse), Bull. Eur. Physiopathol. Respir., 20: 157.

Coccagna, G., Mantovani, M., Brignani, R., Parchi C., and Lugaresi, E.,

16

1972a, Continuous recording of the pulmonary and systemic arterial pressure during sleep in syndromes of hypersomnia with periodic breathing, Bull. Physiopathol. Respir., 8: 1217.

Coccagna, G., Mantovani, M., Brignani, F., Parchi, C., and Lugaresi E., 1972b, Tracheostomy in hypersomnia with periodic breathing, Bull. Physiopathol. Respir., 8: 1217.

Fleming, P.J., Cade, D., Bryan, M.H., and Bryan A.C., 1980, Congenital central hypoventilation and sleep state, Pediatrics, 66: 425.

Fletcher, E.C., De Behuke, R.D., Lovoi, M.J., and Gorin, A., 1985, Undiagnosed sleep apnea in patients with essential hypertension, Ann. Int. Med., 103: 190.

Gislason, ~ 1987, Sleep apnea. Clinical symptoms, epidemiology and ventilatory aspects, Thesis, Uppsala, pp. 1-48.

Guilleminault, C., Heldridge, F.L., and Dement, W.C., 1973, Insomnia with sleep apnea. A new syndrome, Science, 181: 856.

Kales, A., Bixler, E.O., Cadieux, R.J., Schneck, Shaw, L.C., Loke, T.W., Vela-Bueno, A., and Soldatos, C.R., 1984, Sleep apnea in hypertensive population, Lancet, 11: 1005.

Koskenvuo, M., Kaprio, J., Partinen, M., Langinvainio, H., Sarna, S., and Heikkila, K., 1985, Snoring as a risk factor for hypertension and angina pectoris, Lancet, 1: 893.

Koskenvuo, M., Kaprio, J., Talakivi, T., Partinen, M., Heikkila, K., and Sarna, S., 1987, Snoring as a risk factor for ischemic heart disease and stroke in men, Br. Med. J., 294: 16.

Lugaresi, E., Cirignotta, F., Coccagna, G~ and Piana, C., 1980, Some epidemiological data on snoring and cardiocirculatory disturbances, Sleep, 3: 221.

Lugaresi, E., Coccagna, G., and Mantovani, M., 1978, Hypersomnia with periodic apneas, in: "Advances in Sleep Research", vol. 4, Spectrum, New York.

Lugaresi, E., Mondini, S., Zucconi, M., Montagna, P., and Cirignotta, F., 1983, Staging of heavy snorers disease: a proposal, Bull. Eur. Physiopathol. Resp., 19: 590.

McEvoy, R.D., and Thornton, A.T., 1984, Treatment of obstructive sleep apnea syndrome with nasal continuous positive airway pressure, Sleep, 7: 313.

Mondini, S., Zucconi, M., Cirignotta, F., Aguglia, U., Lenzi, P.L., Zauli, C., and Lugaresi, E., 1983, Snoring as a risk factor for cardiac and circulatory problems: an epidemiological study, in: " Sleep-Wake Disorders; Natural History, Epidemiology and Long-term Evolution", Guilleminault, C., Lugaresi, E., eds., Raven Press, New York.

Newsom-Davis, J., Goldman, M., Loh, L., and Casson, M., 1975, Diaphragm function and alveolar hypoventilation, ~ ~ Med., 45: 87.

Norton, P.G., and Dunn, E.V., 1985, Snoring as a risk factor for disease, an epidemiological survey, Br. Med., 291: 630.

Plum, F., and Leigh, R.J., 1981, Abnormalities of central mechanisms. in: "Regulation of Breathing",!!, Hornbein, T.F., ed., Decker, New York.

Riley, R., Guilleminault, C., Powell, N., and Simmons, F.B., 1985, Palatopharyngoplasty failure, cephalometric roentgenograms and obstructive sleep apnea, otolaryngol. Head Neck Surg., 93: 240.

Sanders, M.H., 1984, ·Nasal CPAP effect on patterns of sleep apnea, Chest, 86: 839.

Sullivan, C.E., Berthon-Jones, M., lssa, F.G., and Evis, L., 1985, Reversal of obstructive sleep apnea by continuous positive airway pressure applied through the nares, Lancet, I: 862.

Tilkian, A.G., Motta, J., and Guilleminault, C., 1978, Cardiac arrhythmias in sleep apnea, in: "Sleep Apnea Syndromes", Guilleminault, C., Dement, W. eds., Liss, New York.

17

Vela-Buena, A., Kales, A., Soldatos, C.R., Dobladez-Blanco, B., Campos Castello, J., Espino Hurtado, P., and Olivan Palacios, J., 1984, Sleep in the Prader-Willi Syndrome, Arch. Neural., 41: 294.

White, D.P., Central sleep apnea, in: "The Medical Clinics of North America", vol. 69, Thawley J.E. ed., Saunders, Philadelphia.

Wilhoit, S.C., Brown, E.D., and Suratt, P.M., 1984, Treatment of obstructive sleep apnea with continuous nasal airflow delivered through nasal prongs, Chest, 85: 170.

Williams, A.J., Houston, D., Finberg, S., Lam, C., Kinney, J.L., and Santiago, S., 1985, Sleep apnea syndrome and essential hypertension, Am. ~ Cardiol., 55: 1019.

18

HYPOXEMIA DURING SLEEP

G. Bonsignore, 0. Marrone, V. Bellia, and F. Cibella

Istituto di Pneumologia, Universita di Palermo Istituto di Fisiopatologia Respiratoria del C.N.R. Palermo, Italy

In most normal subjects arterial oxyhemoglobin saturation (Sa02) remains substantially stable throughout all sleep time: some exceptions however, may be observed especially in elderly subjects, who may show desaturations, sometimes associated with apneas (Block et al., 1979; Krieger et al., 1983; Catterall et al., 1985). Conversely, Sa02 drops during part of, or even the whole sleep time, are a common finding in pathological conditions like obstructive sleep apnea syndrome (OSAS) and chronic obstructive pulmonary disease (COPD).

The subject of this paper is hypoxemia during sleep in OSAS and COPD. The causes, some consequences, and the treatment of hypoxemia in these diseases are different. Therefore we will discuss the two diseases separately, emphasizing the most prominent differences between them.

PATHOGENESIS OF SLEEP HYPOXEMIA

Obstructive Sleep Apnea Syndrome

In OSAS numerous hypoxemic episodes recur during sleep, together with obstructive apneic episodes. It is, therefore, easy to recognize the asphyxia consequent to the apneas as the most direct cause of hypoxemic episodes. The following factors contribute to determination of the severity of desaturations: a) baseline Sa02 level before the occurrence (Strohl and Altose, 1984); b) the lung volume at which occlusion occurs (Findley et al., 1976); oxygen consumption during the apnea (Hurewitz and Sampson, 1987); d) the duration of apnea. In this perspective a particular role is played by obesity which impairing thoraco-pulmonary inflation and probably pharyngeal patency, increasing oxygen cosumption and, sometimes, depressing the excitability of neurochemical control mechanisms (Sharp et al., 1980), may trigger the onset of apneas or worsen their consequences. The duration of apneas is determined mainly by the level of arousability which, in turn, is affected by the degree of sleep deprivation, hypoxemia and hypercapnia during daytime, alcohol consumption, individual reactivity to mechanical and chemical stimuli (Sullivan and Issa, 1980).

19

Chronic Obstructive Pulmonary Disease

Most sleep desaturations in COPD occur during REM stages (Wynne et al., 1979; Douglas et al., 1979; Catterall et al., 1983). The nature of functional derangement producing them is still under debate. In fact, although hypoventilation episodes have been observed during REM sleep (Catterall et al., 1983; Hudgel et al., 1983; Fletcher et al.,1983), it remains controversial whether they are the only causes of desaturations, or a worsening of VAIQ ratio also contributes (Hudgel et al., 1983; Fletcher et al., 1983; Catterall et al., 1985). Since a decrease of minute ventilation during REM sleep has been observed in normals (Douglas et al., 1982), it is conceivable that this phenomenon is more marked in patients with COPD: in fact, during REM sleep intercostal and accessory muscles of respiration are physiologically inhibited, and respiration relies almost exclusively on the diaphragmatic action (Tusiewicz et al., 1977; Johnson and Remmers, 1984). Since in most patients with COPD the diaphragm operates in unfavourable mechanical and metabolic conditions, it can be assumed that ventilation during REM in the patients decreases more than in normals. In addition, a decrease in FRC during REM sleep has been hypothesized (Muller et al., 1980); this could cause a greater part of ventilation to occur below closing volume, further impairing the VA/Q mismatch (Fletcher et al., 1983).

CONSEQUENCES OF HYPOXEMIA


The consequences of hypoxemia in OSAS have not been completely clarified. One reason is that hypoxemia interacts with other factors to produce some disturbances, and its individual role is difficult to distinguish.

Hypoxemic episodes could play a role in the intellectual deterioration of patients with OSAS, but sleep disruption is likely to be a more important cause.

In addition to hypercapnic and mechanical stimuli generated during the apnea, hypoxemia contributes to the evocation of post-apneic arousal and to sleep disruption (Sullivan and Issa, 1980). In our experience, in some patients arousal is more likely to occur when post-apneic Sa02 is low; however, in other patients there is no correlation between postapneic Sa02 levels and frequency of arousals (Marrone et al., 1985).

It is not known whether, in the long run, reccurent hypoxemia of OSAS may induce a blunting of awake hypoxic ventilatory responses; it seems possible that in some subjects the asphyctic episodes caused by the apneas determine a reduction in the ventilatory hypercapnic responsiveness (Rapoport et al., 1986).

The cardiovascular system is one of the most heavily affected by hypoxemia. Repetitive peaks of pulmonary hypertension have been observed during sleep in OSAS, and hypoxemia certainly plays a major role in determining them (Coccagna et al., 1972; Schroeder et al, 1978; Marrone et al., 1987). In 5 out of 6 patients who underwent pulmonary artery catheterization during the night we found a significant correlation between pulse pressure and the concomitant level of hypoxemia. This could have been determined by the effect of hypoxemia on pulmonary artery stiffness mediated by the sympathetic nervous system (Szidon and Flint, 1977). It has been advocated that, as a consequence of nocturnal hypertension, stable diurnal hypertension may ensue. However, this

20

relationship has been excluded by some authors in cases with normal daytime blood gas tensions (Bradley et al., 1985).

Cardiac arrhythmias are common in OSAS, and the most important role in determining them seems to be played by hypoxia and the alternating vagal and sympathetic hypertonicity (Tilkian et al., 1977; Guilleminault et al., 1983). Hypoxemia could also contribute to adrenergic discharge and to recurrent systemic hypertensive peaks following apneas (Fletcher et al., 1987). As the disease progresses, an increase in adrenergic secretion under steady conditions could occur, leading to a stable systemic hypertension in many OSAS subjects.


In COPD nocturnal hypoxemia is of particular concern because sleep may be associated with important worsening of diurnal hypoxemia or even with the appearance of hypoxemia in patients with sufficient Pa02 tensions during wakefulness.

Studies concerning the incidence of nocturnal arrythmias in COPD have shown an increased incidence of arrhythmias in association with hypoxemic episodes (Flick and Block, 1979; Tirlapur and Mir, 1982; Shepard et al., 1985). It has been shown that nocturnal hypoxemic episodes in COPD are accompanied by increases in pulmonary arterial pressure, which are abolished by 02 therapy (Boysen et al., 1979; RUhle et al., 1986). It has been hypothesized (Boysen et al., 1979) that nocturnal hypoxemia could contribute to the development of cor pulmonale.

The role of hypoxemia in determining sleep disruption in COPD is controversial since some authors have found an improvement of sleep structure with 02 therapy (Calverley et al., 1982; Bellia et al., 1985), whereas others have found no change in the frequency of arousals (Fleetham et al., 1982).

EVALUATION OF NOCTURNAL HYPOXEMIA


Hypoxemic episodes, defined as falls in Sa02 ~4~ with respect to a baseline level (Block et al.,1979), are frequent and have a short duration in OSAS. Typically nocturnal oximetry of patients with OSAS shows a large number of desaturations, usually lasting some tens of seconds. This kind of oximetric pattern is strongly suggestive of OSAS. In typical cases, due to the patients' hypersomnia, sleep, though disturbed, may be recorded throughout the whole night, whereas possible awakenings intermingled between apneas are of very short duration. Therefore, the oximetric recording can be confidently assumed as representative of actual sleep-related phenomena. On the oximetry, when featuring the trend on a compact time scale, wakefulness and sleep can be distinguished because Sa02 is constant during wakefulness, while it shows a saw-tooth profile during sleep; in addition, NREM and REM sleep periods can be identified with good approximation because the desaturations occuring during REM sleep are characteristically more severe (Farney et al., 1986) (Fig.1).


In COPD hypoxemic episodes can be analysed, taking into account either falls of Sa02 (~ 4~ with respect to a baseline level), or decreases (~ 10~) with respect to a preceding stable level lasting at

21

a T%ME ( ~or-• )

Fig. 1. Sleep structure and concomitant Sa02 pattern during one night in a patient with OSAS.

least one minute. The second criterion has been used to define and identify so called "transient hypoxemic episodes"(Douglas et al., 1979; Catterall et al., 1983). However, we rather prefer the first method of analysis since it does not neglect the less important falls of Sa02, which by contrast may sometimes be of a longer duration; moreover, due to the characteristics of the desaturations i n OSAS (see above), the second method does not fit to the analysis of Sa02 in this condition since it would miss most of the desaturation episodes. Hypoxemic episodes in COPD usually are few and last some minutes, so that they are easily distiguished from those provoked by apneas. Mostly they are concentrated in REM sleep, where they also show the worst severity (Fig.2). Desaturations appearing in COPD do not differ from those that can be observed in other diseases like kyphoscoliosis, pulmonary fibrosis or cystic fibrosis (Kryger, 1985)) .

Concerning the discriminating power of oximetry itself, it must be considered that unlike in OSAS, long periods of wakefulness commonly recur during the night in COPD. During NREM sleep Sa02 decrease from the wakefulness level, but the magnitude of t his decrease - often small - as well as the substantial stability of Sa02 during both wakefulness and NREM, sometimes make their distiction as a sleep-related phenomenon on the oximetric recording difficult; conversely, REM sleep is usually well characterized, since it is associated with long and severe desaturation episodes, mainly in patients of the "blue and bloated" type (Douglas et al., 1979; Catterall et al., 1983), so that the appearance of such episodes makes possible to suspect its occurence. For these reasons the monitoring of EEG together with the oximetry is more important in COPD than in OSAS to s eparate the Sa02 values beloning to wakefulness from those of sleep.

Whereas the pattern of desaturations (numerous and short in OSAS, few and long in COPD) allows a clear distinction between patients wit h OSAS and those with COPD or other diseases, the severity of nocturnal hypoxemia does not. In fact, the different desaturation patterns of these pathological conditions may be associated with comparable degrees

22

9t>

&t:~

N

"" ! I ... 1

··rr w :rn~ , ::: 1. 1. ! = 2 u ...

::; 3 ~ u: ...

RE-f-Mt -------------~ :: .. 7 ..

T IME <n..-s >

Fig. 2. Sleep structure and concomitant Sa02 pattern during one night in a patient with COPD.

of Sa02 impairment, as expressed by parameters like mean sleep Sa02, lowest nocturnal Sa02, percentage of sleep time below a particular Sa02 level.

THERAPEUTIC APPROACHES


Because of the common cause, treatment of hypoxemia in OSAS is closely related to the prevention of obstructive apneas. However, attempts to reverse hypoxemi a independently of the treatment of apneas have been performed with 02 administration. A higher pre-apneic Sa02, set by 02 administration, could be responsible for a less severe hypoxemic level f ollowing each apnea. However, as a side effect, some authors have observed a potentially dangerous prolongation of the apneas after 02 administration, that could discourage from its use (Schroeder et al., 1978); later this finding has not been confirmed by other investigators , who have found that 02 administration improves oxygenation and other symptoms, and does not prolong apneas to a troublesome extent (Martinet al., 1982; Smith et al., 1984; Alf ord et al., 1986). We have found an incons istent response t o 02 administration in our OSAS subjects. Some of t hem show a marked improvements in their sleep Sa02; in these subjects we found an unchanged or sometimes abbreviated mean apnea duration following 02 administration (Fig.3); therefore the increased baseline level of Sa02 resulted in less severe drops of Sa02 as a consequence o f apneas. Ot her patients barely show any Sa02 variation aft er administration; in these subjects we observed a significantly prolonged mean apnea duration (Fig. 4); in such patients 02 administration seems contraindicated, since it may be associated with a more important degree of C02 retention during the apneas, and t heoretically also wit h a higher incidence of cardiac arrhythmias (Schroeder et a l. , 1978).

Other forms of treatment of OSAS improve oxygenation trough their effect on apneas. They inc lude (St rohl et al., 1986 ) weight loss,

23

1.1710

: • ~1'111 ~ , ~,

.... 0 .. "'

.... 0 . "'

" 3

'I

e e. 7 e TYME <Mr-• >

Air

o,

Fig. 3. Sa02 behavior during air breathing (above) and its improvement during oxygen administration (below) in two consecutive nights in a patient with OSAS. In the two nights mean apnea durations were respectively 29 ± 14 and 31 ± 9 s during NREM sleep (NS); 79 ± 30 and 51 ± 32 s during REM sleep (p<.Ol).

administration of drugs (medroxyprogesterone, protryptyline, nicotine, almitrine etc.), surgical therapy on upper airway abnormalities, and ventilation with continuous positive airway pressure applied through the nose (Bonsignore et al., 1987).


Treatment of nocturnal hypoxemia in COPD is mainly performed with 02 administration. It has been demonstrated that long term 02 therapy in hypoxemic subjects improves life expectance and prevents worsening of pulmonary hypertension (Nocturnal Oxygen Therapy Trial Groups, 1980; Medical Research Council Working Party, 1981) whereas it is not clear 'whether it is effective in decreasing daytime pulmonary pressure levels (Medical Research Council Working Party, 1981). In the short run, 02 therapy prevents the nocturnal hypertensive peaks in pulmonary pressure as a consequence of the abolition or improvement of the hypoxemic episodes (Boysen et al., 1979; Fletcher and Levin, 1984; RUhle et al., 1986). In addition it does not seem to induce any C02 retention in patients out of exacerbation (Goldstein et al., 1984) and could induce some improvement in sleep structure (Calverley et al., 1982; Bellia et al., 1985).

CONCLUSIONS

Sa02 derangements during sleep are represented by two main different patterns. The first one may be identified by that of OSAS (high

24

1 00

• o -~o

-e .

= •

Fig. 4. Sa02 behaviour during air breathing (above) and lack of improvement during oxygen administration (below) in two consecutive nights in a patient with OSAS. In the two nights mean apnea duration were respectively 23 ± 8 and 32 ± 7 seconds during NREM sleep (p<.001); 48 ± 30 and 57 ± 30 seconds during REM sleep (NS).

frequency- short duration desaturations), the second one by that of COPD (low frequency- prolonged duration desaturations).

OSAS and COPD represent two distinct pathological conditions, as concerns both the pattern of nocturnal desaturations and their pathogenesis . Despite these different characteristics the mean level of hypoxemia achieved during the night may be similar in the two conditions and result in some common consequences. A correct quantification of sleep hypoxemia can be precisely performed when both the EEG and Sa02 are recorded; however in OSAS the short wakefulness duration and Sa02 characteristics allow to calculate with a good approximation mean hypoxemia levels during sleep from the oximetry alone. Treatment of hypoxemia in OSAS and COPD shall be different due to their different pathogenetic mechanisms.

REFERENCES

Alford, N.J., Fletcher, E.C., and Nickeson, D., 1986, Acute oxygen in patients with sleep apnea and COPD, Chest, 89: 30.

Bellia, V., Marrone, 0., Milone , F., Coppola, P., Oddo, S., and Ferrara, G., 1985, Effetti dell'ossigenoterapia notturna sui disordini r espiratori nella broncopneumopatia cronica ostruttiva, Lotta contra la Tuberc . ~ Malattie Polm. Soc., 55: 390.

Block, A.J., Boysen, P .G., Wynne, J.W., and Hunt, L.A., 1979, Sleep apnea, hypopnea and oxygen desaturation in normal_§ubject s. A

25

strong male predominance, ~Engl. ~ Med., 300: 513. Bonsignore, G., Marrone, 0., Bellia, V., Giannone, G., Ferrara, G., and

Milone, F., 1987, Continuous positive airway pressure improves the quality of sleep and oxygenation in obstructive sleep apnea syndrome, !tal.~ Neural. Sci., 8: 129.

Boysen, P.G., Block, A.J., Hunt, L.A., and Flick, M.R., 1979, Nocturnal pulmonary hypertension in patients with chronic obstructive pulmonary disease, Chest, 76: 536.

Bradley, T.D., Rutherford, R., Grossman, R.F., Lue, F., Zamel, N., Moldofsky, H., and Phillipson, E.A., 1985, Role of daytime hypoxemia in the pathogenesis of right heart failure in the obstructive sleep apnea syndrome, Am. Rev. Respir. Dis., 131: 835.

Calverley, P.M.A., Brezinova, V., Douglas, N.J., Catterall, J.R., and Flenley, D.C., 1982, The effect of oxygenation on sleep quality in chronic bronchitis and emphysema, Am. Rev. Respir. Dis., 126: 206.

Catterall, J.R., Calverley, P.M.A., MacNee, W.C., Warren, P.M., Shapiro, C.M., Douglas, N.J., and Flenley, D.C., 1985, Mechanisms of transient nocturnal hypoxemia in hypoxic chronic bronchitis and emphysema,~~ Physiol., 59: 1698.

Catterall, J.R., Calverley, P.M.A., Shapiro, C.M., Flenley, D.C., and Douglas, N.J., 1985, Breathing and oxygenation during sleep are similar in normal men and normal women, Am. Rev. Respir. Dis., 131: 86.

Catterall, J.R., Douglas, N.J., Calverley, P.M.A., Shapiro, C.M., Brezinova, V., Brash, H.M., and Flenley, D.C., 1983, Transient hypoxemia during sleep in chronic obstructive pulmonary disease is not a sleep apnea syndrome. Am. Rev. Respir. Dis., 128: 24.

Coccagna, G., Mantovani, M., Brignani, F., Parchi, C., and Lugaresi, E., 1972, Continuous recording of the pulmonary and systemic arterial pressure during sleep in syndromes of hypersomnia with periodic breathing, Bull. Physiopath. Resp, 8: 1159.

Douglas, N.J., Calverley, P.M.A., Leggett, R.J.E., Brash, H.M., and Flenley, D.C., 1979, Transient hypoxemia during sleep in chronic bronchitis and emphysema, Lancet, I: 1.

Douglas, N.J., White, D.P., Pickett, C.K., Weil, J.V., and Zwillich, C.W., 1982, Respiration during sleep in normal man, Thorax, 37: 840.

Farney, R.J., Walker, L.E., Jensen, R.L., and Walker, J.M., 1986, Ear oximetry to detect apnea and differentiate rapid eye movement (REM) and non-REM (NREM) sleep. Screening for the sleep apnea syndrome, Chest, 89: 533.

Findley, L.J., Ries, A.L., Tisi, G.M., and Wagner, P.D., 1983, Hypoxemia during apnea in normal subjects: mechanism and impact of lung volume, ~ ~ Physiol., 55: 1777.

Fleetham, J., West, P., Mezon, B., Conway, W., Roth, T., and Kryger, M., 1982, Sleep, arousal and oxygen desaturation in chronic obstructive pulmonary disease, Am. Rev. Respir. Dis. 126: 429.

Fletcher, E.C., Gray, B.A., and Levin, D.C., 1983, Non apneic mechanisms of arterial oxygen desaturation during rapid-eyemovement sleep, ~ ~ Physiol., 54: 632.

Fletcher, E., and Levin, D., 1984, Cardiopulmonary hemodynamics during sleep in subjects with chronic obstructive pulmonary disease. The effect of short and long-term oxygen, Chest, 85: 6.

Fletcher, E.C., Miller, J., Sharf, J.W., and Fletcher, J.C., 1987, Urinary catecholamines before and after tracheostomy in patients with obstructive sleep apnea and hypertension, Sleep, 10: 35.

Flick, M.R., and Block, A.J., 1979, Nocturnal vs diurnal cardiac arrhythmias in patients with chronic obstructive pulmonary disease, Chest, 75: 8.

26

Goldstein, R.S., Ramcharan, V., Bowes, G., McNicholas, W.T., Bradley, D., and Phillipson, E.A., 1984, Effect of supplemental nocturnal oxygen on gas exchange in patients with severe obstructive lung disease, ~Engl.~ Med., 310: 425.

Guilleminault, C., Connally, S.J., and Winkle, R.A., 1983, Cardiac arrhythmias and conduction disturbances during sleep in 400 patients with sleep apnea syndrome, Am. ~ Cardiol., 52: 490.

Hudgel, D.H., Martin, R.J., Capehart, M., Johnson, B., and Hill, P., 1983, Contribution of hypoventilation to sleep oxygen desaturation in chronic obstructive pulmonary disease, J. ~ Physiol., 55: 669.

Hurewitz, A.N., and Sampson, M.G., 1987, Voluntary breath holding in the obese,~~ Physiol., 62: 2371.

Johnson, M.W., and Remmers, J.E., 1984, Accessory muscle activity during sleep in chronic obstructive pulmonary disease (COPD), ~ ~ Physiol., 57: 1011.

Krieger, J., Turlot, J.C., Mangin, P., and Kurtz, D., 1983, Breathing during sleep in normal young and elderly subjects: hypopneas, apneas and correlated factors, Sleep, 6: 108.

Kryger, M.H., 1985, Sleep in restrictive lung disorders, Clin. Chest Med., 6: 675.

Marrone, 0., Ferrara, G., Milone, F., Macaluso, C., Bellia, V., and Bonsignore, G., 1987, Effects of obstructive sleep apneas and of continuous positive airway pressure on pulmonary hemodynamics, Bull. Europ. Physiopath. Resp., 23(suppl. 12): 419S.

Marrone, 0., Milone, F., Silvestri, R., Oddo, S., Coppola, P., Rizzo, A., and Bonsigniore, G., 1985, Is awakening essential to the relief of hypoxemia?, Bull. Europ. Physiopath. Resp., 21 (suppl.): 42H

Martin, R.J., Sanders, M.H., Gray, B.A., and Pennock, B.E., 1982, Acute and long-term ventilatory effects of hyperoxia in adult sleep apnea syndrome, Am. Rev. Respir. Dis., 125: 175.

Medical Research Council Working Party. Long-term domiciliary oxygen therapy in chronic hypoxic cor pulmonale complicating chronic bronchitis and emphysema, 1981, Lancet, I: 681.

Muller, N.L., Francis, P.W., Gurwitz, D., Levison, H., and Bryan, C., 1980, Mechanisms of hemoglobin desaturation during rapid eye movement sleep in normal subjects and in patients with cystic fibrosis, Am. Rev. Respir. Dis., 121: 463.

Nocturnal Oxygen Therapy Trial Groups. Continuous or nocturnal oxygen therapy in hypoxemic chronic obstructive lung disease: a clinical trial, 1980, Ann. Intern. Med., 93: 391.

Rapoport, D.M., Garay, S.M., Epstein, H., and Goldring, R.M., 1986, Hypercapnia in the obstructive sleep apnea syndrome. An evaluation of the "Pickwickian syndrome", Chest, 89: 627.

RUhle, K.H., Klein, D., Kohler, D., Costabel, V., and Matthys, H., 1986, Effects of 2 and 4 1/min oxygen breathing on pulmonary artery pressure. Blood gases and sleep stages in patients with chronic obstructive lung disease, Eur. ~ Respir. Dis., 69 (Suppl.146): 443.

Schroeder, J.S., Motta, J., and Guilleminault, C., 1978, Hemodynamic studies in sleep apnea, in: "Sleep apnea syndromes", C. Guilleminault, W.C. Dement eds., AR Liss Inc., New York.

Sharp, J.T., Barrocas, M., and Chokroverty, S., 1980, The cardiorespiratory effects of obesity, Clin. Chest Med., 1: 103.

Shepard, J.W., Garrison, M.W., Grither, M.A., Evans, R., and Schweitzer, P.K., 1985, Relationship of ventricular ectopy to nocturnal oxygen desaturation in patients with chronic obstructive pulmonary disease, Am. J. Med., 78: 28.

Smith, P.L., Haponick~F., and Bleecker, E.R., 1984, The effect of oxygen in patients with sleep apnea, Am. Rev. Respir. Dis., 130: 958.

27

Strohl, K.P., and Altose M.D., 1984, Oxygen saturation during breathholding and during apneas in sleep, Chest, 85: 181.

Strohl, K.P., Cherniack, N.S., and Gothe,~ 1986, Physiologic basis of therapy for sleep apnea, Am. Rev. Respir. Dis., 134: 791.

Sullivan, C.E., and Issa, F., 1980, Pathophysiological mechanisms in obstructive sleep apnea, Sleep, 3: 235.

Szidon, J.P., and Flint, J.F., 1977, Significance of sympathetic innervation of pulmonary vessels in response to acute hypoxia, ~ ~ Physiol., 43: 65.

Tilkian, A.G., Guilleminault, C., Schroeder, J.S., Lehrman, K.L., and Simmons, F.B., and Dement, W.C., 1977, Sleep inducted apnea syndrome. Prevalence of cardiac arrhythmias and their reversal after tracheostomy, Am. J. Med., 63: 348.

Tirlapur, V.G., and Mir~A.M., 1982, Nocturnal hypoxemia and associated electrocardiographic changes in patients with chronic obstructive airway disease,~ Engl.~ Med., 306: 125.

Tusiewicz, K., Moldofsky, H., and Bryan, A.C., 1977, Mechanics of the rib cage and diaphragm during sleep, ~ ~ Physiol., 43: 600.

Wynne, J.W., Block, A.J., Hemenway, J., Hunt, L.A., and Flick, M.R., 1979, Disordered breathing and oxygen desaturation during sleep in patients with chronic obstructive lung disease, Am. J. Med., 66: 573. ----

28

THE UPPER AIRWAY MUSCLES: THEIR ROLE IN SLEEP-RELATED

RESPIRATORY DYSRHYTHMIAS

Neil S. Cherniack and David W. Hudgel

Department of Medicine, Case Western Reserve University School of Medicine, Cleveland, USA

The upper airways are a convoluted set of channels which air from the atmosphere must traverse to contact the gas-exchanging surfaces of the lungs. The upper airways contain muscular structures which participate not only in breathing, but in speech, chewing and swallowing as well. In part on account of their complex anatomy, the upper airways make up 40% to 70% of the total resistance to air flow even during resting breathing in the awake state. In addition, there are sectors of narrowing in the nasal, oral, and laryngopharynx where even minor anomalies in configuration can substantially affect resistance to air flow (Lunteren and Strohl, 1986; Widdicombe, 1986; Cherniack, 1984).

Because the walls of the upper airways are not rigid and because its channels contain mobile structures, such as the tongue and vocal cords, airway resistance can change substantially as the balance of forces in the upper airway alters with posture and breathing. For example, the subatmospheric intraluminal pressure and suction produced during inspiration by contraction of the thoracic muscles can lead to narrowing of the compliant tissues of the upper airway or the displacing of structures within its passages that increase airflow resistance or even cause obstruction.

The pressure changes themselves as well as the distortion of tissue can elicit reflex effects which also alter breathing. Van de Graaff (personal communication) has suggested that mechanical traction on the hyoid muscles caused by outward inspiratory movements of the sternum to which some of those muscles attach may also decrease airway resistance.

The passages of the upper airways are lined by muscles which when they contract can dilate or compress the channels through which air flows. Many of the upper airway muscles exhibit tonic activity which stiffens the airway. Some of these muscles also have phasic activity synchronous with inspiration and have a dilating action which tends to oppose the occlusive effects of the negative intraluminal pressure that is present during inspiration. It is also possible that some muscles which are active in expiration may promote obstruction. The upper airway muscles respond to many of the same mechanical or chemical stimuli as the thoracic muscles, but not infrequently the response differs quantitatively or qualitatively from that of the thoracic pumping muscles.

29

During sleep the dilating activity of the upper airway muscles is generally reduced more than the activity of thoracic muscles (Lunteren and Strohl, 1986; Cherniack, 1984). The postural changes associated with sleep, like the assumption of the supine position, produce gravitational forces which tend to decrease upper airway cross-sectional areas. Both these changes contribute to the increased upper airway resistance during sleep, so that the residual inspiratory activity of the upper airway muscles becomes more important than during wakefulness in maintaining patency. This is especially true when the airway is anatomically compromised or narrower than it usually is.

In this paper we will briefly review the responses of upper airway muscles to different chemical and mechanical stimuli, contrasting their behavior to those of the thoracic muscles. We will also suggest possible mechanisms by which the upper airway muscles may be involved in the central and obstructive apneas of sleep. Finally, we will indicate where more investigation would increase our understanding of the pathophysiology of sleep apnea.

RESPONSE OF THE UPPER AIRWAY MUSCLES TO HYPERCAPNIA AND TO HYPOXIA

Like the diaphragm many of the muscles of the upper airway (e.g. geniohyoid, genioglossus, sternohyoid, alae nasi, posterior cricoarytenoid) display phasic inspiratory activity which increases with respiratory stimulation (Weiner et al., 1982). Peak inspiratory activity tends to be reached sooner in the upper airway muscles (Lunteren and Strohl, 1986). These differences in within-breath timing may help maintain airway patency.

Changes with hypercapnia in the activity of the posterior cricoarytenoid muscle parallel those of the diaphragm. When a wide range of C02 stimulation is studied, the posterior cricoarytenoid begins and plateaus at nearly the same levels of PC02 as the diaphragm. In anesthetized animals, muscles innervated by the hypoglossal and genioglossal nerves tend to begin and plateau at higher levels of PC02 than the phrenic, the motor nerve to the diaphragm. The inspiratory activity of geniohyoid, sternohyoid, and alae nasi also rises with C02, but higher levels of PC02 are needed to initiate phasic activity in these muscles compared to the diaphragm (Graaff et al., 1984; Lunteren et al., 1987). Because of these differences in onset and plateau levels of C02, the relationship between phrenic versus hypoglossal activity is curvilinear. Thus, at lower levels of PC02 phrenic activity increases more than hypoglossal activity, while at higher levels of PC02 the opposite is true (Weiner et al., 1982). It is of interest that when phasic inspiratory activity is present, the inspiratory discharge of the upper airway muscles frequently slightly precedes that of the diaphragm (Lunteren and Strohl, 1986).

Applications of inhibitory and excitatory agents to the ventral medullary surface, the presumed location of central chemoreceptors, in anesthetized animals has consistently shown greater effects on the hypoglossal nerve than on the phrenic (Lunteren and Strohl, 1986; Haxhiu et al., 1984a; 1986).

What accounts for these differences in response is unclear. One factor may be the characteristics of the motor neurons themselves or in the number and distribution of projections they receive from chemoreceptors and from central respiratory interneurons. It is possible, though, that they are caused by anesthesia or by the release of

30

endorphins during surgical procedures needed to isolate nerves or implant muscle electrodes. A number of studies have shown that anesthesia and respiratory depressants, such as ethanol, GABA, and diazepam (Haxhiu et al., 1986; Brouillette and Thach, 1980; Bruce et al., 1982; Hwang et al.,1983), have greater effects in reducing the activity of upper airway muscles than the diaphragm. However, recent experiments show that the curvilinear relationship between responses of the genioglossus and diaphragm during hypercapnia persist in acute studies in anesthetized cats even after naloxone administration. Moreover, studies in the decerebrate goat, and most studies in the awake or sleeping cat and goat, show similar differences in behavior of the diaphragm and genioglossus to the C02 stimulus (Eldridge, 1986; Parisi et al., 1987, Haxhiu et al., 1984b; 1987 in the press). However, in these latter studies, the full range of the C02 stimulus (from the PC02 needed to begin rhythmic activity until its plateau) has not always been studied, and behavioral effects on breathing could have occurred.

The same caveats apply to studies in humans. In one set of studies in which genioglossus electrical activity was measured with wire electrodes inserted in the body of genioglossus, and diaphragm activity with eosphageal electrodes, no differences in the changes in inspiratory phasic activity of the two muscles to C02 were observed (Onal et al., 1981 a,b). The hypoglossal nerve is known to contain fibers that are tonically active. The tonic electrical activity of the genioglossus was not included in this set of studies in humans. The differences in the sensitivity of the two methods of recording used, the exclusion of tonic activity, and the relatively narrower range of C02 studied might have obscured dissimilarities in the response of the two muscles in humans. It is of interest that when both the tonic and phasic electrical activity of the airway muscles and the diaphragm are measured with surface electrodes during sleep, differences in behavior in humans can be discerned that are similar to those seen in animals (Hudgel et al., 1987a).

Nonetheless, it is clear that state changes have different effects on the behavior of upper airway and thoracic muscles. Arousal (change from the sleeping to the waking state) increases genioglossal electrical activity more than the diaphragm (Cherniack, 1984). This is presumably an effect that is mediated via higher brain centers, but the mechanism is uncertain.

It is of interest that even in the anesthetized cat, the effects of higher brain centers on the respiratory activity of upper airway muscles can be seen (Mitra et al., 1986). Decerebration depresses hypoglossal nerve responses to C02 and increases the level of C02 needed to initiate phasic activity. Decortication, on the other hand, accentuates the effects of hypercapnia on hypoglossal activity. Neither decortication nor decerebration had an appreciable effect on phrenic nerve responses.

C02 affects respiration both by an action on central and pheripheral chemoreceptors. Stimulation of the carotid body by lobeline or nicotine accentuates the activity of the hypoglossal nerve more than the phrenic, while there is an intermediate effect on the recurrent laryngeal nerve (Lunteren et al., 1984a). However, studies of C02 response during hyperoxia suggest that the carotid body may contribute more to the excitation by hypercapnia of the hypoglossal than the phrenic (Bruce et al., 1982).

Responses to hypoxia of the upper airway muscles have not been studied as much as responses to hypercapnia (Weiner et al., 1982; Onal et al., 1981b). Hypoxia can have both an excitatory effect on respiration by action on pheripheral chemoreceptors and a depressive central effect.

31

The depressive effects of hypoxia and hypercapnia on upper airway muscles have not been systematically investigated.

RESPONSES OF UPPER AIRWAY MUSCLES TO MECHANICAL STIMULI

Differences are also observable in the responses of upper airway and thoracic muscles to mechanical stimuli. Stretch receptor stimulation reduces the electrical activity of upper airway muscles before it cuts short phrenic nerve activity (Lunteren et al., 1984c). Irritant and J receptor stimulation produced by capsaicin or histamine administration affects the upper airway muscles and the diaphragm (Haxhiu et al., 1983; Mitra et al., 1985). With bronchoconstriction, the peak activity of the diaphragm, genioglossus, posterior cricoarytenoid, and alae nasi all increase. In addition, peak activity is reached sooner (Haxhiu et al., 1983). J receptor stimulation by capsaicin produces temporary cessation of activity of the hypoglossal and phrenic nerves but causes tonic excitation of recurrent laryngeal activity (Mitra et al., 1985).

Production of a negative pressure in the upper airway prolongs the inspiratory time of both thoracic and upper airway dilating muscles even when hypercapnia is present (Lunteren et al., 1984b). Although the negative pressure tends to decrease the rate of rise of phrenic activity, there is either no change or an increase in the rate of rise of the genioglossus, alae nasi, and posterior cricoarytenoid electrical activity. Receptors supplied by the trigeminal or superior laryngeal nerves can trigger this reflex. This reflex is of particular interest since it would be expected to occur during upper airway obstruction and might help keep the upper airways patent.

Strong stimulation of the superior laryngeal nerve which is sufficient to abolish phrenic nerve activity also simultaneously abolishes hypoglossal nerve activity. In contrast, hypocapnia, ventral medullary cooling, and GABA, which also inhibit respiratory activity, abolish phasic inspiratory activity in the hypoglossal before the phrenic (Haxhiu et al., 1986).

Changes in blood pressure occur during sleep which might affect the activity of baroreceptors. Baroreceptor stimulation appears to depress upper airway muscle activity more than the diaphragm (Salamone et al., 1983).

Stimulation of somatic and visceral afferent nerves also frequently increases upper airway muscle activity more than the diaphragm. The effects of this afferent input may be diminished during sleep (Haxhiu et al., 1984; Cherniack et al., 1984b).

ROLE OF UPPER AIRWAY MUSCLES IN CENTRAL AND OBSTRUCTIVE SLEEP APNEA

Apneas of the central and obstructive types are the most obvious abnormalities of breathing that occur during sleep (Cherniack, 1984). It has become clear that central apneas in which all respiratory activity ceases, and obstructive apneas which are characterized by upper airway occlusion with continuing respiratory activity, can occur together, in the same individual in the same night and within the same apnea (Cherniack, 1984; Hudgel et al., 1987a). Many apneas begin with a central component and are prolonged by an obstructive phase. Although it was once believed that two different forms of apnea were produced by entirely separate mechanisms, it now seems probable that they have a

32

common or1g1n. Changes in the activity or placement of the upper airway muscles can convert central apnea to obstructive apnea (Cherniack, 1984; Longobardo et al., 1982).

There may, however, be important differences in the or1g1n of single as opposed to recurrent apneic events. The apneas we will discuss are of the recurrent variety (Cherniack, 1984).

It has been noted that some individuals exhibit central apneas while asleep supine but obstructive apneas when in the lateral recumbent position (Issa and Sullivan, 1986). Based on this observation, it has been proposed that certain varieties of upper airway obstruction can lead to hypoventilation or actual cessation of respiratory activity. Tightly juxtaposed tissues may produce distortion and trigger receptors, for example, innervated by the superior laryngeal nerve resulting in central apnea. Slight displacements of tissues might reduce inhibition sufficiently to allow respiration to resume without relieving the obstruction. While this mechanism might allow a single apneic event to occur, it is difficult to see how recurrent apneas could be produced.

It has been noted in sleeping infants that repeated swallowing episodes may be associated with recurrent reflexly produced central apneas (Thach and Menon, 1985). It is not clear that recurrent swallowing episodes occur in sleeping adults; nonetheless, it is possible that repeated attempts to swallow secretions that have accumulated in the airway, particularly in obstructed airways, could lead to both recurrent central and obstructive apneas. One would believe that in such a sequence of events central apneas would follow obstructive apneas, while in fact the opposite is usually seen.

Using comb filtering, Pack has observed cyclic changes in ventilation during wakefulness in elderly individuals who had apneas during sleep (Packet al., 1987). Chapman, working in our laboratory using power spectral analysis, could also detect oscillations both during wakefulness and during sleep in both younger individuals (under 30 years) and in the elderly (over 65 years). Chapman noted such oscillations in 4 of 10 younger individuals awake and in 5 of 10 during sleep. Four of 7 elderly persons had significant ventilatory oscillations awake, while 9 of 10 had them while asleep.

The mechanism of the cyclic changes in ventilation observed by Pack et al. and by Chapman is unclear. The respiratory rhythm can be detected in many different central nervous system neurons, and it is possible that similar oscillations in activity occur in motor neurons supplying upper airway muscles. Sufficiently strong oscillations could conceivably produce recurrent central apneas. Even minimal differences in phase or in amplitude in the oscillation of activity of motor neurons supplying thoracic and upper airway muscles could produce recurrent obstructive apneas.

Pack believed these oscillations could not be due to instability in feedback control (see below) because the oscillations were too long (the length of each cycle in Pack's study varied between 40 to 60 seconds). In feedback instability, cycles are usually shorter. However, Longobardo et al. (1982) have shown that even when circulation time between the lung and the chemoreceptors are assumed to be normal, cyclic changes in ventilation can occur over very long periods (40-90 seconds) if the apneas are obstructive, as they were in most of Pack's subjects. The length of obstructive apneas depends on the rate at which the balance of forces across the airway changes to cause airway reopening and not just on circulation time.

33

We have proposed that one mechanism for recurrent oscillations in breathing could be instability in feedback control of respiration (Longobardo et al., 1982). Since cyclic changes in ventilation would produce cyclic changes in gas exange, and hence in arterial blood and brain gas tensions, even slight differences in the responsiveness of upper airway and thoracic muscles to chemical stimuli could then provoke obstructive as well as central apneas.

On theoretical grounds, instability in ventilatory control (cyclic changes in breathing) can be produced by prolonged circulation time, increased controller responses, i.e. the effect of blood gas changes on ventilation (~V/~PC02 or ~V/~P02), or changes in the resting level of PC02 or P02 which in effect exaggerate the effect of ventilation on blood gas tensions (the gain of the respiratory plant, i.e. the mechanical and chemical substrate in which the controller operates) (Cherniack et al., 1984a; Cherniack and Longobardo, 1986). These kinds of changes have been shown to produce recurrent apneas in tracheotomized anesthetized animals. In humans, recurrent central apneas (Cheyne-Stokes breathing) is observed, not uncommonly, in patients with congestive heart failure with long circulation times. Periodic breathing can be induced most easily in individuals with higher ventilatory responses to C02 (higher controller gains). Hypoxia which increases both controller and plant gains also induces periodic breathing. Bulow (1963) showed that recurrent apneas were more likely to occur during sleep in those who had the greatest elevation in resting PC02 (increased plant gain).

The factors that cause feedback instability are interrelated so that recurrent apneas can be produced with less increase in resting PC02 as circulation time is increased (Cherniack et al., 1984a; Cherniack and Longobardo, 1986).

Disturbances in the form of periods of apnea or hyperventilation are needed to produce the instability that triggers recurrent apneas. Apneas will gradually shorten and disappear if the initial disturbance is slight and there is no further upset to breathing. However, when the initial disturbance is greater, the apneas will reoccur for longer periods of time or even indefinitely. Again there is an interrelationship between the strength of the disturbance and system parameters, such as controller gains, circulation time, and plant gain, which determine the length of reappearance of apneas (Cherniack et al., 1984a; Cherniack and Longobardo, 1986).

As explained earlier, the cyclic changes in blood gases can also trigger obstructive apnea if there are differences at least in onset of activity of upper airway and thoracic muscles. Hudgel et al. (1987a) have shown there are cyclic changes in resistance that occur even in what seems to be recurrent central apneas so that resistance is highest during periods of hypoventilation. This suggests that like ventilation the activity of upper airway muscles oscillates.

It makes no differences in this concept of the cause of recurrent apneas if obstructive apneas are terminated because of direct effects of blood gas changes on the drive to upper airway muscles or indirectly by the effects of blood gas changes on arousal mechanism, which in turn alters upper airway muscle activity. It should be recognized that airway obstruction once in occurs tends to prolong the period of apnea, which exaggerates the changes in gas tensions that occur during apnea, and reinforces the self-perpetuating tendencies of the instability.

Theoretically and in practice, recurrent central apneas diminish in duration and frequency or disappear entirely when the C02 content of

34

inspired gas is increased. Hudgel et al. (1987b) have also shown that C02 inhalation tends to terminate obstructive apneas. According to the instability theory, this might happen for at least two reasons. First, C02 breathing would tend to eliminate central apneas or periods of hypoventilation and so reduce the disturbance to ventilation. Second, C02 breathing might disproportionately increase the activity of upper airway muscles thereby preventing obstruction.

FUTURE AREAS OF INVESTIGATION

While there are now many reasons to believe that the responses of the upper airway muscles are crucial both in the origins of apnea and in determining whether apneas are central or obstructive, available data concerning mechanisms is far from complete.

It is not as yet clear whether the phasic and tonic activity of the upper airway muscles are of equal importance in maintaining airway patency. Nor is it clear whether once a fixed level of upper airway muscle activity is achieved, upper airway remains patent despite increasing contractions of the thoracic muscles, or if upper airway patency always requires equivalent increases in upper airway and thoracic muscle activity. The problem is exacerbated because of difficulties in comparing either the changes in electrical activity or force of the upper airway and thoracic muscles.

Frequently the electrical activity of two nerves or muscles is compared by expressing changes as a percent of control values. Two nerves will always show the same rise as percent of control if their activity begins at the same level of PC02 regardless of the real change in activity. On the other hand, if one nerve begins to discharge at a higher PC02 level than the other, its activity as a percent of control will be greater.

In any case, upper airway obstruction depends on differences or imbalances in forces produced, rather than in electrical activity. However, in many instances, changes in electrical activity parallel changes in mechanical action. The moving average of electrical activity of the whole phrenic nerve correlates well with changes in tidal volume and the forces produced by the thoracic muscles on a given occasion. This has been well demonstrated in the case of the diaphragm and the alae nasi (Lunteren et al., 1985).

Electrical activity is not necessarily a good index of the mechanical action of a muscle. Studies in animals in which both electrical activity and length changes have been measured show that muscles, like the sternohyoid, can lengthen in inspiration even though there is a considerable increase in inspiratory electrical activity (Lunteren et al., 1987). Also, because of the complex anatomical arrangements of both the upper airway and chest wall muscles, it is difficult to infer net changes in overall force from the measurement of single muscles.

Strohl et al. have developed techniques of separating the pressures generated by the upper airway and thoracic muscles in the dog so that the mechanical effects of each group can be evaluated (Teeter et al., 1987). These studies show that there is a curvilinear relationship between pressure developed by the sets of muscles during hypercapnia. However, because of major differences in the anatomy of the upper airway and muscles, these results cannot be confidently extrapolated ~o humans.

35

Finally, it may be that sleep itself is heterogenously caused, and the effects of sleep on the respiratory activity of upper airway and thoracic muscles depend on the nature of the substances producing sleep.

REFERENCES

Brouillette, R.T., and Thach, B.T., 1980, Control of genioglossus muscle inspiratory activity, ~ ~ Physiol., 49: 801.

Bruce, E.N., Mitra, J., and Cherniack, N.S., 1982, Central and peripheral chemoreceptor inputs to phrenic and hypoglossal motoneurons, ~ ~ Physiol., 53: 1504.

Bulow, K., 1963, Respirations and wakefulness in man, Acta Physiol. Scand., 59: 1.

Cherniack, N.S., 1984, Sleep apnea and its causes, ~ Clin. Invest. 73: 1501.

Cherniack, N.S., Gothe, B., and Strohl, K.P., 1984a, Mechanisms for recurrent apneas at altitude, in: "High Altitude and Man", J.B. West, S.Lahiri, eds., American Physiological Society, Bethesda.

Cherniack, N.S., Haxhiu, M.A., Mitra, J., et al., 1984b, Responses of upper airway, intercostal and diaphragm muscle activity to stimulation of oesophageal afferents, ~ Physiol. (London), 349: 15.

Cherniack, N.S., and Longobardo, G.S., 1986, Abnormalities in respiratory rhythm, in: "Handbook of Physiology, Respiratory System", Vol. II, Control of Breathing, American Physiological Society, Bethesda.

Eldridge, F.L., 1986, Hypoglossal nerves in unanesthetized decerebrate cats, Acta Bioi. Med. Exp. 11: 85.

Graaff, W.B.van, Gottfried, S.B., Mitra, J., et al., 1984, Respiratory function of hyoid muscles and hyoid arch, ~ ~ Physiol., 57: 197.

Haxhiu, M.A., Deal, E.C., Van de Graaff, W.B., et al., 1983, Bronchoconstriction: upper airway dilating muscle and diaphragram activity, ~ ~ Physiol., 55: 1837.

Haxhiu, M.A., Lunteren, E.van, Mitra, J., et al., 1984, Comparison of the responses of the diaphragm and upper airway muscles to central stimulation of the sciatic nerve, Respir. Physiol., 58: 65.

Haxhiu, M.A., Lunteren, E.van, Mitra, J., et al., 1984b, Responses to chemical stimulation of upper airway muscles and diaphragm in awake cats, ~ ~ Physiol., 56: 397.

Haxhiu, M.A., Lunteren E.van, Mitra, J., and Cherniack, N.S., 1987, Comparison of the response of diaphragm and upper airway dilating muscle activity in sleeping cats, Respir. Physiol. (in press).

Haxhiu, M.A., Mitra, J., Lunteren, E.van, et al., 1984a, Hypoglossal and phrenic responses to cholinergic agents applied to ventral medullary surface, Am.~ Physiol., 247: R939.

Haxhiu, M.A., Mitra, J., Lunteren, E.van, et al., 1986, Responses of hypoglossal and phrenic nerves to decreased respiratory drives in cats, Respiration, 50: 130.

Hudgel, D.W., Chapman, K.R., Faulks, C., and Hendricks, C., 1987a, Changes in inspiratory muscle electrical activity and upper airway resistance during periodic breathing induced by hypoxia during sleep, Am. Rev. Respir. Dis., 135: 899.

Hudgel, D.W., Hendricks, C., and Dudley, A., 1987b, Alteration of apnea time in obstructive sleep apnea by 02 and C02 inhalation, Am. Rev. Respir. Dis., 135: A185.

Hwang, J., St.John, W.M., and Bartlett, D. Jr., 1983, Respiratory-related hypoglossal nerve activity: influence of anesthetics, ~ ~ Physiol., 55: 785.

Issa, F.G., and Sullivan, C.E., 1986, Reversal of central sleep apnea using nasal CPAP, Chest, 90: 165.

Longobardo, G.S., Gothe, B., Goldman, M.D., and Cherniack, N.S., 1982,

36

Sleep apnea considered as a control system instability, Respir. Physiol., 50: 311.

Lunteren, E.van, Haxhiu, M.A., and Cherniack, N.S., 1985, Respiratory changes in nasal muscle length,~~ Physiol., 59: 453.

Lunteren, E.van, Haxhiu, M.A., and Cherniack, N.S., 1987, Mechanical function of hyoid muscles during spontaneous breathing in cats. ~ ~ Physiol., 62: 582.

Lunteren, E.van, Haxhiu, M.A., Mitra, J., et al., 1984a, Effects of dopamine, isoproterenol, and lobeline on cranial and phrenic motoneurons, ~ ~ Physiol., 56: 737.

Lunteren, E.van, and Strohl, K.P., 1986, Muscles of the upper airways, Clinics in Chest Med., 7: 171.

Lunteren, E.van, Van de Graaff, W.B., Parker, D.M., et al., 1984b, Nasal and laryngeal reflex responses to negative upper airway pressure, ~ ~ Physiol., 56: 746.

Lunteren, E.van, Strohl, K.P., Parker, D.M., et al., 1984c, Phasic volume-related feedback on upper airway muscle activity, ~ ~ Physiol., 56: 730.

Mitra, J., Prabhakar, N.R., Haxhiu, M., and Cherniack, N.S., 1985, The effects of hypercapnia and cooling the ventral medullary surface on capsaicin induced respiratory reflexes, Respir. Physiol., 60: 377.

Mitra, J., Prabhakar, N.R., Haxiu, M.A., and Cherniack, N.S., 1986, Comparison of the effects of hypercapnia on phrenic and hypoglossal activity in anesthetized, decerebrate and decorticate animals, Brain Res. Bull., 17: 181. --

Onal, ~ Lopata, M., and O'Connor, T .D., 1981a, Diaphragmatic and genioglossal electromyogram responses to C02 rebreathing in humans, ~ ~ Physiol., 50: 1052.

Onal, E., Lopata, M., and O'Connor, T.D., 1981b, Diaphragmatic and genioglossal electromyogram responses to isocapnic hypoxia in humans, Am. Rev. Respir. Dis., 124: 215.

Pack, A.I., Millman, R.P., Silage, D.E., Shore, E.T., Knight H., 1987, Oscillatory changes in ventilation awake and sleep, in: "Concepts and Formalization and Control of Breathing", G. Benchetrit, P. Bacconier, J. Demongeot, eds., Manchester University Press.

Parisi, R.A., Neubauer, J.A., Frank, M.M., Edelman, N.H., Santiago, T., 1987, Correlation between genioglossal and diaphragmatic response to hypercapnia during sleep, Am. Rev. Respir. Dis., 135: 378.

Salamone, J.A., Strohl, K.P., Weiner, D.M., et.al., 1983, Cranial and phrenic nerve responses to changes in systemic blood pressure, ~ ~ Physiol., 55: 61.

Teeter, J.P., Fouke, J.M., and Strohl, K.P., 1987, Comparison of volume changes in the upper airway and thorax, ~ ~ Physiol., 62: 284.

Thach, B.T., and Menon, A., 1985, Pulmonary protective mechanisms in infants, Am. Rev. Respir. Dis., 131: 555.

Weiner, D., Mitra J., Salamone J., et al., 1982, Effect of chemical stimuli on nerves supplying upper airway muscles, J. ~ Physiol., 52: 530.

Widdicombe, J.G., 1986, Reflexes from the upper respiratory tract, in: "Handbook of Physiology, Respiratory System", Section 3, Respiration, Vol. II, American Physiological Society, Bethesda, Maryland.

37

EFFECTIVENESS AND SIDE-EFFECTS OF NASAL CONTINUOUS POSITIVE

AIRWAY PRESSURE THERAPY IN 66 PATIENTS WITH SLEEP APNEA

INTRODUCTION

H. Becker, U. Koehler, J.H. Peter, M. Steinberg,and P. von Wichert

Medizinische Poliklinik, Philipps Universitat Marburg, FRG

Since upper airway obstruction has been recognized as the crucial factor in the development of obstructive sleep apnea, tracheostomy -first reported as a therapeutical measure by Kuhlo et al. in 1969- is regarded as the therapy of choice in patients with sleep apnea. The recognition of severe social and medical complications as long-term consequences of tracheostomy paved the ground for an intensive search for alternative methods during the years which followed. Therapeutical success could be achieved with certain medications and with pharyngeal dilatation by way of surgery, but neither of these methods was as successful as tracheostomy (Cohn, 1986; Strohl et al., 1981).

In 1981, Sullivan and colleagues first reported instances of successful therapy of obstructive sleep apnea by means of nasal continuous positive airway pressure (nCPAP) (Sullivan et al., 1981). This form of therapy was then employed successfully in various medical centers (Krieger et al., 1984; Remmers et al., 1984; Becker et al., 1988 in the press). Reports on the efficiency of long-term treatment, however, vary to a sometimes marked degree from center to center (Issa et al., 1987; Schweitzer et al., 1987).

In our hospital, patients with sleep apnea have for some time been treated according to a prechosen standardized therapeutical scheme, including the acquisition of relevant data regarding the effectiveness and the side effects of nCPAP therapy.

METHODS

We studied 66 male patients with an established result of sleep apnea. Their mean age was 52.8 years (34-75 years), and their mean bodyweight 125.7 (84-188) in percent of normal weight. Their initial result of sleep apnea averaged 46.5 (12-113) episodesfh, and the duration of the longest apnea episode per patient averaged 65.4 s (12-180 s). One patient had central sleep apnea, while 65 suffered from the obstructive or mixed type of sleep apnea.

39

Fig.l.

801 AI longest apnea (s)

:~ [l_ [L Without w i th

CPAP w i thou t with

Sample recording illustrating the effect of nCPAP application on respiration during sleep. AI - apnea index. See text.

Before onset of therapy, we conducted extensive internal examinations including ECG at rest and during exercise, chest X-ray, echocardiography, lung function test, blood gas analysis, and comprehensive laboratory findings as well as ENT-status and neurological examinations.

After one night of polysomnographic recordings we conducted three nights of nCPAP therapy under conditions of intensive care, as individual cases of critically prolonged nocturnal hypoxemia during the initial phase of nCPAP treatment have been reported in the literature (Krieger et al., 1983). The recordings included thoracic and abdominal respiration (inductive plethysmography), nasal airflow, partial arterial oxygen tension or saturation, nCPAP values, and an ECG.

Those patients who continued treatment after this initial phase were asked to report side-effects of nCPAP therapy after a minimum of one month of treatment. Information on complaints was collected by means of a questionnaire. Throughout the time of treatment, which was 1-17 months in the present case, the patients contacted the attending physician in our outpatient department at regular intervals.

RESULTS

The Effect of nCPAP Treatment on Sleep Apnea.

After three nights of treatment, a 90-1001 reduction of apnea could be achieved in 62 patients, and a 751 reduction in one. The average apnea index decreased from 46.5 to 1.6 episodes/h. The remaining apneas were predominantly of the central type, and the longest apnea episodes were distinctly shorter than before treatment (17.6 son average, as compared with 65.4 s before). Figure 1 illustrates the effect of nCPAP on respiration during sleep. One patient suffered from acute rhinitis, and another from claustrophobia. In both cases, therapy had to be interrupted after two days. An airway pressure of 15 mbar, the maximum available with the device we used, proved insufficient in one patient with cheilognathopalatoschisis, previously operated upon. In the remaining 63 patients, the effective nCPAP pressure values as determined during the adjustment phase were between 5 and 15 mbar. Figure 2 shows the distribution of effective nCPAP values.

Long-term nCPAP Treatment and Side-effects.

Out of 63 patiens who could be effectively treated by means of nCPAP, 60 decided to continue therapy in their homes. The patients were

40

asked to report on therapeutical success and possible complaints at regular intervals. One patient died during sleep after he had stopped nCPAP application on his own account during a spell of respiratory infection.

After 1-17 months of treatment, the following side-effects were observed: The most frequent complaint was pressure sores (69%), followed by "drying out" of the nasal and pharyngeal mucosa (46%) and rhinitis (34%). Conjunctivitis occurred in 12%, an increased frequency of respiratory infections was found in 7%, and sinusitis in 5% of the patients. None of these side effects were of a severe kind, and discontinuation of therapy was not necessary in any of these instances.

DISCUSSION AND SUMMARY

Nasal continuous positive airway pressure is an effective therapy in obstructive sleep apnea. It can be used in a vast majority of patients, rare cases with specific problems excepted. Almost 90% of the patients treated with this method consent to long-term treatment. Regular attendance to the patients' problems in a specific unit of the outpatient department may play an important part with regard to this high rate.

Side-effects were found to be comparatively frequent , but did not interfere with therapy or were recognized early during the regular check-ups and could be effectively treated. It can be expected that some of these side-effects can be avoided by further improvements in the technical apparatus involved, such as the nasal masks. Recent improvements seem to point into this direction.

n

12

10

8

6

4

2

0

nCPAP - Pressure

:5 6 7 8 9 10 11 12 13 14 1~ mba.r

Fig. 2. Distribution of effective nCPAP values as established after the adjustment phase.

41

During phases of acute infection, which may necessitate discontinuation of nCPAP, patients with severe sleep apnea run an acute risk comparable to the risk of patients with untreated sleep apnea and therefore demand particular attention and intensive monitoring and treatment durig such phases.

As nCPAP treatment was also found to have a beneficial influence on cardiocirculatory morbidity, such as cardiac arrhythmias and hypertension (Becker et al., 1987), it can be recommended as an efficient therapeutical instrument which should be employed as early as possible in patients with sleep apnea. Under conditions of regular monitoring and care it can be expected that this therapy will be well accepted and well tolerated.

REFERENCES

Becker, H., Figura, M., Himmelmann, H., et al., 1988, Die nasale 'Continuous positive airway pressure' (nCPAP)-Therapie-Praktische Erfahrungen bei 54 Patienten, Prax. Klin. Pneumol. , (in press).

Becker, H., Koehler, U., Peter, J.H., and Wichert, P., von, 1987, Reversibility of cardiac arrhythmias in sleep apnea (SA) under nasal continuous positive airway pressure therapy (CPAP), Eur. ~ Clin. Invest., 17: 3.

Cohn, M.A., 1986, Surgical treatment in sleep apnea syndrome, in: "Abnormalities of Respiration During Sleep", E.C. Fletcher, ed., Grune & Stratton, Orlando.

Issa, F., Grunstein, R., Bruderer, J., Costas, L., McCauly, V., Berthon-Jones, M., and Sullivan, C., 1987, Five years experience with home nasal continuous positive airway pressure therapy for the obstructive sleep apnea syndrome, in: "Sleep-Related Disorders and Internal Diseases", Peter, J.H., Podszus, T., and von Wichert, P., eds., Springer, Berlin-Heidelberg-New York.

Krieger, J., Sautegeau, A., Sauder, P., Weitzenblum, E., and Kurtz, D., 1984, Syndromes d'apnees du sommeil. Traitement par la pression positive continue par voie nasale, Presse Med., 13: 2559.

Krieger, J., Weitzenblum, E., Monassier, J.P., Stoeckel, C., and Kurtz, D., 1983, Dangerous hypoxemia during continuous positive airway pressure treatment of obstructive sleep apnoea, Lancet, II: 1429.

Kuhlo, W., Doll, E., and Frank, M.C., 1969, Erfolgreiche Behandlung eines Pickwick-Syndroms durch eine Dauertrachealkanule, Deut. Med. Wochenschr., 94: 1286.

Remmers, J.E., Sterling, J.A., Thorarinson, B., and Kuna, S.T., 1984, Nasal airway positive pressure in patients with occlusive sleep apnea, Am. Rev. Respir. Dis., 130: 1152.

Schweitzer, P.K., Chambers, G.W., Birkenmeier, N., and Walsh, J.K., 1987, Nasal continuous positive airway pressure (CPAP) compliance at six, twelve and eighteen months, Sleep Res., 16: 186.

Strohl, K.P., Hensley, M.J., Saunders, N.A., Scharf, S.M., Brown, R., and Ingram, R.H., jr., 1981, Progesterone administration and progressive sleep apnea, JAMA, 245: 1230.

Sullivan, C.E., Issa, F.G., Berthon-Jones, M., and Eves, I., 1981, Reversal of obstructive sleep apnea by continuous positive airway pressure applied through the nares, Lancet, I: 862.

42

THE INFLUENCE OF OBESITY ON DISORDERED BREATHING IN PATIENTS

WITH OBSTRUCTIVE SLEEP APNEA SYNDROME

Anna Brzecka

Pulmonary Clinic of Medical Academy, Wroclaw, Poland

Obesity has been long recognized as a clinical characteristic of many patients with obstructive sleep apnea (Guilleminault et al., 1976; Lugaresi et al., 1978). The purpose of this study was to relate the severity of the obstructive sleep apnea syndrome (OSAS) to the degree of obesity.

METHODS

The patients were 112 men and 10 women ageing from 18 to 80 years (mean 49 ± 13 years). The weight ranged from 61 kg to 193 kg (mean 114 ± 29 kg). In all of them the diagnosis of OSAS was established at the Minnesota Regional Sleep Disorders Center in Minneapolis, USA.

Overnight sleep studies were performed using standard polysomnographic techniques, including recordings of electroencephalogram, electrooculogram, electromyogram. Respiratory movements of the rib cage and abdomen, and their sum were measured by the respiratory inductive pletysmograph (Respitrace). Thermistors were used to detect the oronasal airflow. Arterial oxygen saturation was measured by ear pulse oximeter (Biox 2, Biox 3).

Polysomnograms were scored for sleep stages using standard criteria (Rechtschaffen and Kales, 1973); respiratory tracings were evaluated for the presence of the obstructive apnea. From oximeter tracings the

Table 1. Sleep time (in minutes) and architecture in the two groups of patients.

IO < 20 IO > 20 p

TST 362 ± 129 320 + 127 NS Stage 1 17 ± 16 22 ± 21 NS Stage 2 61 ± 19 65 ± 23 NS Stage 3+4 8 ± 19 4 ± 9 <0.05 Stage REM 14 ± 7 10 ± 7 <0.05

43

average and the lowest oxygen saturation were measured. The percent of sleep time spent in apnea and the apnea index (the number of apneas per hour of sleep) were counted.

Index of obesity, IO, was calculated kg by the cubic of his height in meters. two groups, based on the I0<20 or I0>20. obese group (n=65) and 26.6±5.2 for more

by dividing a subject's mass in The patients were divided into The IO was 17.4±2.4 for less

obese group (n=57).

Statistical significance of differences between the two groups of patients was evaluated by the t-test for independent means.

RESULTS

The results of studying sleep time and architecture in the group with I0<20 and in the group with 10>20 are shown in Table 1. No significant differences were noted between both groups with respect to the total sleep time (TST) and Stages 1 and 2 NREM sleep, whereas the stages of slow-wave sleep and rapid-eye movements sleep were shorter in the group of the more obese patients. These differences were statistically significant.

The mean apnea index and the mean percent of sleep time spent in apnea (shown in Table 2) were greater in the group of more obese patients. These differences were statistically significant.

The mean number of obstructive apneas was greater in the patients with 10>20 both in NREM and in REM sleep; no differences between the two groups were observed with regard to the apnea duration (see Table 3).

The mean oxygen saturation in awake state was not significantly different in both groups (93+12% for the patients group with 10<20 and 94.+5% for the group with 10>20). Average and the lowest arterial oxygen desaturation were lower in the group of more obese patients (as shown in Table 4) and these differences were statistically significant.

COMMENTS

The present series of 122 patients confirmed the findings of marked obesity in the majority of patients with OSAS. The precise mechanism by which excess weight contributes to the genesis of upper airway occlusion during sleep is not clear. The high indicence of obesity among patients with OSAS ( Guilleminault et al., 1976; Lugaresi et al., 1978), and the beneficial effect of weight reduction (Harman et al., 1982; Fairman and Sugerman 1982) indicate that in the presence of a mild anatomic abnormality that reduces upper airway size, the superimposition of obesity may be sufficient to allow the development of obstructive apneas during sleep. Obesity may predispose susceptible persons to upper airway occlu-

44

Table 2. Apnea Index (AI) and percent of total sleep time spent in apnea (A%TST) in the two groups of patients.

AI A % TST

IO < 20

61 ± 25 46 ± 21

10 > 20

79 ± 28 57 ± 18

p

<0.01 <0.01

Table 3. Number and duration of apneas during TST in the two groups of patients

stage IO < 20 IO > 20 p

Apnea NREM 263 + 190 356 ± 209 <0.01 number REM 35 ± 30 46 + 29 <0.05

Average NREM 26 + 6 25 + 7 NS duration (s] REM 36 + 13 42 ± 21 NS

Longest NREM 48 + 12 50 + 25 NS duration [s] REM 67 ± 39 79 ± 41 NS

Table 4. Arterial oxygen saturation (Sa02) during sleep in the two groups of patients

stage IO < 20 IO > 20 p

Average NREM 85 + 8 78 ± 10 <0.01

Sa02(%) REM 76 + 16 65 ± 14 <0.01

Lowest NREM 78 + 11 67 + 15 <0.01

Sa02(%) REM 68 + 21 52 ± 19 <0.01

45

sion indirectly, secondary to a decrease in lung volume (Hoffstein et al., 1984).

The patients in both groups had severely disturbed sleep architecture with predominant Stages 1 and 2 NREM sleep. More obese patients had less rapid-eye-movements sleep and had lower percentage of slow wave sleep. The reasons for more profound sleep disturbances are probably related to the more frequent obstructive apneas in the more obese group.

The present study correlated the degree of obesity with apnea number, the percent of sleep time spent in apnea, and arterial oxygen desaturation. Thus, although the higher degree of obesity did not separate the patients with respect to the duration of apneas, the severity of sleep apnea phenomenon could be related to the excess of weight.

REFERENCES

Fairman, R.P., and Sugerman, H.J., 1982, Gastroplasty for obstructive sleep apnea and morbid obesity (abstract), Am. Rev. Respir. Dis., 125(Suppl): 108.

Guilleminault, C., Tilkian, A., and Dement, W.C., 1976, The sleep apnea syndromes, Ann. Rev. Med., 27: 465.

Harman, E.M., Wynne, J.W., and Block, A.J., 1982, The effect of weight loss on sleep disordered breathing and oxygen desaturation in morbidly obese men, Chest, 82: 291.

Hoffstein, V., Zamel, N., and Phillipson, E.M., 1984, Lung volume dependence of pharyngeal cross-sectional area in patients with sleep apnea, Am. Rev. Respir. Dis., 130: 175.

Lugaresi, E., Coccagna, G., and Mantovani, M., 1978, "Hypersomnia with periodic apneas", Spectrum Publications, New York.

Rechtschaffen, A., and Kales A., 1973, "A manual for standardized technology, technique and scoring system for sleep stages of human subjects", Brain Information Service, Los Angeles.

46

CHRONIC OBSTRUCTIVE PULMONARY DISORDERS (COPD) AND SLEEP

A. Gianotti, P. Moscatelli, and N. Franconieri

Universita di Genova, Italy

ABSTRACT*

Alveolar hypoventilation and alterations in the distribution of ventilation perfusion ratios are the predominant mechanisms of hypoxemia during sleep with related transient episodes of pulmonary hypertension. We study, by all night polysomnography, 17 patients with severe chronic obstructive pulmonary diseases (mean awake Pa02 9 KPa, mean FEV, 30 per cent predicted normal). Preliminary data show that only 5 subjects had significant episodes of apnea during sleep. We did not find the cyclic precipitous falls in Sa02 so characteristics of sleep apnea syndrome, althought in 14 patients (82%) there are sleep disturbances (especially during REM sleep). We think that the majority of patients with COPD do not have the classical sleep apnea syndrome. Furthermore, since the frequency of arousal does not decrease significantly following nocturnal oxygen therapy, we emphasize that, in our study, the 5 patients with sleep apnea syndrome, had also the higher PaC02. It could be possible that not hypoxemia, but hypercapnia is the principal arousal stimulus.

* The full MS has not been received

47

POLYSOMNOGRAPHIC FINDINGS IN PATIENTS WITH CHRONIC OBSTRUCTIVE

PULMONARY DISEASE (COPD)

E. Gozlikirmizi, N. Yildirim, H. Kaynak, S. Madazlioglu, H. Denkta•, and F. Yenel

Department of Neurology and Pulmonary Diseases Cerrahpasa Medical Faculty, University of Istanbul Turkey

Beginning from 1970's, studies on respiratory physiology during sleep and physiopathological events in pulmonary diseases have given background to the therapy of patients with pulmonary problems (Williams, 1978). In some specialized centers the patients with COPD are studied extensively and the effects of sleep on respiratory functions and the sleep disorders originating from respiratory problems in these patients have gained new dimensions (Wynne et al., 1979). Some of these findings have shown that polysomnography can help the clinician (Guilleminault et al., 1980).

SUBJECTS AND METHOD

14 patients (13 men and 1 woman) with COPD, (the diagnosis was verified by complete clinical and laboratory methods, Tab. 1.), were studied. The cases were all non-obese, mean age of 64 ± 10 (range 46 to 79 yrs). The parameters, i.e. EEG, ECG, chin EMG, leg EMG, nasal and oral airflows, thoracic and abdominal movements, EOG, were monitored while the patients slept. During this step, the behavior of patients was inspected and recorded, taking into account the presence or absence of snoring. Later on, the number, duration and the distribution of sleep stages (Rechtschaffen and Kales, 1968) and of the respiratory events (apnea and hypopnea) were evaluated. After this evaluation, the clinical characteristics of the cases were grouped depending upon C02 and 02 blood level. The polysomnographic data were correlated with the respiratory parameters. In this preliminary report, no statistics could be done because the number of cases was insufficient.

Apnea during sleep has been defined as an interruption of air flow for longer than 10 s and hypopnea as a decrease in respiratory frequency, but not a complete cessation of breathing.

RESULTS

Apneas were quantified for each patient, and a sleep apnea index was calculated as the number of apneas per 60 minutes. The same calculations were done for hypopneas and the (A+H) index for each patient was shown to

49

(11

0

Tab

le

1.

------------------

RV

/ FV

C1

FVC

1/

No

Nam

e A

ge

Pa0

2 Pa

C02

pH

TL

C (m

l)

FVC

TIB

L

aten

cy

T.S

.T

A.I

H

.I

(A+

H)I

'.t

'.t

----

----

----

----

----

----

----

----

----

----

----

----

----

----

----

----

----

----

-1

c.s

57

52

49

.6

7.2

8

65

1047

40

3

88

' 3

5'

52

' 2

7.8

6

.9

34

.7

2 M

.:il

76

42

.7

54

7.3

4

67

771

61

41

8'

26

' 9

2'

8.2

5

.8

14

3 :i)

.A

66

66

.3

64

7.3

1

77

413

38

40

0'

20

' 2

72

' 7

.2

20

.7

27

.9

4 H

.N

79

38

.5

79

7.2

5

78

413

38

46

4'

25

' 3

87

' 1

0.0

6

.9

16

.9

5 K

.Y

46

57

.7

58

.7

7.3

5

-10

19

43

33

1'

10

' 2

32

' 0

0 0

6 D

.A.S

52

6

7.7

5

2.7

7

.29

71

84

1 36

1

59

' 1

4'

86

' 0

0 0

7 M

.O

58

41

.6

69

.2

7.2

8

67

909

59

17

0'

19

' 1

49

' 0

0 0

8~

A.K

. 65

6

5.2

4

9.9

7

.33

-

496

44

39

3'

1'

26

0'

0.2

1

.0

1.2

9

O.N

.S

67

61

.3

40

7.4

0

-49

1 38

3

23

' 2

8'

17

5'

2.7

2

.7

5.4

10

L

.K

71

62

39

7.4

1

69

716

33

28

2'

31

' 2

02

' 1

.4

7.1

8

.5

11

R

.K

54

75

38

.7

7.4

0

-64

8 44

4

43

' 7

8'

31

4'

0.7

0

.2

0.9

12

I.

O

75

50

41

.6

7.2

8

-77

1 36

3

54

' 5

' 2

72

' 0

3.3

3

.3

13

A.D

60

6

7.7

4

1.4

7

.37

-

936

37

24

7'

12

' 2

10

' 2

0 2

14

H.O

74

6

5.0

3

2.0

7

.41

54

12

48

58

35

5'

18

' 2

09

' 0

.2

0.5

0

.7

---------------------------------------------------------------------------------------------

~ F

emal

e

~ E06<U

Et6

·~~~~~~)~)J)~)J;,.J,.J,.J,.J,.J,.J,.J~

Fig. 1. See text

be greater than 5 in six of the 14 cases. Four patients were from the hypercarbic group and two from the hypoxic one. In other words, the abnormal respiratory events ratio during sleep was 4/8 in hypercarbic group and 2/6 in the other one.

The sleep apnea syndrome occured in all four patients in the hypercarbic group. Even disregarding hypopneas, the apnea index of these patients was over 5. However, the abnormal respiratory events in the form of hypopnea was predominant for two cases in the hypoxic group. From the hypercarbic patients demonstrating the sleep apnea syndrome, the (A+H) I values were found as 34.7-14-27.9 and 16.9. The hypoxic patients with the (A+H) I over 5 had the exact values equal to 5.4 and 8.5. The other eight patients (hypercarbic and hypoxic) whose (A+H) I was under 5 had individual values 0 for three patients, under 1 for two patients and between 1 and 3 for another three patients.

Among hypercarbic cases, four patients having (A+H) I over 5 were evaluated. Two of them had apneas of both central and obstructive types while the other two had apneas of central type. These two later cases have also shown apneas lasting 10-15 seconds during short drowsiness periods defined by EEG. These apneas had affected the initiation of sleep (see Fig. 1).

When the apneas and the hypopneas are distributed to sleep stages I, II, III+IV and the REM sleep, the (A+H) I is found as 36, 11, 10 and 7, respectively. Besides the linear decrease observed in hypoxic patients

51

the (A+H) I in hypercarbic patients shows a plateau as the sleep deepens (Fig. 2).

Hypoxic and hypercarbic patients with or without respiratory disorders have been compared by means of pulmonary function tests and blood gas values but no meaningful differences were found.

40

35

30

25

20

15

10

5

CONCLUSION

~ I I I I I I I I I

--n=6

--- n =4

- -- -- n=2

I

Q.._

'

I I I I I b..--~---<>

-..,

III•IV REM

Fig. 2. See text

It has been observed that sleep apnea syndrome occurs in 1/2 of the hypercarbic patients whilst abnormal respiratory events (frequent hypopneas) occured in 1/3 of the hypoxic patients . These respiratory disturbances during s leep were predominant during stages I and II.

The age of 65 is critical for the patients with COPD and the night time respiratory disorders observed are more frequent in elderly patients.

SUMMARY

14 patients (13 men and 1 woman) with chronic obstructive pulmonary disease (COPD) were monitored during their spontaneous night sleeps. The polysomnographic findings and clini ca l data were compared between two subgroups i.e. hypercarbic and hypoxic patients. Half of the hypercarbic patients have shown sleep apnea syndrome and 1/3 of hypoxic patients have shown frequent hypopneas. The abnormal respiratory events are predominant in the sleep stages I and II among these patients. After the age of 65 in patients with COPD, respiratory disorders during sleep are more frequent.

52

ACKNOWLEDGEMENTS

This work was supported by Istanbul University Research Fund. Project Number 76-77j284III

REFERENCES

Guilleminault, C., Cummiskey, J., Motta, J., 1980, Chronic Obstructive Airflow Disease and Sleep Studies, American Review of Respiratory Disease, 122: 397.

Rechtschaffen, A., Kales, A., 1968, A Manual of Standardized Terminology Techniques and Scoring Systems for Sleep Stages of Human Subjects. USPHS Pub. No.204, U.S. Government Printing Office, Washington, D.C.

Williams, R.L., 1978, Sleep Disturbances in Various Medical and Surgical Conditions, in: Sleep Disorders, R.L. Williams, I. Karacan, ed., John Wiley and Sons, New York.

Wynne, J.W., Block, A.J., Hemenway, J., Hunt, L.A., Flick, M.R., 1979, Disordered Breathing and Oxygen Desaturation During Sleep in Patients with Chronic obstructive Lung Disease (COLD), The American Journal of Medicine, 66: 573.

53

HIGH AND FLUCTUATING MUSCLE NERVE SYMPATHETIC ACTIVITY IN THE

SLEEP APNEA SYNDROME: A PATHOGENETIC MECHANISM IN THE DEVELOPMENT

OF HYPERTENSION ?

Jan Hednerl, Johan Sellgren2, Hasse Ejnell3, and Gunnar Wallin4

Departments of Clinical Pharmacology!, Anesthesiology2 Oto-Rhino-Laryngology3 and Clinical NeurophysiologyijSahlgren's University Hospital, Goteborg, Sweden

Muscle sympathetic activity was measured in 4 patients with sleep apnea syndrome (SAS). All patients exhibited increased base line sympathetic activity when awake as well, as during sleep, compared to healthy control subjects. Increased sympathetic activity was also measured during central and obstructive apneic events during sleep. Immediately after arousal and initiation of ventilation, the sympathetic activity decreased to very low values. The increased sympathetic autonomic discharge seen in sleep apnea patients may be of pathophysiological importance for the hypertension and nocturnal ventricular arrhythmias reported to occur in SAS patients.

A comprehensive explanation for the hemodynamic disturbances occuring in the sleep apnea syndrome (SAS) is not readily available. Several mechanisms have been suggested in the pathophysiology of the systemic hypertension (Koles et al., 1984; Koskenvua et al., 1987) or ventricular arrhythmias (Scharf, 1984) often seen in these patients. Cyclic variations of heart rate, i.e. bradycardia during apnea followed by the abrupt onset of tachycardia have been reported in patients with SAS (Tilkian et al., 1977). The initial bradycardia during apnea can be effectively inhibited by atropine sulphate (Guilleminault et al., 1984), while tachycardia associated with the end of apnea and the arousal response is unaffected by propranolol (Guilleminault et al.,1984). Cyclical heart rate patterns are not seen in patients with impairment of the autonomic nervous control of the heart (heart transplant patients, SHY-Drager syndrome) (Guilleminault et al., 1984).

Both hypoxemia and hypercapnia associated with apneic episodes can lead to increased sympathetic autonomic discharge via activation of central and peripheral chemoreceptors (Scharf, 1984). The decreased plural pressure associated with a MUller manoeuvre to overcome an obstructive apnea leads to decreased right atrial pressures and to an augmented venous thoracic inflow (Guyton et al., 1957). Moreover, the low pleural pressure results in a significant increase in left ventricular end-diastolic and end-systolic volume (Scharf et al., 1979). Direct evidence for increased sympatho-adrenal discharge during sleep apneas has not yet to our knowledge, been presented.,

55

The aim of_the present study was to evaluate the symphathetic activity directly by measuring efferent postganglionic muscle sympathetic activity in the peroneal nerve in patients with SAS.

PATIENTS AND METHODS

Four patients were referred to the Sleep Clinic of Sahlgren's University Hospital, GHteborg, Sweden, because of sleep problems such as day time somnolence, disrupted nocturnal sleep, and heavy snoring, or spouse observation of apneic events during sleep. The patients were subjected to an overnight continuous monitoring of oxygen saturation and heart rate. They were diagnosed as having a sleep apnea syndrome, i.e. a pathologically high apnea - hypopnea index (number of apneic and hypopneic episodes per hour of sleep). Furthermore, a separate 2- 4 hour daytime polygraphic monitoring was performed.

The measurements of muscle sympathetic activity during sleep were performed during a daytime recording. An oral consent was obtained from each patient. Approval for these studies was obtained from the Ethics Committee at the Medical Faculty, Gothenburg University. The patients were connected to an oxymeter (Bios 3700). Nasal and oral airflow were monitored with thermistors. Respiratory movements were monitored via thoracic strain gauge encircling the chest and blood pressure was measured with a strain gauge pressure transducer connected to brachial arterial catheter. Multiunit postganglionic sympathetic efferent activity was led off from peroneal nerve muscle fascicles with a tungsten microelectrode with an unisolated tip diameter of about 1-5 um. Reference electrodes with larger unisolated tips were inserted subcutaneously 1-2 em from recording electrodes. The electrodes were connected to a differential preamplifier with a gain of 1000 and to an amplifier with a gain of 50. Nerve traffic was monitored with a storage oscilloscope and a loudspeaker. The raw nerve signal was fed trough a band pass filter with a bandwidth of 700-2000 Hz and through an amplitude discriminator to reduce remaining noise. A resistance-capacitance integrating network with a time constant of 0.1 s was used to derive mean voltages of nerve activity. These methods have been described earlier (Sundlof and Wallin, 1977). All measurements were recorded with an inkwriting oscillograph and an FM tape recorder. Data was analysed manually and with the aid of a digital computer and a processing digital oscilloscope (Norland 3001).

RESULTS

Both during wakefulness and during sleep not associated with apneic breathing pattern, an unusually high and variable activity in muscle nerve sympathetic activity was found. During sleep apnea a characteristic pattern in the integrated muscle nerve sympathetic activity was found (Fig. 1).

In general, the activity pattern did not differ between central and obstructive apneic events. Sympathetic activity was continuously increased during the apnea to be suddenly interrupted in association with the onset of respiration. No apparent changes in heart rate were seen in association with apneic events in these patients. However, mean blood pressure was highly variable and seemed to show some association with the apneic breathing pattern. In these patients, EEG monitoring was not performed and it is not known whether the apneas and oxygen desaturations were associated with arousal.

56

DISCUSSION

Mu&C;Ie "''"' symp•tt"'ht •ct!Vrly

Hearl rate

(mrn· 1 )

Ae$P«aiOty movemtn1$

N051!1

I Aif IIOW

\ tdoulh

Q)liyQtft salur at-on

SaO. (")

200t 100

0

.·~ l ~ fJt '

\00[ 50 [

0

1001 ! .. ~ 50

Fig.1. A representative recording of apneic events during sleep in a patient with SAS. Shown is the integrated muscle-sympathetic activity, arterial blood pressure, heart rate, thoracic movements, nose and mouth air flow and oxygen saturation curve. The time scale is indicated at the bottom of the figure.

Moderate hypoxemia is known to be associated with a sympathetic autonomic circulatory response consisting of increases in systemic arterial pressure, cardiac output, cardiac contractility and peripheral vascular resistance (see Sylvester et al., 1979 for review). The mechanism of these hemodynamic changes associated with hypoxemia is not fully clarified. However, the response might involve a release of catecholamines as well as a sympathetic response mediated via the carotid body reflex (Sylvester et al., 1979). Systemic hypertension has been shown to be overrepresented in patients with SAS (Koles et al., 1984; Koskenvua et al., 1987) and hemodynamic studies of SAS patients during wakefulness and sleep indicate that the expected progressive fall in arterial pressure during normal sleep was largely absent (Tilkian et al., 1976). In a study by Tilkian et al. (1976) a characteristic pattern of cyclic arterial pressure elevation which disappeared after several minutes of normal ventilation or wakefulness was seen with each apneic episode. The episodes of obstructive sleep apnea were also associated with bradycardia most likely due to increased vagal tone secondary to the MUller manoeuvre, i.e. an attempted breath against a closed glottis. Resumption of apnea, however, was associated with a sinus tachycardia

57

which was suggested to be related to a removal of marked vagal tone. Such an interpretation is supported by the finding that bradyarrhythmias were inhibited by a vagal blockage by atropine (Tilkian et al., 1977)). The present data indicate that cyclic systemic hypertension might be secondary to an increased sympathetic discharge and peripheral vasoconstriction triggered by hypoxemia, and possibly also by respiratory acidosis associated with the apnea. The prominent "off-switch" of muscle nerve sympathetic activity may indicate a baroreceptor mechanisms decreasing sympathetic outflow. Whether such a repetitive activation of baroreceptors results, in turn, in pathological set points for the baroreceptor response to increases in pressure remains to be studied. However, it is a possible mechanism to explain the increased and highly variable activity in muscle nerve sympathetic activities measured in these patients.

The mechanism behind the continuous increase of sympathetic activity recorded during the course of apnea may also be due to direct chemosensory mechanisms. In animal experiments splanchnic nerve activity has been shown to increase linearly with increased inspired C02 in anaesthetised rats (Elam et al., 1981). In this, the activity of the dominating noradrenergic nucleus in the pontine region of the rat, locus coeruleus, as well as peripheral splanchnic nerve activity was measured during hypercapnic and hypoxic stimuli. Both types of stimuli increased central noradrenergic activity, while an increase in splanchnic nerve activity appeared only during hypercapnic conditions. As increased arterial C02 is a modest but yet well identified phenomenon of sleep apnea (Tilkian et al., 1976); this might be one of the mechanisms involved in the increased muscle nerve sympathetic activity recorded in SAS patients.

In conclusion, an increased, highly variable muscle nerve sympathetic activity was found in four SAS patients. A typical pattern with a continouously increasing sympathetic activity during the apnea, followed by a sudden "off-switch" of the activity, was recorded in all patients so far studied. The increased sympathetic autonomic activity associated with SAS may be of importance in the development of the sustained pulmonary hypertension and possibly the systemic hypertension commonly seen in these patients.

ACKNOWLEDGEMENTS

The work has been supported by The Swedish National Association against Heart and Chest Diseases. The secretarial aid of Elly Kihl is gratefully acknowledged.

REFERENCES

Elam, M., Yao, T., Thoren, P., and Svensson, T.H., 1981, Hypercapnia and hypoxia: Chemoreceptor-mediated control of locus coeruleus neurons and splanchnic sympathetic nerves, Brain Res., 222: 373.

Guilleminault, C., Winkle, R., Connolly, S., Melvin K., and Tilkian, A., 1984, Cyclical variation of the heart rate in sleep apnea syndrome. Mechanisms and usefulness of 24h electrocardiography as a screening technique, Lancet, 2: 126.

Guyton, A.C. Lindsey, A.W., Abernathy, B., and Richardson, T., 1957, Venous return at various right atrial pressures, and the normal venous return curve, Am. ~ Physiol., 189: 609.

58

Koles, A., Cadieux, R.J., Shaw, III L.C., Vela-Bueno, A., Bixles, E.O., Scheck, D.W., Loche, T.W., and Soldatos, C.R., 1984, Sleep apnea in a hypertensive population, Lancet, 1: 1005.

Koskenvua, M., Kaprio, J., Telakivi, T., Partinen, M., Heikkila, K., and Sarna S., 1987, Snoring as a risk factor for ischaemic heart disease and stroke in men, Brit. Med. ~. 294: 16.

Scharf, S.M., Brown, R., Tow, D.E., and Parisi, A.F., 1979, Cardiac effects of increased lung volume and decreased pleural pressure in man, ~ ~ Physiol., 47: 257.

Scharf, s,, 1984, Influence of sleep state and breathing on cardiovascular function, in: "Sleep and breathing", Saunders and Sullivan, eds., Marcel Dekker Inc.

Sundlof, G., and Wallin, B.G., 1977, The variability of muscle nerve sympathetic activity in resting recumbent man,~ Physiol., 272: 383.

Sylvester, J.T., Scharf, S.M., Gilbert, R.D., 1979, Fitzgerald, R.S., and Traystman, R.J., Hypoxic and CO hypoxia in dogs: hemodynamics, carotid reflex and catecholamines, Am.~ Physiol., 236: 422.

Tilkian, A.G., Guilleminault, C., Schroeder, J.S., Lehrman, K.C., Simmons, F.B., and Dement W.C., 1976, Hemodynamics in sleep-induced apnea, Ann. Intern. Med., 85: 714.

Tilkian, A.R., Guilleminault, C., Schroeder, J.S., Lehrman, K.L., Simmons B.L., and Dement W.C., 1977, Sleep induced apnea syndrome: relevance of cardiac arrhytmias and their reversal after tracheostomy, Am.~ Med., 63: 348.

59

CEPHALOMETRY FOR EVALUATION OF GEOMETRY OF THE UPPER AIRWAY

A. Kukwa, I. Fleszar,

B. de Berry-Borowiecki, R. H. I. Blanks, A. Komorowska, and M. Ryba

ENT Department, Institute of Surgery, Medical Academy Warsaw, Poland

Department of Anatomy and Surgery, University of California Irvine, USA

Many authors confirm that the main cause of peripheral or central obstructive sleep apnea (OSA) is the neuromuscular dysfunction in oropharyngeal region (Brouillette and Thach, 1979; Remmers et al., 1978). However, in OSA patients anatomostructural abnormalites are present which may be primary to this physiological abnormality.

Changes in geometry of the mandibulofacial skeleton affect the framework on which the musculature of tongue, uvula, pharynx and the hyoid bone is suspended. On the basis of these anatomical disproportions of the frame, functional pattern of the oropharyngeal musculature is modified. Disharmony in distribution of the spots to which oropharyngeal muscles are attached is responsible for abnormalities of the lumen of the upper airway. Although the majority of OSA patients have no synonymously definable and documented lesion, yet many of them present a significant number of anatomical abnormalities involving upper airway segment (Wilms et al., 1982).

Both qualitative and quantitative description of sleep apnea can be easily done by means of polysomnography (PSG), whereas defining the area, or only the level of pharyngeal wall collapse, still present many difficulties.

Initially endoscopic examination seemed sufficient to diagnose the level of obstruction, but it does not. Although the pharyngofiberoscopy with the registration of Mullers manoeuvre and the cinefluoroscopy have still remained the most frequently applied method, with the use of them it is not possible to define precisely either the spot, or the extensiveness of the airway collapse. For this reason in UCI in 1983 a method of examining facial bones applied by orthodonts - the so-called cephalometry - was adopted. Our modification of this technique allows quantitative description of the geometry of the upper airway.

The cephalometric measurements obtained from OSA patients were compared with analogical data obtained from the control group. The

61

analysis of the results revealed that as many as 28 out of 59 chosen measurements were indicating statistically significant differences. We conclude that OSA patients have a significantly enlarged tongue and elongated soft palate. Also, the maxilla is displaced posteriorly, compared to the control group. In OSA patients there is an evidence of micrognathia and retrognathia. However, the most relevant element of the examination seems to be the possibility of determining the position of the bone: in OSA patients it tends to be displaced downwards and backwards, compared to the control group.

These findings encouraged us to further investigations, especially because similar results were obtained by Riley and coworkers (1983). The sitting position of the patients during x-ray seemed unsuitable for full evaluation of patency of the upper airway. Therefore we decided to perform cephalometric examinations in such a position of the patient which most closely resembles his position during sleep.

METHODS

The patients examined were young people with breathing disorders during sleep which appeared in consequence of narrowing the upper airway after the pharyngoplasty (applied for correction of cleft palate).

Lateral x-ray was performed by means of the polytomograph. The patients were in supine position. All the measurements: linear, angular and the cross-section were performed in the way already defined in our previous paper (DeBerry-Borowiecki et al., in the press). Before the examination each patient was specially prepared: he was instructed to swallow the saliva, then relax his throat muscles and stop breathing at the end of his normal expiration. The lamp of the polytomograph was placed in a standard way, in such a distance from the patient's head that the structures were imaged in 1:1 proportion. The measurements were taken after indicating all anthropometric points. The numerical data thus obtained were compared with the average from the previous studies.

Fig. 1 shows the anthropometric landmarks used to define the geometry of the upper airway.

Table 1 . HYOID BONE: craniometric differences in location of hyoid bone in control groupe and OSA patients in sitting and supine position

MEASUREMENT OSA PATIENTS CONTROL GROUP OSA PATIENTS

(distance, angle, IN SITTING n = 12 IN SUPINE cross-section) POSITION POSITION

(mean ± S.D.) (mean ± S.D.)

Ar - H (mm) 105 ± 8 91 ± 10 102.1 + 8.4 s - H (mm) 129 ± 8 116 ± 14 116.9 + 13.2 Go - H (mm) 41 ± 7 34 ± 10 30.9 + 8.8 Gn - H (mm) 53 + 9 52 ± 9 49.8 ± 6.3 H - phW (mm) 34 + 5 33 ± 6 31.5 ± 8.5 ~ Go-Gn-H (deg) 32 + 6 20 ± 8 24.1 ± 14.6 ~ N-S & Ar-H (deg) 83 + 5 75 ± 9 83.3 ± 7.1 ~ N-S-H (deg) 93 ± 4 89 ± 5 88.8 + 3.5 ~ Cn-GO & N-S (deg) 34 ± 8 32 ± 11 31.8 ± 3.5

62

RESULTS

Fig. 1. Anthropometric landmarks: A - subspinale, ANS - Anterior nasal spine, Ar - articulare, B - supramentale, Et - tip ot epiglottis, Fh -Frankfurt horizontal, Gn - gnathion, Go - gonion, H - hyoid bone, N - nasion, 0 - orbitale, P - porion, PhW - posterior pharyngeal wall, PNS - posterior nasal spine, S - sella, Sc - sphenoidal crista, Sr - sphenoidal rostrum, Tb - tongue base

It was generally maintained, hitherto, that the bulk of the tongue within mesopharynx constitutes the main danger of obstruction of the airway. The results obtained in the present study indicate that in the supine position of the patient (which is a natural sleeping position) his hyoid bone becomes significantly displaced.

The difference due to the patient's position is especially visible in that the numerical proportion the Go-H and Gn-H measurements gets reversed, compared to the previously presented data in OSA patients. Displacing the hyoid bone backwards creates new static, as well as dynamic, conditions within the pharyngeal area. It is the glossa-hyoid muscle attached to the hyoid bone which causes displacing the bulk of the tongue both backwards and upwards.

63

DISCUSSION

In our material we have not observed any evidence of lowering the hyoid bone, which was found previously. It seems that in some OSA patients collapsing of the lumen of the airway reveals during sleep due to anatomical changes which lead to loss of tightness of the pharyngeal wall, while the hyoid bone is not displaced.

Cephalometric studies should be perfomed not only in patients whith OSA, but also in all cases with oropharyngeal abnormalities involving lympathic and another soft tissue overgrowth. This examination may give much information about patency of the upper airway and a very strong suggestion on how to correct it surgically. The applied technique of anthropometric studies can be used for total evaluation of patency of the upper airway. No matter what kind of examination is used for visualisation of the skeleton and contour of the upper airway, as far, as the measurements give us sufficient information about it. In our opinion, for these studies we may use a regular lateral x-ray or all kinds of films as well as CT-scan. Yet, the main point is, that all landmarks have to be marked and all measurements have to be compared to the same data taken from the patients of the control group. The possibility that this technique may be used in many ways seems to confirm its versatility. It is very useful for the evaluation of patency of the upper airway in diagnosis, and also for estimation of postoperative status of patients.

REFERENCE

Brouillette, R.T., and Thach, B.T., 1979, A neuromuscular mechanism maintaining extrathoracic airway patency, ~ ~ Physiol., 46: 772.

DeBerry-Borowiecki, B., Kukwa, A., and Blanks, R.H.I., 1988, Cephalometric analysis for diagnosis and treatment of obstructive sleep apnea, Otolaryngol. Head Neck Surg. (in press).

Riley, R., Guilleminault, C., Herran, J. et al., 1983, Cephalometric analysis and flow volume in obstructive sleep apnea patients, Sleep, 6: 303.

Remmers, J.E., DeGroot, W.T., Sauerland, E.K., et al., 1978, Pathogenesis of upper airway occlusion during sleep, ~ ~ Physiol. 44: 931.

Wilms, J.E., Popovich, J., Conway, W., Fujita, S. et al., 1982, Anatomic abnormalities in obstructive sleep apnea, Ann. Otol. Rhinol. Laryngol., 91: 595.

64

THE CLINICAL RELEVANCE OF VERY SEVERE PURE DIAPHRAGM WEAKNESS

C. Laroche, N. Carroll, A. Mier, C. Brophy, and M. Green

Brompton Hospital, London, U.K.

ABSTRACT~

Previous studies suggest that patients with bilateral diaphragm paralysis are at high risk of developing chronic respiratory failure due to alveolar hypoventilation. However such studies mostly describe patients with generalised neuromuscular disorders in which other respiratory muscles are also weak. We therefore studied three patients with diaphragm paralysis (two due to neuralgic amyotrophy, one following traumatic diaphragm rupture) but no weakness of the other respiratory muscles. All were breathless on exertion (MRC Grade 4) with orthopnoea and abdominal paradox but with no symptoms of nocturnal hypoventilation. Maximum expiratory mouth pressure was >100% pred. consistent with normal intercostal and abdominal muscle strength. Vital capacity was 45-65% pred., falling by 36-55% when supine. Transdiaphragmatic pressure (Pdi) was zero during (a) tidal breathing, (b) inspiration to TLC,_and (c) supermaximal stimulation of both phrenic nerves. It was also very low during a maximum sniff (10, 7.5 and 12.5 em H20 respectively, normal >100 em H20). Maximum inspiratory mouth pressure was 35-70% pred. Arterial blood samples sitting at rest showed a p02 of 10.6-12.3 kPa and normal pC02 of 4.7-5.5 kPa. Continuous overnight monitoring on two consecutive nights, in normal sleeping positions (one on his side, two propped up), using an ear oximeter (Biox 3) and a Hewlett Packard capnometer, showed no significant p02 desaturation nor any increase in pC02. These patients have now remained stable for one to four years without developing any signs of nocturnal hypoventilation or respiratory failure, nor have daytime blood gases deteriorated. We conclude that in the absence of generalised respiratory muscle weakness, severe diaphragm weakness does not lead to respiratory failure.

* The full MS has not been received

65

CARDIAC INVOLVEMENT IN OBSTRUCTIVE SLEEP APNEA SYNDROME,

(OSAS) - CASE REPORT

M. MigdaJ, K. Kubicka, W. Kawalec, L. OrJowski, M. Zubrzycka, and P.S. Gutkowski

Child Health Center, Warsaw, Poland

The causes of OSAS in children are numerous and range from neuromuscular and endocrine disorders to craniofacial anomalies or nasal congestion (Hunt and Brouilette 1982). Tonsil and adenoid hypertrophy is probably the most frequent cause of OSAS in children. The studies of Noonan (1965) and Menashe et al. (1985) established that enlarged tonsils

I

m ~ """""'\.;"; \.;'; ~ \.!'> \r """\., .~ (V'--J\j ~ ~ :lVR ' ' I I I

-v~~~ v"""-.r" V4 ' 1

~ (\-.J ~ ('----; ('--' (\----'\.; ~ (V"""\-~ (\l"'u 1\i

Vs ' '-"-. ~ ~ ~ /""'-'\.. r

aVF

Fig. 1. ECG before treatment.

can lead to upper airway obstruction during sleep. Chronic alveolar hypoventilation and cor pulmonale may occur as a consequence. Other cardiovascular involvement reported in children with OSAS are systemic and pulmonary hypertension, cardiomegaly, arhythmias.

67

There are only a few studies (eg. Noonan, 1965) reporting evaluation of cardiac changes (including ECG or chest X-ray ) in children with OSAS. The aim of the present study was to point out echocardiographic and ECG changes in a very severe OSAS patient prior to, and after tonsillectomy . An eight years old boy was admitted to the Department of Cardiology because of nocturnal cyanotic attacks, signs of congestive heart failure and transient arterial hypertension. Previously he suffered from recurrent pneumonias and hypertrophy of tonsils and adenoids . On

Fig. 2. Echocardiographic pictures before treatment.

admission his clinica l state was poor . Dyspnea and peripheral cyanosis without clubbing were observed. There was no chest deformity or trill. The left border of the heart was shifted to the left axillary line. The pulmonic component of the second heart sound was markedly a ccentuated. There was no murmur. The lungs were clear on auscultation. The liver was palpable 4 em below the right costal margin . The tonsils were markedly enlarged. Laboratory investigations were normal except red blood cells count and hematocrit elevation (9 700 000 and 581, respectively). Arterial hypercapnia with compensated r espiratory acidosis was al so present . On ECG right vent r icular and right atrial hypertrophy was found.

Chest X-ray revealed heart enlargement . During lung function measurement reduced patency of upper airways ( i n terms of decr eased

68

MIF5o/MEF5o ratio) was found. Echocardiographic examination revealed normal anatomy of the heart and great vessels, except hypertrophy of the ventricular walls and interventricular septum. Enlargement of right ventricle and right atrium was found. The interventricular septum was bulging toward the left ventricle.

Fig. 3. Echocardiographic pictures before treatment.

Cardiac catheterisation and angiocardiography showed no evidence of right to left or left to right shunting. Pressures in the right ventricle and pulmonary artery were normal at rest. Oxygen saturation in arterial blood was 90%. After administration of 100% oxygen the saturation rose to 100%. During cardiac catheterisation an incident of cyanosis with the right ventricular and pulmonary artery pressure elevation to the systemic level occured. The pressure elevation in the right atrium to mean value of 20 mm Hg was also observed. The oxygen saturation in pulmonary trunk fell to 271. After the intubation hypoxemia disappeared.

The previous medical history and all physical and laboratory findings prompted us to perform nocturnal polysomnography. Thoracic movements, ECG, transcutaneous P02 (tc P02), capnography (C02), 02 saturation (Sat 02) and pulmonary artery pressure (PA press) were simultaneously recorded. The polysomnographic results allowed to diagnose OSAS (number of obstructive apneas > 5 s per 100 min of recording: 18.6; mean duration of apnea: 13.9 s.; range 5-60 s) with marked decrease of Sat 02 (max diff.75%), tc P02 (max diff.45 mmHg) and rise of end tidal C02 (max 80%) and PA pressure (up to 80 mm Hg). Tonsillectomy was done. The second polysomnography performed 10 days later showed considerable improvement (number of obstructive apneas

69

n V1

uVF __..,_

Fig. 4. ECG after tonsillectomy.

Fig. 5. Echocardiographic picture after tons illectomy

70

> 5 s/100 min of recording: 6.2; mean duration of apnea: 6 s.; range 5-8 s.). tc P02 and Sat 02 were in normal ranges.

After the operation neither cyanosis nor signs of a congestive heart failure were found. Two months after surgery ECG and all laboratory tests were normal.

Four months later echocardiographic examination showed normal size of the right atrium and right ventricle. The position of the interventricular septum was normal.

In conclusion, the patient had symptoms and signs of chronic hypoventilation and cor pulmonale. The diagnosis of OSAS was based upon patient's history, physical examination and polysomnography. After tonsillectomy all cardiac abnormalities demonstrated in this child had completely disappeared. However, their long term reversibility needs further observations.

REFERENCES

Hunt, C.E., and Brouillette, R.T., 1982, Abnormalities of breathing control and airway maintenance in infants and children as a cause of cor pulmonale, Pediatric Cardiology, 3: 249.

Menashe, V.D., Farrehi, C., and Miller, M., 1965, Hypoventilation and cor pulmonale due to chronic upper airway obstruction,~ Pediatr., 67: 198.

Noonan, J.A., 1965, Reversible cor pulmonale due to hypertrophied tonsils and adenoids: studies in two cases, Circulation, 32: 164.

71

RESPIRATORY PATTERNING AND ARTERIAL OXYGENATION DURING SLEEP

IN LARYNGECTOMISED PATIENTS

INTRODUCTION

W. Oldfield, L. Sawicka, M.S. Meah, and W.N. Gardner

Department of Thoracic Medicine, Kings College, School of Medicine and Dentistry, Denmark Hill, London, England

Previous studies (Gardner, 1983; Gardner and Meah, 1985) showed that laryngectomised subjects at quiet rest, in comparison with a matched control group, had more rapid expiratory flow with a slightly prolonged expiratory time resulting in extended end-expiratory pauses. In some subjects these amounted to frank apnea despite full wakefulness, and expiratory time was often excessively variable. These subjects breathed through chronic tracheostomies with significant loss of upper airway resistance and the more rapid expiratory flow was consistent with a passive response to this loss.

There appears to be no literature on breathing during sleep in these subjects. In normal subjects, breathing becomes irregular in REM sleep and it is not inconceivable that removal of the larynx and diversion of airflow from the upper airways with their range of controlling reflexes might have more profound effects during sleep than in the waking state, leading possibly to prolonged apnea, arterial oxygen desaturation and increase in arterial PC02.

In these experiments we have studied the behavior of the major respiratory parameters during natural sleep in laryngectomised subjects.

METHODS

We studied 4 male subjects (ages 66, 65, 56, 45) who had undergone laryngectomy for cancer of the larynx at least 6 months previously. All had been smokers but at the time of study were fit with normal chest radiographs and no evidence of recurrence of the tumour. All had prolonged end-expiratory pauses in the awake state at quiet rest with means over approximately 50 breaths of between 1 and 2.5 seconds.

Breathing was measured throughout the night with a respiratory inductive plethysmograph (Respitrace) displayed on a chart recorder and analysed manually. It was calibrated at the beginning and end of the night using the least squares regression method against a pneumotachograph temporarily attached over the patients stoma. End-tidal C02 <PETC02) was measured by Gould capnograph or Centronics mass

73

Table

~SPIRA TORY END-TIDAL OXYGEN F~QUENCY PC02 SATURATION (%)

Waking 15.4 ± 2.0 35.1 ± 1.0 95.4 ± 0.8 (14.0 - 18.3) (34.0 - 36.4) (94.6 - 96.4)

Non-~M 15.3 ± 1.8 35.5 ± 1.7 94.5 ± 0.6 (13.0 - 17.1) (34.0 - 36.4) (93.8 - 95.3)

~M 17.3 ± 2.2 37.7 + 4.7 94.3 ± 0.6 (15.3 20.0) (33.6 - 44.5) (93.8 - 95.2)

Mean ± SD (minimum - maximum)

spectrometer sampling from a light cardboard ring attached to the outside of the stoma. Internal connections in the trachea were avoided. Oxygen saturation was measured with an Ohmeda Biox 3700 pulse oximeter.

Sleep was assessed by measurement of EEG and eye movements. Standard EEG electrodes were attached in the C3-A2 or C4-A1 positions. Two eye electrodes were attached at an angle across the orbits to monitor EOG. All electrodes were connected to an HB2 head box linked to an SLE Neuroscribe pen recorder. Sleep was staged into waking, non-~M and ~M according to the criteria of Rechschaffan and Kales.

Signals were divided between 2 chart recorders, the speed of which were increased every 15 minutes to allow more detailed analysis. Analysis was perfomed by hand over a 2 minute period every 15 minutes throughout the night.

Patients arrived in the study room an hour before their normal bedtime and sat quietly watching TV. They were connected to the apparatus and then slept in a normal bed, waking at their usual time. They were not sleep deprived and slept normally, reporting a good nights sleep. They had previously been studied in the laboratory and were familiar with the equipment. In two cases, a familiarisation study the previous night was performed.

RESULTS

Subjects slept for 4-8 hours, consistent with their normal sleep length. The proportions of ~M to non-~M were consistent with normal subjects and they showed 4-6 ~M cycles per night.

Changes of oxygen saturation, PETC02, respiratory frequency, and when measured, tidal volume, were consistent with those reported in the literature for the various sleep stages in intact subjects (see table). Saturation did not fall below 94, tidal volume fell slightly from waking to non-REM and to ~M, respiratory frequency increased slightly in non~M and ~M, and PETC02 rose about 3 mmHg between waking and ~M. There was no marked apnea in any sleep stage, although the Respitrace did not allow the more sophisticated computer analysis of expiratory flow

74

previously reported from this laboratory. It is concluded that removal of the larynx and exclusion of the upper airways from the respiratory circuit have no major effects on respiratory control mechanisms during any stage of sleep and do not affect the quality or duration of sleep.

REFERENCES

Gardner, W.N., 1983, Role of the larynx in control of pattern of breathing during C02 inhalation in human,~~ Physiol., 54: 1726.

Gardner, W.N., and Meach, M.S., 1985, Clin. Sci., 69: Suppl. 12, 8P.

75

VAGAL REACTIVITY DURING SLEEP: STv~Y OF THE OCULOCARDIAC

AND THE HERING - BREUER REFLEX IN PRETERMS

INTRODUCTION

J. Ramet1, J.P. Praud2, A.M. D'Allest2, A. Carofilis2, M. Dehan2 , and Cl. Gaultier2

1Department of Pediatrics AZ-VUB-Vrije Universiteit Brussel, Belgium

2Department of Physiology, Hospital A. Beclere, Clamart France

Episodes of spontaneous or stimulation-related bradycardia are often observed in preterm infants. Vagal cardiac and respiratory reflexes are of major importance in this age group. The Hering-Breuer inspiratory inhibitory reflex is ¥nown to decrease with maturation. Another vagal reflex, the oculocardiac reflex, was performed during active sleep in 17 healthy preterms. The purpose of this study was to evaluate the maturation of the oculocardiac reflex in preterms and to assess the relationship betwen this cardiac vagal reflex and the Hering-Breuer inspiratory-inhibitory reflex.

METHODS

Seventeen preterms who where free from neurologic or cardiopulmonary disease were studie (Table 1). The Vagal Stimulation Test (VST) was performed during natural sleep with the infants in supine position. A standardized technique of ocular compression was used: identical pressures were applied to both eyes using water-filled pressure-sensors. The pressure on both eyes was maintained without fluctuation for a period of ten seconds (Fig. 1).

The Hering-Breuer inspiratory inhibitory reflex (HBR) was studied by the end-expiratory occlusion technique using a slide valve connected to a pneumotachograph and a face mask. The HBR response was expressed as the percentage of prolongation of inspiration from the control inspiratory time (TI) to the occluded TI· (Fig. 2).

VST and HBR were performed during active sleep as determined by EEG analysis and clinical behavioral observation.

77

Table 1. Characteristics and Data Preceding and During Vagal Stimulation Test

---------~-------------------------------------------

SUBJ. SEX GNA PNA PCA CONTROL* ASYSTOLE* '.tRR No (WK) (d) (WK) (msec) (msec)

1 M 30.5 6 31.5 400 1733 433.3 2 F 31 8 32 440 1867 424.3 3 M 31.5 5 32 435 1800 413.8 4 F 26.5 46 33 400 1033 258.3 5 F 31.5 11 33 440 1666 378.6 6 M 32 19 34.5 435 1866 428.0 7 F 34 7 35 445 1233 277.1 8 F 35 10 36.5 390 933 239.2 9 M 35 15 37 440 1066 242.3

10 M 32 36 37 440 1071 242.4 11 F 36.5 6 37.5 435 666 153.1 12 M 30 52 37.5 405 869 214.6 13 M 35.5 17 38 440 845 192.0 14 M 37 10 38.5 435 1200 275.9 15 F 35.5 20 38.5 410 833 203.2 16 F 29.5 70 39.5 435 733 168.5 17 M 29 73 39.5 395 666 168.6

-----------------------------------------------------

MEAN 32.5 24 36.0 425 1181 277.3 SD 3.0 23 2.8 19 436 99.2

-----------------------------------------------------------

* control RR interval: mean value preceding the vagal stimulation test. asystole: duration of the longest asystole during the vagal stimulation

test. '.t RR: maximal prolongation of RR interval during VST expressed as

percentage of the preceding control RR interval GNA: gestational age at birth PNA: postnatal age at recording PCA: postconceptional age at recording

78

uo

uo

ECO

pre1aur• ___ - - ---------' lett

t t Fig. 1. Vagal stimulation test by ocular compression, experimental

tracing in an infant; top to bottom: EEG-EEG-ECG- ocular compression left-ocular compression right. The arrows indicate onset and end of the ocular compression. The brace indicates the control period in which the RR interval was calculated and the asterisk - the longest asystole or maximal pr?longation of RR interval during the test.

PLOIW

VOLUME rnl

. o

[

./ \ / '-~

I~ Tlo

Fig. 2. Hering-Breuer reflex; Trc: inspiratory time of control breath, Ttotc: respiratory cycle time of control breath Tr 0 : inspiratory time of occluded effort (%BR = Tr 0 - Trc/Tr0 o 100)

79

a: a: ~

80

400

300

200

• • • •

•

\ ',

•

', '\ • ~ •

·~ \ ·"' ~ • \

100

30 32 34 36 38

POSTCONCEPTIONAL AGE (weeks)

Fig. 3. Postconceptional age (PCA) is plotted against %RR, i.e. maximal prolongation of RR interval expressed as % of the control RR interval.

40

RESULTS

400

800

I ~

200

100

n=17

r=O·-

P(0.01

• • •

• ••

10 20

%HII

•

ao

• • •

40 150

Fig. 4. %HB plotted against %RR, i.e. response of the Hering-Breuer reflex vs. maximal prolongation of RR interval expressed as %of the control RR interval.

Values of asystole and% RR are indicated in table 1. We examined the relationship between duration of the longest asystole and %RR against gestational and postconceptional ages: no relationship was found between the duration of the longest asystole or %RR and gestational age. Postconceptional age was significantly correlated to %RR (r = -0.87 p < 0.001) (Fig. 3). A significant linear relationship between the response of the two vagally-mediated reflexes (VST and HBR) was observed (r = 0.88, p < 0.001) (Fig. 4).

CONCLUSION

The strength of the response to cardiac vagal stimulation by ocular compression during active sleep decreases with postconceptional age in preterms. The evolution of two vagally-mediated reflexes is significantly correlated. The results of an oculocardiac reflex studies by a standardized technique in preterms should be compared to the present predicted values.

81

RESUMPTION OF VENTILATION AT THE END OF OBSTRUCTIVE SLEEP

APNEAS IS NOT DETERMINED BY DIAPHRAGMATIC FATIGUE

INTRODUCTION

S. Sanci, F. Cibella, 0. Marrone, G. Cuttitta, R. Modica, S. Romano, and V. Bellia

Istituto di Fisiopatologia Respiratoria del C.N.R., Palermo Istituto di Pneumologia, Universita di Palermo, Italy

In patients affected by obstructive sleep apnea syndrome (OSAS) a progressive increase of the force developed by the respiratory muscles is usually observed during the occlusive phase before upper airway patency is resumed (Onal and Lopata, 1986).

The inspiratory pressures generated by the action of respiratory muscles increase toward the end of the apnea, reaching values potentially close to those known to determine an impairment of muscular contraction.

However, it has not yet been established whether in these conditions diaphragmatic fatigue occurs. In fact, Guilleminault (1980) first described a significant decrease in the centroid frequency of the spectrum of diaphragmatic electromyogram (EMGdi), compatible with the onset of diaphragmatic fatigue, in a patient with OSAS during an obstructive apnea occuring in non REM (NREM) sleep; conversely, Martin et al. (1982) did not report any modification suggesting diaphragmatic fatigue, in the same spectrum of the EMGdi analyzed in terms of high/low frequency ratio. Moreover, Vincken et al. (1987) speculated from their results that in OSAS, when the respiratory muscles develop a certain strength, close to the fatigue threshold, arousal occurs in order to prevent the onset of diaphragmatic fatigue.

Our study was designed to contribute to the solution of this controversy by evaluating the diaphragmatic contraction during obstructive apneas in NREM sleep both from the spectral and the mechanical point of view.

METHODS

Four men and one woman with OSAS were studied during nocturnal sleep. Sleep state was identified by electroencephalogram (EEG), electro-oculogram (EOG), and submental EMG by the conventional standard criteria.

EMGdi was performed via an esophageal electrode and the frequency power spectrum of the signal was measured in order to detect the development of diaphragmatic fatigue from the changes in the power

83

135 t f 120

f 105

t t 1 * f

90 f f f t ~ 75 f f f f

...J

:i 60

45

30

IS

-3 -2 -I 0 2 3 4 5 6 7 8 9 10 II 12 13 + BREATHS

Fig. 1. Time course of the H/L ratio variation during obstructive apneas in the investigated sample. On the abscissa, number of breaths from the third preapneic (-3) to the third postapneic (13). On the ordinate, H/1 ratio expressed as percentage of the value at the third preapneic breath. Data are expressed as mean ± SE.

contained in the high (150-360 Hz) and low (20-50 Hz) frequency bands (H/L) (Scheitzer et al., 1979).

Transdiaphragmatic pressure (Pdi) was measured by the conventional balloon catheter technique. Each patient performed, during wakefulness, maximal voluntary inspiratory efforts in the supine position in order to obtain the maximal transdiaphragmatic pressure (Pdi max). From the Pdi tracing the tension-time index of the diaphragm (TTdi), i.e. the product of PdijPdimax and Ti/Ttot (Bellemare and Grassino, 1982a), was calculated. A TTdi value > 0.18 was assumed as a potentially fatiguing threshold (Bellemare and Grassino, 1982b).

Airflow (V) was also monitored by a Fleisch No. 2 pneumotachograph. All the signals were recorded on a strip-chart recorder (7758B Hewlett Packard) and on magnetic tape (3968A Hewlett Packard) for later playback and analysis.

A breath by breath analysis from the third preapneic to· the third postapneic breath was performed in total of 30 obstructive apneas for the five patients.

RESULTS

Fig. 1. illustrates the time course of the HfL ratio values obtained for the five patients during the preapneic, apneic and postapneic

84

-o II-

.35

.30 * * * .25 * * * * * * * .20 *

* * * ------------:-------------.--------------.----------------------------------------

.15 * * * * * * . IS I

* * * .1215

e.ee 2 3 5

SUBJECTS

Fig. 2. TTdi values recorded in the five patients. Each point refers to the last occluded breath of an apnea. The broken line refers to the selected TTdi thresold (0.18).

breaths. The H/L values tended to decrease at the beginning of the apnea, then progressively increased up to the baseline preapneic value toward the end of the apnea.

TTdi values higher than 0.18 were observed in 15 of the 30 analyzed apneic episodes; in eight apneas this occured only in the last occluded effort, whereas in the remaining seven in the last two to six.

Fig. 2. shows the TTdi values obtained for the five patients during the last occluded effort before airflow was resumed: the threshold value was reached in all apneas in one patient (#4) whereas for the remaining four a wariable behavior between apneic episodes was observed.

DISCUSSION

The results of the present study suggest that: a) diaphragmatic fatigue (as expressed by a significant decrease of the

H/L ratio in frequency spectrum of EMGdi) does occur in patients with OSAS during NREM sleep;

b) the arousal reaction is not strictly related to a potentially fatiguing contraction (i.e. showing a TTdi higher than 0.18).

As concerns the first point, our results are in agreement with those of Martinet al. (1982) who found a significant reduction of the diaphragmatic H/L ratio from the third preapneic to the first occluded breath in six patients with OSAS. We found a similar reduction in our patients, with the lowest H(L values during the first or second occluded breu+h. Modifications of the diaphragmatic power spectrum with an increase of the low frequencies and a consequent decrease in the H/L

85

ratio are usually expected when diaphragmatic fatigue occurs (Roussos and Macklem, 1986); these changes have been related to a reduction in neuromuscular conduction velocity that is due to an imbalance between diaphragmatic energy demands and energy supplies. In our patients the lowest values in arterial oxygen saturation, i.e. the metabolic conditions most likely to predispose to fatigue, were observed only at the end of the apneic episode; conversely, changes in H/L values were observed at the beginning of the apnea. This consideration does not lend support to the view that these EMGdi changes are to be interpreted in terms of diaphragmatic fatigue. In addition it must be noticed that the time course of these modifications was not coincident with the appearance of the mechanical evidence of potentially fatiguing contractions represented by high TTdi values.

An alternative explanation is that the reduction of the H/L ratio might be related to an alteration of the central nervous system output to the phrenic nerve that may characterize the onset of the apnea in OSAS (Martinet al., 1982).

As far as the mechanical aspects of diaphragmatic contraction are concerned, in a recent paper Vincken et al. (1987) found a direct relationship between the arousal reaction and TTdi values reached during the last occluded effort. They found that when the TTdi was close·to the fatigue threshold (0.18), arousal occurred in all their patients, suggesting then, that the inspiratory muscles triggered a reflex when reaching a certain tension - in order to avoid the onset of fatigue.

We found a wide variability among our patients between the TTdi values developed during the last occluded efforts preceding the arousal reaction (Fig. 2). For each patient the TTdi reached at the end of the occlusive phase differed from apnea to apnea and only in one patient it was always higher than 0.18. However, in this particular case we found TTdi values over the threshold starting as early as from the sixth breath before the end of the apnea, thus reducing the likelihood of a causative relationship between the condition of contraction and the onset of arousal.

Therefore, our results suggest that there is a wide range of stimuli probably cooperating to a various extent in producing arousal and resumption of upper airway patency in different patients and in different apneas.

REFERENCES

Bellemare, F., and Grassino, A., 1982a, Effect of pressure and timing of concentration on human diaphragm fatigue,~~ Physiol.: Respirat. Envirom. Exercise Physiol., 53: 1190.

Bellemare, F., and Grassino, A., 1982b, Evaluation of human diaphragm fatigue,~~ Physiol.: Respirat. Envirom. Execise Physiol., 53: 1196.

Guilleminault, C., 1980, Sleep apnea syndromes: impact of sleep and sleep states, Sleep, 3: 227.

Martin, R.L., Fernandez, E., Hudgel, D.W., and Hill, P., 1982, The association of sleep apnea to the High/Low ratio of the diaphragmatic EMG, Am. Rev. Respir. Dis., 125: 107A.

Onal, E., and Lopata, M., 1986, Respiratory muscle interaction during NREM sleep in patients with occlusive apnea,~~ Physiol.: Respirat. Envirom. Exercise Physiol., 61: 1891.

86

Roussos, C., and Macklem, P.T., 1986, Inspiratory muscle fatigue, in.: "Handbook of Physiology", vol.III, Part 2, Fishman A.P., Macklem P.T., Mead J., Geiger S.R., eds., American Physiological Society, Washington, D.C.

Schweitzer, T.W., Fitzgerald, J.W., Bowden, J.A., Lynne-Davies, P., 1979, Spectral analysis of human inspiratory diaphragmatic electromyograms, J.Appl. Physiol.: Respirat. Envirom. Exercise Physiol., 46: 152.

Vincken, W., Guilleminault, C., Silvestri, L., Cosio M., Grassino A., 1987, Inspiratory muscle activity as a trigger causing the airways to open in obstructive sleep apnea, Am. Rev. Respir. Dis., 135: 372.

87

THE EFFECT OF CHRONIC PULMONARY DENERVATION ON THE PATTERN

OF BREATHING DURING SLEEP IN MAN

INTRODUCTION

S.A. Shea, R.L. Horner, E. McKenzie, N.R. Banner, M.H. Yacoub and A. Guz

Department of Medicine, Charing Cross and Westminster Medical School, London Cardiothoracic Unit, Barefield Hospital, Middlesex, UK

In most mammals vagal afferents from the lungs play an important role in the control of breathing - without them breathing becomes significantly slower and deeper (eg. Hering and Breuer, 1868) and more variable (Kelsen et al., 1982). However, in healthy man at rest most evidence suggests that the effect of vagal afferents upon the pattern of breathing is negligible (Guz et al., 1964; Cross et al., 1976; Winning et al., 1985). These studies have used different techniques to achieve acute reversible vagal block and they have not been completely selective for the pulmonary innervation.

The operation of combined heart-lung transplantation offers the opportunity to study the pattern of breathing in the chronic absence of vagal afferents from the lungs: tracheal innervation remains intact. In the present study we have quantified the level, the pattern and the variability of breathing in recipients of human heart-lung transplantation during relaxed wakefulness and during sleep (when the behavioral influences upon breathing are minimal). These data have been compared with similar data obtained from normal controls, and also with a second group who had undergone heart transplantation alone, to control for the effects of intrathoracic surgery and drug treatment.

SUBJECTS

The study was performed on healthy recipients of human heart-lung transplantation (HL); 3 males, 5 females, aged between 22-44 years. The 'time post transplantation ranged 1-24 months. These patients were matched as closely as possible to 8 healthy patients who had undergone heart transplantation (H); 4 males, 4 females, aged 17-51 years, 1-17 months posttransplantation, and also to the second control group of 8 healthy normals (N); 3 males, 5 females, aged 19-43 years.

89

HL ······· H N ---

RESTING WAKEFULNESS

Vr(ml) 500

2 1 0 1 2 Tt (sec) TE (sec)

STAGE 4 SLEEP

Vr(ml) 500

2 1 0

Tt (sec)

STAGE 2 SLEEP

Vr (ml) 500

3 2 1 0 Tt (sec)

REM SLEEP

Vr(ml) 500

3 2 1 0 Tt (sec)

3

3

Fig. 1. HL, heart-lung transplant patients; H, heart transplant patients; N normal controls: 8 subjects per group, 5 min per subject per state. Tr, inspiratory time ; TE, expiratory time; VT, tidal volume. Fisher's Least Significant Difference (LSD), derived from analysis of variance, is portrayed graphically. All lines lie within these regions, hence there are no s ignificant differences between groups during any state .

90

MEASUREMENTS

Breathing was quantified during wakefulness and sleep with respiratory inductance plethysmography (Respitrace) which was calibrated against a wedge spirometer using the multiple linear regression technique. Blood gases were estimated noninvasively; end-tidal PC02 (PETC02) was measured from expired air sampled at the nose, transcutaneous PC02 (PTcC02) was measured during sleep using an electrode attached to the medial surface of the forearm, and arterial oxygen saturation (Sa02) was estimated with a pulse ear oximeter. The state of wakefulness or sleep was established by monitoring the EEG (C3/A2), EOG (F7/A1 and F8/A1) and the submental EMG.

PROTOCOL AND ANALYSIS

Each subject was monitored on two consecutive nights but only data from the 2nd night were analysed. Five minute periods from each of 4 states (Relaxed wakefulness, Stage 2 sleep, Stage 4 sleep and REM sleep) were selected for breath-by-breath analysis. For the wakefulness studies the subjects were supine and under standardised resting conditions with a minimum of sensory stimulation (Shea et al., 1987). During sleep sections were selected which were close in time to a Respitrace calibration and when the subjects were supine without snoring. The mean % change in Respitrace volume calibration overnight was 0.3% ± 7.6% (n=24).

In order to compare mean levels and the breath-by-breath variability (coefficient of variation) between the 3 groups (HL, H, and N) analyses of variance were employed upon each of the variables during the four states; Fisher's least significant differences (p=0.05) were used to test for significant differences between groups. The shape of the frequency distributions (coefficients of skewness and kurtosis) of TI, TE and VT were compared with Mann Whitney U tests.

RESULTS

The results of the analyses of variance upon the mean levels of the respiratory parameters are shown in Fig.1.: there were no significant differences between groups in any of the variables during wakefulness or sleep. Also, there were no significant differences in blood gases (PErC02, PTcC02, Sa02) between the three groups throughout wakefulness and sleep.

The results of analyses of variance upon breath-by-breath variability are shown in Fig. 2.: out of 36 comparisons between groups there were only 3 significant differences (p<0.05). Coefficients of skewness and kurtosis were used as indices of the shapes of the frequency histograms. The results of the analyses are shown in Table 1. Out of 72 comparisons between groups (Mann Whitney U tests) there were 3 significant differences (0.02<p<0.05). It is likely that this few number of statistically significant differences will have emerged by chance (Type I error) and we believe that they do not signify real physiological differences.

91

CD

1\.)

Tl

c.v.

(%

)

RE

ST

ING

W

AK

EF

ULN

ES

S

50

STA

GE

2 S

LEE

P

STA

GE

4 S

LEE

P

RE

M S

LEE

P

50~ 5

0~

50~

30

30

30

----

----

----

---

·---

--- -

------

------

------

--····•

·········

10

:mt _

___ 10

,,,

,,;,,,

,,,

10

~I:;m:

: -----

HL

H

N

HL

H

N

HL

H

N

TE

c.

v.

(%)

50

50 ~

-50

:1-~::~-,

:~ :1

ti!f:-~·

00 1r:~r-

-1,

: ltl' -

--H

L H

N

50

50~

:b_

: ;;;{

i:=::-

HL

H

N

HL

H

N

~~. :~

------------

--(%

) 10

:mr

-----

::::::::

::::::

HL

H

N

Fig

. 2

. F

ish

er'

s le

ast

si

gn

ific

an

t d

iffe

ren

ces

(p=

0.0

5)

are

draw

n ab

ove

and

belo

w t

he a

ver

age

co

eff

icie

nt

of

vari

ati

on

fo

r th

e n

orm

al

con

tro

l g

rou

p.

The

si

gn

ific

an

t d

iffe

ren

ces

from

no

rmal

ar

e in

dic

ate

d b

y ast

eri

sk.

Ab

bre

via

tio

ns

as

in F

ig.l

.

co

w

Tab

le 1

. C

ompa

riso

ns

of

the

shap

e o

f fr

equ

ency

his

tog

ram

s be

twee

n gr

oups

fo

r se

lecte

d v

ari

ab

les.

Tr

TE

vT

m,

H

N

m,

H

N

Ill.

H

N

Ske

wne

ss

0.3

7

-0.1

0

-0.2

4

1.5

3

1.0

8

1.9

8

1.1

2

0.0

0

0.2

4

AWAK

E K

urt

osi

s 2

.46

0

.99

0

.43

4

.56

2

.73

7

.17

8

.02

1

.37

0

.91

Ske

wne

ss

-0.5

1+

-0

.15

0

.02

0

.24

0

.73

0

.45

0

.19

-0

.12

0

.19

ST

AG

E 2

SLEE

P K

urt

osi

s 0

.68

0

.41

0

.09

1

.83

2

.04

1

.10

1

.41

1

.31

0

.53

Ske

wne

ss

-0.3

4

0.0

6

-0.0

1

-0.0

6

0.0

2

-0.0

6

0.1

9

-0.0

8

0.2

5

STA

GE

4 SL

EEP

Ku

rto

sis

0.6

4

2.7

4.

1.6

4

1.1

0

1.1

3

1.1

3

4.2

0

11

. 1o+

1

.25

Ske

wne

ss

-0.5

3

-0.8

7

-0.5

5

2.5

3

2.5

4

2.0

8

0.3

8

-0.1

1

0.1

3

REM

SL

EEP

Ku

rto

sis

0.2

4

1.5

2

0.4

4

8.9

6

13

.08

6

.40

-0

.11

0

.50

0

.02

• -

p<

0.0

5,

Ill.

com

pare

d to

H;

+ -

p<

0.0

5,

N c

ompa

red

to I

ll.

or

H.

Ab

bre

via

tio

ns

as

in F

igu

re 1

.

CONCLUSION

In these patients with heart-lung or heart transplantation there were no abnormalities in blood gases during relaxed wakefulness or throughout sleep. We have found that breathing in man with chronic pulmonary denervation is remarkably normal in level, pattern and variability during relaxed wakefulness and during all sleep stages.

REFERENCES

Cross, B.A., Guz, A., Jain, S.K., Archer, S., Stevens, J., and Reynolds F., 1976, The effect of anaesthesia of the airway in dog and man: a study of respiratory reflexes, sensations and lung mechanics, Clin. Sci. Mol. Biol., 50: 439.

Guz, A., Noble, M.I.M, Trenchard, D., Cochrane, H.L., and Makey, A.R., 1964, Studies on the vagus nerves in man: their role in respiratory and circulatory control, Clin. Sci., 27: 293.

Hering, E., and Breuer, J., 1970, Self-steering of respiration through the nervus vagus, (First published 1868), in: "Breathing; Hering-Breuer Centenary Symposium", Porter R.J., and Churchill A., eds. , London.

Kelsen, S.G., Shustack, A., and Hough, W., 1982, The effect of vagal blockade on the variability of ventilation in the awake dog, Resp. Physiol., 49: 339.

Shea, S.A., Walter, J., Murphy, K., and Guz, A., 1987, Evidence for the individuality of breathing patterns in resting healthy man, Resp. Physiol., 68: 331.

Winning, A.J., Hamilton, R.D., Shea, S.A., Knott, C., and Guz, A.,

94

1985, The effect of airway anaesthesia on the control of breathing and the sensation of breathlessness in man, Clin. Sci., 68:215.

CONCHOPLASTY IN THE TREATMENT OF THE OBSTRUCTIVE SLEEP APNEA

SYNDROME

H. Skarzynski, W. Jeglinski, A. Kukwa, G. Opalski, M. Ryba, Z. Szlenk, P. Radzimowski, and M. Lisicka

Departments of Otolaryngology and Cardiology Medical Academy, Warsaw, Poland

Upper airway compromise on the level of the nose may be the result of developmental deffects, chronic infections, or trauma. In some cases the patency of the nasal airway may be improved by means of a non-invasive treatment. In this paper we describe the possibilities of surgical treatment in cases where nasal airway compromise is caused by deviated nasal septum with unilateral compensative hypertrophy of the nasal concha or by bilateral hypertrophy of nasal inferior conchae (Kukwa et al., 1988 in press).

For more than ten years narrowed nasal airway is thought to be one of the causes of the disturbances of breathing during sleep (Guilleminault et al. 1976; Lugaresi et al., 1980; Miller, 1982). It is obvious that treatment of such a narrow airway is of great importance, though the optimal method is still being discussed and has not yet been established. Such a method should not only give permanent relief to the symptoms of obstructed nose, but should also preserve its physiological functions, especially for the humidification of inspired air.

On the basis of the experience of many authors and our own, we worked out a special method of the operation, wich safely decreases the size of nasal conchae without any side effects.

METHODS

Forty-eight patients with nasal airway compromise were treated. In 22 cases obstruction was caused by a bilateral hypertrophy of the nasal inferior conchae. In 26 patients the reason of the obstruction was the deviated nasal septum with contralateral compensative hypertrophy of the nasal concha inferior. In all cases subjective symptoms of sleep apnea syndrome were present. In all cases we performed polysomnographic studies before and after surgery.

The surgical treatment was dependent on the type of obstruction, and involved either septoplasty, or septoplasty complemented by conchoplasty.

95

RESULTS

In ~he surgical treatment of the nasal airway compromise two points are essential: patency should be regained and the physiological functions of the nose preserved. In all cases with hypertrophied conchae we performed a new type of operation, which was worked out in our clinic.

In all cases PSG revealed apneic episodes before surgery, though only in 31 of them (64.5%) sleep apnea syndrome was diagnosed. After the surgery none of the patients complained of the nasal obstruction and none complained of the crusting or dry nose. In PSG studies performed six months after the operation no apneic episodes were observed in 28 patients (58.3%). In the remaining patients PSG studies revealed the presence of very short apneas, but their quantity was significantly lower than before the treatment.

CONCLUSIONS

Proper elimination of the obstruction in the nasal airway, preserving physiological functions of the nose, is essential in the treatment of obstructive sleep apnea syndrome. The applied method of treatment of nasal airway compromise eliminated the obstruction without disturbing physiological functions of the nose (Mabry, 1981; 1982; Martinez et al., 1983). The relief of the symptoms was permanent and there were no complaints of dryness or crusting in the nose.

REFERENCES


Kukwa, A., Skarzyn~ki, H., Szlenk, Z., and Ryba, M., 1988, Surgical treatment of nasal obstruction. Turbinoplasty, Otolaryngol. Pol., (in press).

Lugaresi, E., Cirignotta, F., Coccagna, G., and Piana, C., 1980, Some epidemiological data on snoring and cardiocirculatory disturbances, Sleep, 3: 221.

Mabry, R.L., 1981, Medical management of the stuffy nose, South. Med. ~. 74: 984.

Mabry, R.L., 1982, Inferior turbinoplasty, Laryngoscope, 92: 459. Martinez, S.A., Nissen, A.J., Stock, C.R., and Tesmer, T., 1983, Nasal

turbinate resection for relief of nasal obstruction, Laryngoscope, 93: 871.

Miller, W.P., 1982, Cardiac arrhythmias and conduction disturbances in the sleep apnea syndrome. Ann. ~ Med., 73: 317.

96

THE EFFECTS OF ADENO- AND TONSILLECTOMY IN CHILDREN WITH

SLEEP APNEA SYNDROME

H. Skarzynski, A. Kukwa, G. Opalski, R. Krauze, W. Jeglinski, K. Sfomka, and A. Kalotka-Bratek

Departments of Otolaryngology and Cardiology Medical Academy, Warsaw, Poland

Every case of disturbances of respiration during sleep with apneic episodes longer than 10 s should always draw our attention. After many investigations it became certain that especially children should be observed closely during all types of upper airway infections, when the patency of the airway is compromised and the risk of apneic episodes increases. Even a limited compromise of the upper airway, when chronic, may lead to a serious pathology in the circulatory system and may provoke the manifestation of disturbances in emotional life.

Simmons and Hill (1974) were first to draw the attention of the ENT surgeons to the chronic upper airway compromise. Most of the workers agree that among possible cases of such state are: nasal septum deviation (Simons et al., 1977), hypertrophy of the lymphoid tissue of the pharynx (Eliascher et al., 1980; Rubin et al., 1983), rhinitis allergica (Lavie et al., 1981). The symptoms of sleep apnea are observed in children with large adenoid and tonsills (Eliascher et al., 1980) and in some cases the effects of such obstruction in the upper airway on the circulatory system are quite serious (Miller, 1984).

The aim of our paper is to present the effects of adeno- and tonsillectomy in children with the diagnosis of sleep apnea. In 12 chidren with sleep apnea symptoms, surgical treatment was applied. Adenotomy or tonsillotomy eliminated an airway obstruction and improvement in quality of night sleep was observed. Operated children improved their school performance, demonstrated better daytime activity and ability to concentrate.

METHODS

The material consists of 12 cases (age 5 - 16) treated in the ENT Clinic for recurring infections of the upper airway and progressing upper airway compromise. In 8 cases upper airway compromise was caused by hypertrophied adenoid and in 4 cases by large tonsills. In all patients we observed somnolescence during the day, lower activity and loss of ability to concentrate at school. In two cases arterial blood pressure was slightly higher. The children snored and breathed through the mouth.

97

RESULTS

Every child underwent PSG investigation before and after the surgery. In 6 cases the PSG study exhibited disturbances of breathing. Before the surgery apneic episodes were 8 - 25 s long. The number of apneic episodes per hour ranged from 6 to 15. During the apneic episodes P02 and PC02 changed typically and the range of such changes were in close proportion to the time of the apnea. At the same time we observed disturbances of the heart rhythm. In all cases we did not observe such disturbances after the surgery.

There are five parameters of clinical improvement after surgical treatment: changes in ECG, heart rate, arterial P02 and PC02, and - first of all - respiratory pattern during sleep. At least there of them were observed in the total of 12 patients. In 8 cases all five parameters were found.

DISCUSSION

Early diagnosis of chronic upper airway compromise in essential. Narrowed upper airway results in many disturbances, both emotional and physiological, leading to the retardation of child's development. After the treatment the children presented better activity and much better school performance, which was observed by themselves and by their parents. In many papers (Eliascher et al., 1980; Guilleminault et al, 1976; Miller, 1984) it was outlined that upper airway compromise seriously disturbs the function of circulatory system function. Our patients did not yet present any such disturbances during physical examination·, but a certain pathology was already observed in the PSG studies prior to surgery.

The analysis of the results of our investigation suggests that indications for adeno- and tonsillectomy should be much widened. Such statement may serve as one more voice in the discussion concerning early adenotomy, tonsillectomy or tonsillotomy.

REFERENCES

Eliaschar, J., Lavie, P., Halpern, E., Gordon, C., and Altroy, G., 1980, Sleep apneic episodes as indications for adenotonsillectomy, Arch. otolaryngol., 106: 492. -


Lavie, P., Gertner, R., Zoner, J., and Podoshin, L., 1981, Breathing disorders in sleep associated with "microarousals" in patients with allergic rhinitis, Acta otolaryngol., 92: 529.

Miller, W.P., 1984, Cardiac arrhythmias and conduction disturbances in sleep apnea syndrome, Ann. J. Med., 73: 317.

Rubin, A-H.E., Eliaschar, J~oachim, Z., Alroy, G., and Lavie, P., 1983, Effects of nasal surgery and tonsillectomy on sleep apnea, Bull. Europ. Physiopath. Resp., 19: 612.

Simmons, F.B., and Hill, M.W., 1974, Hyposomnia caused by upper airway obstructions: A new syndrome in otolaryngology, Ann. Otol., 83: 670.

Simmons, F.B., Guilleminault, C., Dement, W.C., Tilkian, A., and Hill, M.W., 1977, Surgical management of airway obstructions during sleep, Laryngoscope, 87: 326.

98

SURGERY THERAPY FOR OBSTRUCTIVE SLEEP APNEA -

PRESENT AND FUTURE

Bernard de Berry-Borowiecki

University of California, Irvine, USA

The phenomenon of airway obstruction in sleep apnea syndrome is caused primarily by central nervous system failure to maintain integrity of pharyngeal airway during sleep. Specifically, normal activation of pharyngeal dilator muscles undergoes profound change from wakefulness to sleep. There is a loss of muscle tone within pharyngeal wall. The airway becomes narrower, necessitating greater inspiratory effort to maintain satisfactory airflow. This in turn exerts a further inward pressure within pharynx and consequently flaccid muscular walls of the pharynx collapse, causing the occlusion. A pharyngeal airway that anatomically has lumen narrower than normal is more likely to yield to the occlusive forces. Conversely airway defense by neuromuscular mechanism (compensating for unfavorable anatomy of the region) may in time lose its efficiency. Thus evolved hypertrophy and possible degenerative changes in the muscles of tongue, palate and pharyngeal walls may precipitate a point where neuromuscular function fails to maintain the airway under certain adverse conditions, e.g., sleep, anesthesia, etc. Historically, theory of a mechanism for pharyngeal airway occlusion in obstructive sleep apnea was embroiled in the controversy between the proponents of so-called active mechanism suggested by Weitzman et al. (1978) and passive mechanism proposed by Remmers et al. (1978) and others.

The surgeons readily embraced a theory of positive mechanism of obstructive sleep apnea event occlusion. It provided an attractive rationale for whole gamut of operations introduced in past years. It has been argued that once the anatomic configuration of the pharyngeal airway is corrected, neuromuscular function preserving the airway improves concurrently, leading to overall successful therapeutic result. We have been concentrating, therefore, on the progress in diagnosing anatomical abnormalities in obstructive sleep apnea patients' airway as the basis for design of corrective surgery. Upper airway endoscopy utilizes fiberoptic technology to exam methodically the entire extent of the upper airway under the conditions of wakefulness (de Berry-Borowiecki et al., 1985) and sleep (Borowiecki et al., 1978). The same methodology utilizing conditions of sleep lab allowed us to correlate the PSG characteristics with dynamically changing pharyngeal airway configuration (Rajewski et al., 1984). Radiographic studies of upper airway in obstructive sleep apnea patients evolved in a manner similar to the endoscopic evaluations. Awake subjects have been evaluated with lateral

99

head and neck radiograms (cephalometry) (de Berry-Borowiecki et al., 1987a in the press; Guilleminault et al., 1984), computerized axial tomography (Suratt et al., 1983) and MRI techniques, as well as contrast cinefluoroscopy of upper airway during sleep (Weitzman et al., 1978; Katsantonis and Walsh, 1984). The results of those studies suggests that the majority of OSA patients display disproportionate anatomy of the upper airway and consequently seem to have multiple sites of obstruction during apnea event (Rajewski et al., 1984).

Typical obstructive sleep apnea patient has compromised nasal airway, secondary to structural septal deformities and hypertrophic and inflammatory mucosal changes. Upper pharyngeal airway may be adversely affected by retropositioned maxilla, long hard palate and hypertrophic and elongated soft palate (de Berry-Borowiecki et al., 1987a in the press). In some individuals there is concomitant hypertrophy of lymphatic tissue of Waldeyer's ring (Mangat et al., 1977; Orr and Martin, 1981). Lower pharyngeal airway may be compromised by presence of micrognathic or retrognathic mandible and low position of hyoid bone (de Berry-Borowiecki et al., 1987a in the press; Riley et al., 1983). Consequently, the tongue base is posteriorly displaced with its large portion being located in hypopharynx. The fulcrum of the tongue mass anterior-superior movement is thus unfavorably mechanically displaced downward. In addition, the observations suggest hypertrophy of the tongue with enlargement of its mass volume and decrease of oral cavity volume. Cross section of the pharyngeal airway is decreased and tendency of its walls to collapse is increased as demonstrated by endoscopic examination during Mueller's manoeuvre. CT scanning performed at various levels of pharyngeal airway shows its decreased diameter. MRI and acoustic reflex studies confirm decreased volume of pharyngeal airway.

Surgical therapy for obstructive sleep apnea evolved from an airway bypass procedure - tracheostomy - to the surgical attempts of correcting pharyngeal and nasal airway. Since 1969 report by Kuhlo et al. of successful reversal of obstructive sleep apnea by tracheostomy, this procedure has been recognized as the single most effective surgical technique employed for therapy of obstructive sleep apnea (de BerryBorowiecki and Sassin, 1983). Properly functioning tracheostomy has 1001 cure rate and when it fails to reverse obstructive sleep apnea, one must assume that preoperative diagnosis is inaccurate. The pitfalls of permanent tracheostomy are multiple and include psychological and esthetic stigmata, functional limitation in normal everyday function, as well as threat of long-term iatrogenic damage to the trachea.

Palatopharyngoplasty was popularized by Fujita et al. (1981), Simmons et al. (1984) and Borowiecki et al. (1978). It offers a very satisfactory solution for chronic snoring, but its therapeutic value for a patient with OSA is limited. It is expected to be successful in about one-fourth of random sample of OSA patients. The cure rate, however, raises to approximately 70% when appropriate selection criteria are employed. PPP strives to reduce mucosal redundancy in the region of the velopharyngeal sphincter. It weakens the sphincteric mechanism by partially removing its muscular elements and scar forming sequelae. In instances where tonsillar hypertrophy is present, it aids to correct anatomical obstruction. This surgical procedure, however, does not change craniofacial structures that may influence configuration of the pharyngeal airway.

Reports of Kuo et al. (1979) and Bear and Priest (1980) stimulated development of surgical techniques to improve hypopharyngeal airway. Segmental mandibular advancement osteotomies in conjunction with infrahyoid release and hyoid bone suspension might improve hypopharyngeal

100

pharyngeal airway without change in occlusion (Riley et al., 1984; de Berry-Borowiecki et al., 1987b in the press) . Vertical sliding mandibular osteotomies achieve similar goal with simultaneous correction of occlusion and when employed in conjunction with anterior maxillary reposition (Riley et al., 1986), a potential for expansion of pharyngeal airway is considerably enhanced. The procedure carries additional benefit in effecting enlargement of the oral cavity volume, thus improving accommodation of the tongue mass in its anterior-superior movement during inspiration.

Neurophysiologic studies of the structures defining pharyngeal airway suggest the existence of two primary forces applied to the pharyngeal wall. One is an outbound vector effectively opening airway and the other one is directed inward resulting in closure of the pharyngeal airway. It is generally accepted that genioglossus muscle is the most important pharyngeal airway dilator (Sauerland and Harper, 1976; Sauerland et al., 1981; Brouillette and Thach, 1980) and its activity is dramatically decreased during obstructive sleep apnea event. The animal experiments (Brouillette and Thach, 1979) have shown that sectioning of both hypopharyngeal nerves decreases upper airway closing pressure. Conversely, one could hypothesize that stimulating XII nerve might result in the increase of upper airway closing pressure, decrease of the airflow resistance and stabilization of pharyngeal airway. We have examined this hypothesis on a rabbit animal model. The time for this communication does not allow for a detailed description of our experiments. In brief, we have measured upper airway pressure, respiratory effort, acoustic characteristic of respiration and 02 saturation under the condition of intravenous anesthesia. Respiratory effort has increased in relation to depth of anesthesia with concurrent rise of negative upper airway pressure. The observed increased upper airway resistance resulted in characteristic inspiratory pharyngeal stridor (which corresponds to snoring sound in a human). At the same time, arterial oxygen saturation decreased as compared to the starting baseline value. Stimulation of the XII nerve resulted in marked decrease of negative upper airway pressure and abolished inspiratory pharyngeal stridor. Correspondingly, respiratory effort has decreased slightly and 02 saturation values increased.

COMMENTARY

In spite of significant progress in surgical therapies for obstructive sleep apnea, there is an obvious need for further improvement. Surgery is only one of several treatment modalities for OSA. At present its success continues to be somewhat unpredictable and long-term results uncertain. Surgical therapy, therefore, should be very selective and it is indicated only after comprehensive evaluation of individual patients. Future surgical program designs should include corrective measures for abnormal upper airway geometry, as well as for improvement of deficient neuromuscular mechanism of upper airway control. It is probably reasonable to hypothesize that these two factors are interdependent in the natural evolution of OSA syndrome. Early recognition of abnormal upper airway geometry predisposing for obstructive sleep apnea might provide opportunity for early airway correction and prevention of OSA development. The multidisciplinary cooperation of clinicians and basic scientists is essential for success of such effort. An individual diagnosed to have obstructive sleep apnea needs comprehensive evaluation including an assessment of geometric airway profile and tests for neuromuscular mechanism function. Such evaluation will allow formulation of individual treatment program and provide improved prognostic yardstick as to the outcome of surgical therapy.

101

Our animal model experiments produce evidence to suggest that direct hypoglossal nerve stimulation improves the function of neuromuscular mechanism stabilizing pharyngeal airway. Intravenous pentobarbital anesthesia does depress activity of central nervous respiratory system. In the conditions of our experiment, this resulted in decreased activity of upper airway muscles, but apparently did not affect adversely the performance of thoracic pumping muscles. This difference in the responses of upper airway and thoracic muscles in our experimental model induces condition closely resembling the phenomenon of obstructive sleep apnea. Thus, apparent pacing of genioglossus activity achieved through XII nerve stimulation offers promise to prevent occurrence of obstructive sleep apnea events and snoring. We hope that continuation of our studies will not only lead to achieving advancement in surgery of obstructive sleep apnea and snoring, but will provide additional information concerning physiology of the upper airway.

REFERENCES

Borowiecki, B., Pollack, C.P., Weitzman, E.D., Rakoff, S., and Imperato, J., 1978, Fibroptic study of pharyngeal airway during sleep in patients with hypersomnia obstructive sleep-apnea syndrome, Laryngoscope, 88: 1310.

de Berry-Borowiecki, B., Kukwa, A.A., and Blanks, R.H.I., 1985, Indications for palatopharyngoplasty, Arch. of Otolaryngol., 111: 659.

de Berry-Borowiecki, B., et al., 1987a, Cephalometric analysis for diagnosis and treatment of obstructive sleep apnea (in the press).

de Berry-Borowiecki, B., et al., 1987b, Surgical therapy for obstructive sleep apnea, (in the press).

de Berry-Borowiecki, B., and Sassin, J.F., 1983, Surgical treatment of sleep apnea, Arch. of Otoleryngol., 109: 508.

Bear, S.E., and Priest, J.H., 1980, Sleep apnea syndrome: Correction with surgical advancement of the mandible, ~Oral Surg., 38: 543.

Brouillette, R.T., and Thach, B.T., 1979, A neuromuscular mechanism maintaining extrathoracic airway patency, ~ ~ Physiol., 46: 772.

Brouillette, R.T., and Thach, B.T., 1980, Control of genioglossus muscle inspiratory activity,~~ Physiol., 49: 801.

Fujita, S., Conway, W., Zorick F., and Roth, T., 1981, Surgical correction of anatomic abnormalities in obstructive sleep apnea syndrome: uvulopalatopharyngoplasty, otolaryngol. Head Neck Surg., 89: 923.

Guilleminault, C., Riley, R., and Powell, N., 1984, Obstructive sleep apnea and abnormal cephalometric measurements, Chest, 86: 793.

Katsantonis, G.P., and Walsh, J., 1984, Cinefluoroscopy, its role in selection of candidates for UPP, Abstract presented at the meeting of the American Academy of Otolaryngology, Las Vegas, October.

Kuhlo, W., Doll, E., and Franck, M.C., 1969, Erfolgreiche Behandlung eines Pickwick-Syndroms durch eine Dauertrachealkanule, Dtsch. Med. Wochenschr., 94: 1286.

Kuo, P.C., West, R.A., Bloomquist, D.C., and McNeil, R.W., 1979, The effect of mandibular osteotomy in three patients with hypersomnia sleep apnea, Oral. Surg., 43: 385.

Mangat, E., Orr, W.C., and Smith, R.O., 1977, Sleep apnea hypersomnolence and upper airway obstruction secondary to adenotonsillar enlargement, Arch. Otolaryngol., 103: 383.

Orr, W.C., and Martin, R.J., 1981, Obstructive sleep apnea associated with tonsillar hypertrophy in adults, Arch. Intern. Med., 141: 990.

102

Remmers, J.E., de Groot, W.J., and Sauerland, E.K., 1978, Pathogenesis of upper airway occlusion during sleep, ~ ~ Physiol., 44: 931.

Riley, R., Guilleminault, C., Herran, J., et al, 1983, Cephalometric analyses and flow volume loops in obstructive sleep apnea, Sleep, 6: 304.

Riley, R.W., Powell, N.B., et al., 1986, Maxillary mandibular and hyoid advancement - an alternative to tracheostomy in OSA syndrome, otolaryngol. Head Neck Surg., 94: 584S

Riley, R.W., Powell, N.B., and Guilleminault, C., 1984, Inferior sagittal osteostomy of the mandible with hyoid myotomy suspension: A new procedure for obstructive sleep apnea, Abstract presented at the meeting of the American Academy of Otolaryngology, Las Vegas.

Rajewski, T.E., Schuller, D.E., Clark, R.W., Schmidt, H.S., and Potts, R.E., 1984, Videoendoscopic determination of the mechanism of obstruction in obstructive sleep apnea, Otolaryngol. Head Neck Surg., 92: 127.

Sauerland, E.K., and Harper, R.M., 1976, The human tongue during sleep. Electromyographic activity of the genioglossus muscle, Exp. Neurol., 51: 160.

Sauerland, E.K., Orr, W.C., and Hairston, L.E., 1981, EMG patterns of oropharyngeal muscles during respiration in wakefulness and sleep, Electroencephalogr. Clin. Neurophysiol., 21: 307.

Simmons, F.B., Guilleminault, C., and Miles, L.E., 1984, The palatopharyngoplasty operation for snoring and sleep apnea: An interim report, Otolaryngol. Head Neck Surg., 92: 375.

Suratt, P.M., Dee, P., Atkinson, R.L., Armstrong, P., and Wilhoit, S.C., 1983, Fluoroscopic and computer tomographic features of pharyngeal obstruction in obstructive sleep apnea, Am. Rev. Respir. Dis., 127: 487.

Weitzman, E.D., Pollack, C.O., Borowiecki, B., et al., 1978, The hypersomnial-sleep apnea syndrome: site and mechanism of upper airway obstruction, in: "Sleep Apnea Sydromes", Guilleminault, C., and Dement, W.C., eds., Alan R. Liss, Inc., New York.

103

ANESTHESIA AND CENTRAL NERVOUS SYSTEM

ABSTRACT*

B.K. Siesjo

Laboratory for Experimental Brain Research University of Lund, Lund, Sweden

Although anesthetics in general affect both function and metabolism of the brain the effects are quite variable. Some anesthetics (e.g. barbiturates) lead to an orderly progression of symptoms, starting with an initial excitation (stage I) progressing to delirium, sometimes accompanied by hallucinations and a cataleptic behavior (stage II), and ending in the unconsciousness of surgical anesthesia (stage III). However, other anesthetics like nitrous oxide and ketamine never cause stage III anesthesia and still others (e.g. enflurane) increase excitability and elicit seizures.

Clearly, anesthetics have complex effects on brain metabolism. This is reflected in a variable effect on metabolic rate and blood flow, ranging from a 50% reduction in oxygen consumption (e.g. barbiturates) to an actual increase in metabolic rate associated with seizure activity.

It is now known that although some anesthetics in high concentration block electron transfer in mitochondria and, thereby, ATP production, the anesthetics effects are not caused by a defective energy production. The generally accepted view is that the primary effect of anesthetics is on neuronal function and that changes in metabolic rate are secondary, the nature of these changes being determined by the type of neurons affected, and their functional organization.

It seems likely that the primary effects of anesthetics involve a change in the cell memebrane conductance to ions. One mechanism whereby some anesthetics depress function is a reduction in Na+ conductance. Local anesthetics are presumed to act in this way. However, other results suggest that at least some anesthetics exert their functional effect by reducing membrane conductance to ca2+. If this occurs at presynaptic sites, the effect would be to reduce transmitter release, and thereby synaptic transmission (e.g. barbiturates). However, at least some anesthetics seem to have postsynaptic effects that mimick those of inhibitory transmitters, acting by increasing Cl- conductance. It has recently become clear that some drugs decrease excitatory drive by acting as postsynaptic glutamate receptor antagonists, more specifically as noncompetitive antagonists at NMDA-operated calcium channels. Interestingly

105

enough, the so-called dissociative anesthetics (ketamine and phencyclidine) belong to this group. Since the effects are exerted at receptors which are preferentially localized to certain regions in the brain, it is understandable that there are marked differences in action between various anesthetic compounds.

* The full MS has not been received.

106

RESPIRATION AND ANESTHESIA

Bogdan Kaminski

Department of Anesthesiology and Intensive Therapy Medical Academy, Warsaw, Poland

Mysteries of the control of breathing enjoy a big group of interested specialists. Besides neurophysiologists for whom it is the daily bread a small crowd is waiting behind the scenes to get the crumbs: pulmonologists, anesthesiologists, pediatricians, thoracic surgeons and certain part of GPs. The information reaching this second circle is - to say the least - delayed.

In one of popular Polish textbooks on anesthesiology (I will spare the authors quoting their names) the control of breathing is explained as follows: "The rhythmic respiratory movements are controlled by the cells of the respiratory center in the floor of the fourth ventricle (Legallois 1824, Flourens 1824-1858). There exist presumably three main parts of the respiratory center: 1) spinal center able to initiate and maintain the respiratory activity of irregular pattern, 2) apneustic center in the middle and lower part of the pons, which not inhibited induces the breathing spasm, or apneustic breathing, 3) pneumotaxic center in the upper one third of the pons, producing cyclic inhibition of the apneustic center".

It is to no comfort that the knowledge of anesthesiology to the disposal of an average physiologist is of a similar freshness and prec1s1on. Along with shortage of concise texts there are also other· reasons for this information barrier, as too elaborate, too technical language and notorious lack of common meetings.

At one of the recent interdisciplinary meetings in USA the physiologists presented their papers dealing mainly with nembutal while anesthesiologists talked about high frequency ventilation. In the lobby both parties sounded alarm that the subject of the presentations is not adequate.

Since I have the privilege and opportunity to talk to such an eminent audience of neurophysiologists I would like to tackle some myths shrouding anesthesiology and ventilation in its course.

NARCOSIS AS IT IS

Few words about terminology. The term "narcosis" coined over two hundred years ago was not bad. Its vehement change to "anesthesia" was

107

mainly aimed at separating the doctors from the random "narcotiseur" who were usually the least able surgeons or the most able orderlies.

Much more important, however, was the change in the methods of temporary and reversible obtunding many functions of the central nervous system. There is no more ether anesthesia, halothane anesthesia or nembutal anesthesia. It is now intricate combination of drugs belonging to nearly 10 different pharmacological groups. Their choice and dosage depends primarily on the experience of a given anesthesiologist, on his predilections, on the needs at the particular moment of operation and - last but not least - on the specific characteristics of the patient. For many decades it is going to be so. Even sophisticated computerized systems are so often overriden by manual control that they are mainly the symbol of status more than a real step toward so called objective anesthesia.

Among the important pharmacological groups finding everyday use in clinical anesthesia are:

1. Narcotics - mainly fentanyl with its biphasic depression of the ventilatory response, but also the robust survivor of the past -morphine. They gain ground thanks to new discoveries in the receptor physiology and revolutionized ways and patterns of application: peridural, oral, time-cycled etc.

2. Proper anesthetics with halothane in the lead. Enflurane and isoflurane failed to displace halothane as they are more expensive and less predictable. Nitrous oxide play an exclusive role in spite of many attempted coups d'etat. Here belongs also ketamine, one of the few anesthetics which exert stimulating effect on chemoreceptors.

The recent advances concerning opiate receptors and opioid compounds may soon result in creating the drugs precisely aimed at relieving pain without obtunding ventilation. Anesthesiologists will also look forward to receive drugs specifically diminshing the ventilatory drive. They could facilitate treatment of COPD patients where prolonged assisted ventilation is often marred by diminished ability to cough and to move around in the bed when standard depressants have to be applied.

3. Tranquilizers and ataractics, among them diazepam being number one. 4. Barbiturates, practically represented only by thiopentone and

methohexital, with nembutal reserved for the laboratory animals. 5. Dehydrobenzperidol playing specific role when it comes to

determining the effect of anesthesia on ventilation. It is known to increase hypoxic response of chemoreceptors.

6. Relaxants which often make our discussion on the ventilation in anesthesia superfluous.

7. Lignocain and similar compounds more and more often used as adjuvants in general anesthesia.

8. Naloxone occupying more of our attention than it is believed. The stress of anesthesia and surgery may result in endorphin release not normally present; under such circumstances naloxone may uncover a suppresion of ventilation caused by release of endogenous opiates. There are also claims that naloxone may find an application in shock states of various origin as well as in numerous intoxications other than produced by narcotics.

9. Atropin, neostygmin, dopamine, few other drugs which undoubtedly modify ventilation and have to be taken into consideration when discussing the control of breathing.

ROUTES OF ADMINISTRATION

Important changes in this respect have to be taken into account by physiologists. Next to traditional routes - inhalation and intravenous -two new ones developed not yet fully evaluated with few more to come.

108

Perispinal route of narcotics administration is a variation of subarachnoid and epidural application of local anesthetics - on one hand, on the other - it resulted from discoveries in the field of opiate receptors. Primarily used in the treatment of chronic pain this way of administration gains popularity in surgical anesthesia. Combined epidural narcotics with routine general anesthesia allows to cut down the dosage of all drugs concerned, effectively saves the patient from postoperative pain, speeds up recovery and the process of regaining homeostasis. Morphine seems to be proper drug since its lower lipid solubility presumably allows it to be retained in CSF for a longer period than it is the case with more-lipid-souble drugs as fentanyl.

New, sometimes quite unusual routes of administration are in the offing, like intrapleural application of local anesthetics to produce extensive intercostal block, intratracheal infusion used in emergency, buccal and sublingual mucosa application and recently - uninterrupted skin application of vasodilating and narcotic agents. Obviously the target receptors remain the same via circulation, but the dynamic pattern, distribution and degradation pathways can be substantially different.

COEXISTING FACTORS IN CLINICAL ANESTHESIA

The title of this heading is biased by the author's profession. It should rather read "anesthesia as the complicating factor in intraoperative disorders".

Clinical anesthesia is carried out in a certain entanglement of factors everyone of which can endanger the patient's life. In the order of their occurrence these are:

1. Anxiety and fear with all their physiological accompaniment. 2. Surgical intervention. It is worth indicating that fear occurs

more frequently than operations, since it is born also by anticipated anesthesia, by the preoperative visits and by cancelled operations.

3. Blood loss and its replacement. Filling the vascular bed brings about many dangers which tend to be overlooked.

4. Periodical hypoperfusion of organs or regions. It is produced centrally, as in hypotension, or locally - due to rotation of the peduncle, compression by inconsiderate assistant or overzealous nurse.

5. Shifts in the fluid and acid-base equilibrium - practically unavoidable for several days after every major surgery.

6. Hyperoxia and hypocapnia produced by so called safe anesthesiologist.

Obviously there exists a number of other factors which are not so frequent or rather which do not appear so frequently in press.

Taken superficially such a recitation could dishearten any physiogist who would rather return to the safe subject of intraperitoneal chloralose in perfectly healthy rabbit subjected to blunt tracheostomy. The reality is different. There are many good reasons making the subject of anesthesia and respiration worth intensive research.

1. The respiratory disturbances in the postoperative period are one of two - the other being induction to anesthesia - most dangerous factors. Suddenly decreased vigilance of the staff, activity of various systems trying to restore homeostasis (e.g. compensatory hemodilution), respiratory and circulatory support discontinued, additional doses of analgetics and narcotics applied - it all adds to the dangers of the immediate postoperative phase.

2. Percentage of aged and cardiologically incompetent patients among

109

those subjected to anesthesia and surgery increases every year. Routine management and routine dosage of drugs could prove lethal in such cases.

3. Purely cognitive reasons. The place of stretch receptors and chemoreceptors in modifying the respiratory pattern in anesthetized humans, hyperoxia as the source of free oxygen radicals in diseased lungs, delayed effects of mandatory artificial ventilation - these are only few of many questions that can be answered by combined efforts of ansthesiologists and physiologists. It has been obvious since long that the complicated array of intraoperative factors is not and cannot be reproduced in laboratory setting. A serious progress can be achieved when the obsolete drugs will be eliminated from the researcher's armamentarium and when the clinicians would use more physiological terms to describe the situation they create and supervise.

Apneic threshold could complement PaC02 determination, respiratory muscle electromyogram would be preferable to the train of four in the thumb and the ventilatory drive could be assessed from the inspiratory effort.

4. Spontaneous ventilation is only allowed in small percentage of general anesthesia. Is it so harmless as we think? Or perhaps the concept of mandatory minute ventilation, so aptly applied in intensive care, would supply the answer for prolonged surgical anesthesia?

5. Nearly one third of surgical interventions is made under local analgesia. The influence of local analgetics on the respiratory pattern and control deserves more attention than is given to it by respiratory physiologists.

6. Management of chronic pain is a fast growing field of anesthesiology. Some techniques described for use in the operating rooms are transferred to the pain clinics (e.g. plexus blocks), some others made their way from the pain clinic, like epidural narcotics. A long list of respiratory problems awaits elucidation in this respect.

Sometimes the clinical experience seems to contradict the sound experimental evidence. Respiratory depression by morphine provides a good example. To the horror of many physiologists and some oldfashioned GPs it is now a common practice to apply morphine intravenously in cases of incipient pulmonary edema. In spite of producing a real depression the gaseous exchange in the lugns improves due to:

1. decreased turbulence in the airways brought about by slowing the respiratory rhythm and diminishing anxiety; tubular character of the laminar flow allows much better ventilation of the lungs at decreased breathing volumes;

2. peripheral pooling of blood instantly improves the pulmonary compliance;

3. diminished sympathetic response results in lowering the metabolic demands.

To sum up, the practical anesthesiology can gain from the contact with experimental physiology:

1. by preventing phenomena which it creates itself, like notorious hyperventilation, unnecessary hyperoxia considered one of anesthesiological virtues, or overenthusiastic use of controlled ventilation;

2. recognizing the dangers inherent in playing with many physiological systems, autonomic nervous system in the first place;

3. adjusting balanced anesthesia according to new physiological findings; so called combined anesthesia opens wide possibilities for modifications and improvements.

Physiology, on the other hand, is offered an insight into a real world of an organism subjected to various forms of stress; it allows to study new frontiers of human adaptability.

110

CONTROL OF BREATHING BY NEUROPEPTIDES

Marie-Pierre Morin-Surun, Jean Champagnat, Arthur Foutz, and Monique Denavit-Saubie

Laboratoire de Physiologie Nerveuse C.N.R.S., Gif-sur-Yvette, France

Respiratory rhythm is partly generated by the periodic actions of classical neurotransmitters which act at the synapse level between different types of respiratory neurons. Periodic glutamate-like excitations (Denavit-Saubie et al., 1978) have been described as well as GABA- and glycine-like inhibitions (Champagnat et al., 1982). A new type of neuroactive substance, neuropeptides, are present in the respiratory networks and are involved in the control of the respiratory rhythm. We have been interested in those which are released in the nucleus tractus solitarii (NTS): cholecystokinin, enkephalins, substance P, neurotensin (Morin-Surun et al., 1983, 1984a,b,c, 1986; Hurle et al., 1985a).

Neuropeptides may act in different ways:

1. They may act on neuronal discharge frequency and in addition on glutamate-like excitations (enkephalins, neurotensin), thus specifically on respiratory modulation.

2. They may act on neuronal frequency without any interaction with a neurotransmitter known to be necessary for respiratory modulation (cholecystokinin). Such neuropeptides are not involved in rhythm generation, however, they may modulate respiration either directly or indirectly by an action on cerebral structures which influence respiration.

METHODS

The pharmacologic effects of neuropeptides were studied by different methods.

1. We tested their effect directly on respiratory neurons, either on their spontaneous pattern of discharge or on the pattern of discharge by local application of classical neurotransmitters involved in respiratory rhythm. In these cases, neurons were recorded by the central barrel of a multibarrelled micropipette. The neuroactive substances were applied in the vicinity of the recorded neuron, either by microiontophoresis, or by microinjection, using small pressure pulses in the adjacent barrel (for further details, see Morin-Surun et al., 1986).

111

2. We examined the effect of their release in the respiratory area on the respiratory motor nerve. For this purpose, the phrenic nerve of anaesthetized, bivagotomized and spontaneously breating cats was recorded. The neuroactive substances were applied in the electrophysiologically identified respiratory area. The tip of the micropipette was broken under light microscopic observation in order to obtain a drop with a diameter ranging from 300 to 400 ~ (for further details, see Morin-Surun, 1986).

3. We examined the possible cerebral structures involved in the effect of neuropeptides on respiration. For this purpose, systemic administrations were performed in different preparations (intact, bivagotomized, decerebrated, Morin-Surun et al., 1986; Champagnat et al., 1984). The ventilation was recorded using pletysmograph or pneumotachograph and the drugs were administered intravenously trough an implanted catheter in order to avoid animal manipulation.

SPECIFIC ACTION OF NEUROPEPTIDES ON RESPIRATORY MODULATION

We have shown that opioid peptides and neurotensin modified the glutamate induced excitation and thus act on synapses involved in respiratory rhytmogenesis.

Opioid Peptides

Enkephalins applied by iontophoresis depress the discharge frequency of respiratory neurons located either in the dorsal respiratory group (DRG) or ventral respiratory group (VRG). It is the maximal discharge which is depressed by this neuropeptide while the basal frequency is not modified. This effect seems to be due to a specific interaction with the periodic excitation induced by glutamate, since enkephalin depresses the neuronal discharge induced by iontophoretic application of glutamate (Denavit-Saubie et al., 1978).

The respiratory ramp is due to periodic excitations (Von Euler, 1986). Enkephalin slows down the formation of the inspiratory ramp by an inhibition of the glutamate-like excitation (Denavit-Sauble et al., 1978). Then, the off-switch mechanism is delayed until its threshold is reached. Enkephalins are located in soma and terminals of neurons included in the respiratory areas (Elde et al., 1976). It is possible that these interneurons are responsible for the modulation of the inspiratory ramp. Besides the involvement of enkephalins in the respiratory control, it has been shown that beta-endorphins are present in terminals in the nucleus tractus solitarii (Palkovits and Eskens, 1987) and can modulate respiration (Holaday, 1985).

The different opioid systems act on different opioid receptors (see review by Roques, 1985). We have investigated the role of two different opioid receptors: the mu receptors, on which all endogenous peptides have the same affinity (Pasternak, 1986), and the delta receptors, which are more specific to enkephalin ligands (Hughes and Kosterlitz, 1983). We have shown that both mu and delta opiate receptors are present, distinct and functionally active on bulbar respiratory-related neurons since iontophoretic application of the specific agonist of mu (TRIMU-4) and delta (DSLET)induced, as do enkephalins, a depression of spontaneous discharge frequency and glutamate induced excitation of respiratory related neurons (Morin-Surun et al., 1984a). These different receptors are differently involved in the respiratory depression occurring after the administration of opiates. The delta agonist, DSLET, injected intraperitoneally in chronic rats, reduced respiratory frequency while

112

Sp/ a.

240

HIO

1mln

G 40nA

Fig . 1. Potentiation of glutamate-induced excitation by neurotensin. Respiratory discharge frequency evoked by glutamate iontophoretic application was potentiated by neurotensin pressure pulses application. Respiratory discharge frequency was expressed in spike per second (sp/s). Drug applications are represented by horizontal bars.

the mu agonist, TRIMU-4, induced a depression of tidal volume (MorinSurun et al., 1984b). These different effects could be related to the distinct distribution of high density of these receptors in the different respiratory areas: the pneumotaxic center, including the nucleus parabrachialis medialis and the Kolliker-Fuse nucleus, contains a very high density of delta binding sites, while the dorsal respiratory nucleus which corresponds to the nucleus tractus solitarii, is more heavily labelled by the mu ligand (Sales et al., 1985). Since application of opioid peptides on the pontine dorsa-rostral surface induced a reduction of respiratory frequency and the application of opioid peptides on the medullary ventral surface reduced the tidal volume (Hurle et al., 1985b), the respiratory effects of specific agonists of mu and delta receptors injected intraperitoneally could be due to a fixation of the delta agonists in the pneumotaxic center and of the mu agonists in the dorsal respiratory group.

Neurotensin

Neurotensin is located in terminals found in the DRG (Yamazoe et al., 1985). We have shown that neurotensin, applied by microiontophoresis or by pressure pulses on respiratory related neurons in the DRG increases their firing frequency. On some neurons, neurotensin prolonged the duration of the inspiratory discharge either at the beginning or at the end of inspiration. Thus, it seems that excitations induced by neurotensin are able to delay the off-switch mechanism (Morin-Surun, 1986). This action is due to a potentiation of glutamate-like excitation responsible for the central inspiratory activity because we found that neurotensin, applied simultaneously with glutamate on a respiratory neuron, increased the firing frequency induced by glutamate in a proportion superior to the sum of the independent excitations induced by each drug (Fig. 1).

113

Neurotensin injected by micropressure into a group of inspiratory neurons of the DRG induced an apneustic pattern of respiration, but when injected in the VRG it did not modify respiratory rhythm. The apneusis can be related to the effects observed on respiratory-related units since the potentiation of glutamate induced excitations block the inhibitory off-switch mechanism. The contents of neurotensin-like terminals in the DRG could be produced by hypothalamic somata (Sawchenko and Swanson, 1982). We hypothesized that neurotensin released by a mechanism independent of respiratory input, can specifically modulate the respiratory rhythm to adapt the respiration to certain homeostatic conditions. For example, the hypothalamus is implicated in thermoregulation. It is well known that neurotensin induces hypothermia (Bissette et al., 1976) and that respiration is modulated during hypothermia (Kiley et al., 1984). Such a modulation may be due to a release of neurotensin into the DRG.

In conclusion, some neuropeptides could modulate respiratory rhythm by an action on glutamate induced excitation, either by an inhibition of the inspiratory ramp mechanism (enkephalin) or by a negative interaction with the off-switch mechanism (neurotensin).

UNSPECIFIC ACTION OF NEUROPEPTIDES ON RESPIRATORY MODULATION

Other neuropeptides such as cholecystokinin or substance P modulate respiration by a direct action on the neuronal firing frequency without interaction with known neurotransmitters implicated in the respiratory neurogenesis.

Cholecystokinin Peptides

The sulfated octapeptide carboxy terminal of cholecystokinin (CCK-8) was shown to be the predominant form of cholecystokinin present and active in the central nervous system of mammal where it was suspected to act as a neurotransmitter (Larsson and Rehfeld, 1979). Other forms of cholecystokinin such as the tetrapeptide carboxyl terminal of cholecystokinin (CCK-4) have also been detected. CCK-8 was found in terminals in the DRG, and thus could be released onto respiratory neurons. Applied by microiontophoresis on DRG respiratory neurons CCK-8 acts as an inhibitory neurotransmitter while CCK-4 acts as an excitatory one. The action of CCK-8 was direct through specific receptor mechanisms since CCK-8-induced inhibitions were not reproduced by application of related peptides and were resistant to antagonists of different inhibitory transmitters such as strychnine or bicucucline (Morin-Surun et al., 1983). Thus CCK-8 released into DRG can modulate the respiratory rhythm without interaction with neurotransmission. The respiratory control by CCK-8 could be considered a "non-specific" respiratory control in comparison with the "specific" respiratory control exerted by enkephalins or neurotensin.

The Substance P

Substance P (SP) is present in the DRG terminals and could be the neurotransmitter for vagal or glossopharyngeal afferent (Ljungdahl et al., 1978). Applied by microiontophoresis, SP increases the discharge frequency of respiratory neurons which respond to vagal or glossopharyngeal stimulation (Morin-Surun et al., 1984c). Respiratory related activities have been shown to involve opiate-sensitive excitatory synapses (Denavit-Saubie et al., 1978). SP excitations in the DRG are not sensitive to the action of opiates while simultaneously, glutamatergic transmission is opiate sensitive. This would suggest that

114

SP is not involved in the rhythmic excitatory synapses responsible for the central inspiratory activity (Morin-Surun et al., 1984). Like cholecystokinin, SP respiratory control could be considered a "nonspecific" respiratory control.

NEUROPEPTIDES COULD MODULATE RESPIRATION BY AN ACTION ON STRUCTURES WHICH PROJECT TO THE NTS

Brainstem respiratory areas are not the only structures which possess neuropeptide receptors. Neuropeptides released in structures which project to respiratory areas can modify the respiratory rhythm. To test this hypothesis, we observed the effects of neuropeptides injected systemically on animal preparations in which these projections were interrupted.

Contribution of Suprapontine Structures to Modulation of Respiration Qy Neuropeptides

Intravenous injection of CCK-8 in chronically implanted cats (for technique see Foutz et al., 1983) stimulated the respiratory frequency; this effect was abolished in decerebrated animals (Hurle et al., 1985a). This demonstrates that CCK-8 can stimulate respiration by an indirect action on suprapontine structures.

Contribution of Vagal Afferents in Neuropeptide Modulation

The participation of vagal afferents in the neuropeptide modulation of respiration has been observed with CCK. In decerebrated cats, CCK-8 induced a decrease of the tidal volume which is also observed in chronically implanted animal; such an effect is suppressed by bilateral vagotomy (Hurle et al., 1985a). This depressant effect of CCK-8 could be due to a peripheral action coming from by vagal nerves. We can also suppose that CCK-8 acts directly on CCK-8 receptors present in vagal nerves (Zarbin et al., 1981).

CONCLUSION

Neuropeptides can modulate respiration by different mechanisms:

1. By a direct action on respiratory neurons. In this case the effect would be dependent on the conditions of local neuropeptide release. If this release is due to "nonrespiratory stimuli", we suppose that the role of neuropeptides would be to adapt respiration to such stimuli (vigilance, cardiovascular or digestive functions). If this release is due to "specific respiratory" stimuli, such as a stimulation of vagal afferents, neuropeptides could be considered to be neurotransmitters directly involved in the specific control of respiration.

2. By an interaction with other neurotransmitters implicated in respiratory neurogenesis. Their respiratory actions also depend on their condition of release. We can distinguish two possibilities: they are tonically released in respiratory areas (e.g. enkephalins) and thus participate in respiratory rhythmogenesis, or they are released under certain conditions (e.g. neurotensin) and thus modulate respiration by an interaction with mechanisms involved in respiratory neurogenesis.

3. By an indirect action on other structures not involved in respiratory neurogenesis but implicated in the control of respiratory rhythm.

11 5

Thus, the role of neuropeptides in respiration is complex. Furthermore, the presence of multiple receptors for a given family of neuropeptides permits a role of neuromodulation by means of different mechanisms. The complexity of the neurotransmission system which controls this vital function allows a refined adaptation to any changes of the homeostatic condition.

REFERENCE

Bissette, G., Nemeroff, C.B., Loosen, P.T., Prange, A.J., and Lipton, M.A., 1976, Hypothermia and intolerance to cold induced by instracisternal administration of hypothalamic peptide neurotensin, Nature, 262: 607.

Champagnat, J., Denavit-Saubie, M., Moyanova, S., and Rondouin, G., 1982, Involvement of aminoacids in periodic inhibitions of bulbar respiratory neurones, Brain Res., 237: 351.

Champagnat, J., Denavit-Saubie, M., Morin-Surun, M.P., and Roques, B.P., 1984, Interaction of opiates with neurotransmission in the nucleus of the tractus solitarius in rats,~ Physiol. Lond., 354: 152P.

Denavit-Saubie, M., Champagnat, J., and Zieglgansberger, W., 1978, Effects of opiates and methionine-enkephalin on pontine and bulbar respiratory neurones of the cat, Brain Res., 155: 55.

Elde, R., Hokfelt, T., Johansson, 0., and Terenius, L., 1976, Immunochistochemical studies using antibodies to leucine-enkephalin: initial observations on the nervous system of the rat, Neuroscience, 1: 349.

Euler, C., von, 1986, Brain stem mechanisms for generation and control of breathing pattern, in: "Handbook of Physiology - The Respiratory System, vol.II -Control of Breathing, part one", A.P. Fishman, N.S. Cherniack, J.G. Widdicombe, and S.R. Geiger, eds., American Physiological Society, Washington, D.C.

Foutz, A.S., Dauthier, C., and Kerdelhue, B., 1983, B-endorphin plasma levels during neuromuscular blockade in unanesthetized cat, Brain Res. 263: 119.

Holaday, J.W., 1985, Endogenous opioids and their receptors, in: "Current Concepts", Kalamazoo, ed., The Upjohn Company, Michigan.

Hughes, J., and Kosterlitz, H.W., 1983, Introduction to opioid peptides, Br. Med. Bull., 39: 1.

Hurle, M., Morin-Surun, M.P., Foutz, A.S., Boudinot, E., and DenavitSaubie, M., 1985a, Different targets involved in the effect of cholecystokinin on respiration, Eur. J. Pharmacal., 118: 87.

Hurle, M.A., Mediavilla, A., and Fl6rez:- 1985b, Differential respiratory patterns induced by opioids applied to the ventral medullary and dorsal pontine surfaces of cats, Neuropharmacol., 24: 597.

Kiley, J.P., Eldridge, F.L., and Millhorn, D.E., 1984, The effect of hypothermia on central neural control of respiration, Respir. Physiol., 58: 295.

Larsson, L.I., and Rehfeld, J.F., 1979, Localization and molecular heterogeneity of cholecystokinin in the central and peripheral nervous system, Brain Res., 165: 201.

Ljungdahl, A., H8kfelt, T., Nilssonn, G., and Goldstein, M., 1978, Distribution of substance P-like immunoreactivity in the central nervous system of the rat. I. Cell bodies and nerve terminals, Neuroscience, 3: 861.

Morin, M.P., De Marchi, P., Champagnat, J., Vanderhaeghen, J.J., Rossier, J., and Denavit-Saubie, M., 1983, Inhibitory effect of cholecystokinin octapeptide on neurons in the nucleus tractus solitarius, Brain Res., 265: 333.

1 16

Morin-Surun, M.P., Boudinot, E., Gacel, G., Champagnat, J., Roques, B.P., and Denavit-Saubie, M., 1984a, Different effects of mu and delta opiate agonists on respiration, Eur. ~Pharmacal., 98: 235.

Morin-Surun, M.P., Gacel, G., Champagnat, J., Denavit-Saubie, M., and Roques, B.P., 1984b, Pharmacological identification of delta and mu opiate receptors on bulbar respiratory neurons, Eur. ~ Pharmacal., 98: 241.

Morin-Surun, M.P., Jordan, D., Champagnat, J., Spyer, K.M. and DenavitSaubie, M., 1984c, Excitatory effects of iontophoretically applied substance P on neurons in the nucleus tractus solitarius of the cat: lack of interaction with opiates and opioids, Brain Res., 307: 388.

Morin-Surun, M.P., Marlot, D., Kessler, J.P., and Denavit-Saubie, M., 1986, The excitation by neurotensin of nucleus tractus solitarius neurons induces apneustic breathing, Brain Res., 384: 106.

Palkovits, M., and Eskay, R.L., 1987, Distribution and possible origin of B-endorphin and ACTH in discrete brainstem nuclei of rats, Neuropeptides, 9: 123.

Pasternak, G.M., 1986, Multiple morphine and enkephalin receptors: biochemical and pharmacological aspects, Ann. N.Y. Acad. Sci., 467: 130.

Roques, B.P., 1986, Pharmacology of different classes of cerebral opioid receptors, Annales d'Endocrinologie, 47: 88.

Sales, N., Riche, D., Roques, B.P., and Denavit-Saubie, M., 1985, Localization of mu and delta-opioid receptors in cat respiratory areas: an autoradiographic study, Brain Res., 344: 382.

Sawchenko, P., and Swanson, L.W., 1982, Immunochistochemical identification of neurons in the paraventricular nucleus that project to the medulla or to the spinal cord in the rat, ~ Camp. Neural., 205: 260.

Yamazoe, M., Shiosaka, S., Shibasaki, T., Ling, N., Tateishi, K., Hashimura, E., Hamacka, T., Kimmel, J.R., Matsuo, H., and Tohyama, H., 1985, Distribution of six neuropeptides in the nucleus tractus solitarius of the rat: an immunochistochemical analysis, Neuroscience, 13: 1243.

Zarbin, M.A., Wamsley, J.K., Innis, R.B., and Kuhar, M.J., 1981, Cholecystokinin receptors: presence and axonal flow in the rat vagus nerve, Life Sciences, 9: 697.

11 7

NEUROTRANSMISSION AND NEUROMODULATION INVOLVED

IN THE CONTROL OF RESPIRATION

Pawe,l' Grieb

Department of Neurophysiology, Polish Academy of Sciences Medical Research Centre, Warsaw, Poland

INTRODUCTION

In classical textbooks of physiology the network generating respiratory rhythm was described as an "oscillator", and it used to be modelled "cybernetically" by analogy to electronic oscillators. Such an interpretation resulted from the then-actual concept of synaptic transmission as a mere continuation of nerve conduction.

During the last decade a substantial body of data has been accumulated concerning the so-called "non-classical" neurotransmission, and various neuromodulatory phenomena. It now appears that synaptic transmission is far more complicated than previously anticipated. Concomitantly, the concept of a central pattern generator (CPG), a neural network capable of generating efferent rhythmic activities and modulating them in response to differential activation of afferent inputs, was introduced. The actual output pattern of a CPG is the result of interplay between intrinsic properties of neurons, synaptic "wiring diagram", afferent influences and various modulatory phenomena.

Below some current concepts of neurotransmission and neuromodulation are reviewed, with reference to the generation and control of respiratory rhythmicity. To visualize the complexity of the respiratory pattern generator (RPG), a short description of a simple, relatively well known invertebrate CPG will be given for comparison.

NEUROTRANSMISSION AND NEUROMODULATION

Definitions concerning chemical communications of neurons were continuously evolving as new mechanisms have been discovered. At first, neurotransmission was described as a chemical transfer of neuronal excitation involving fast and short-lasting alterations of membrane conductances gated by ligands (neurotransmitters), occurring through engagement of receptors at postsynaptic membranes. This is the type of transmission which takes place in nerve-muscle terminals, the first model used for studying synaptic mechanisms. To underscore its "historical" background this type of neurotransmission is frequently called "classical". Later the term "neuromodulation" appeared, referred to as a chemical modification of neuronal excitability which, by itself, does not

119

produce direct effects on membrane conductances, but changes the action of "classical" neurotransmitters. "Extreme" suggestion was that any substance capable of altering neuronal excitability, including C02 and NH3, should be called a neuromodulator. However, in my opinion, this term shall be restricted to first (extracellular) messengers, ie. substances acting through interactions with specialized membrane receptors.

The "Classical" Theory of Neurotransmission

The molecular basis of neuronal excitability lays in the behavior of membrane proteins, ionic channels. The channels govern membrane conductances for small cations (Na+, K+, ca+2) and anions (mainly Cl-). The conductances, in turn, (together with ionic pumps, diffusion, transmitter inactivation and uptake, etc.) govern membrane potentials.

Channels are voltage-gated (voltage-dependent), or ligand-gated (voltage-independent). "Classical" concept of neurotransmission assumed that neurotransmitters are ligands of voltage-independent channels located at postsynaptic membranes. A neurotransmitter "quanta" released from a presynaptic terminal diffuse through a synaptic cleft and, by acting on receptors at postsynaptic membrane, induce fast and brief increases in a ligand-gated membrane conductance (ie. activate a channel). The subsequent flow of ionic current leads either to depolarization, or to hyperpolarization of postsynaptic membrane, creating "quantal" EPSP or IPSP. These miniature potentials are then subject to spatial and temporal summation at the bulk of the cell membrane. If a threshold depolarization is reached, voltage-gated channels open and action potential is generated.

The "synaptic language" seemed to be rather simple (EPSP, silence, or IPSP). It was thought that the clue of signal processing lays primarily in "the wiring diagram" describing how neurons are connected by excitatory and inhibitory synapses. As Brown (1986) states, "chemical transmission of this form may be compared with the role of a transistor in a large hard-wired network, implying that the entire nervous system might be fully comprehensible given sufficient knowledge of the circuitry ... ". Since a neuron usually carries thousands of synapses, the mechanism described above by itself would provide for tremendous si~1al processing capability. Already epistemologically difficult situation drastically worsened due to the research done in the last decade.

Multiple Ionic Channels

It now appears that several different types of ionic channels are present at an average neuron. A given neurotransmitter may activate many different types of ionic conductances, even at the same cell (cf. Brown, 1986). Furthermore, some neurotransmitters may modify (suppress or enhance) endogenous membrane currents which are active at the resting state.

The set of channels determines the intrinsic properties of a neurone, ie. its characteristic pattern of response to various stimuli. Some neurons display highly organized electrophysiological behaviors triggered by elementary stimuli (like, for example, a change in membrane potential, Dekin and Getting, 1984); in "true" pacemaker cells rhythmicity is stimulus-independent.

Channels are subject to complex regulations by second (intracellular) messenger systems. Some of them respond to the concentration of intracellular ligands like ca+2 or ATP. Such channels upon activa-

120

tion induce conductance changes of markedly slower onset and longer duration, and generate the so-called slow IPSP's and EPSP's. Recent reports indicate that synaptic mediators may not act on ionic channels directly. Instead, coupling of activated receptors to their responsive channels through specialized membrane-associated GTP-binding proteins (N- and G-proteins) may occur. This coupling by itself may undergo complicated regulations, employing second messengers c-AMP, inositol-1,4,5-triphosphate and diacylglycerol, multiple specialized enzymes, and a cascade of reactions, consequences of which are not restricted to a single elementary change in membrane ionic conductance (Berridge, 1984; Brown, 1986). In fact, spatial and temporal summation of EPSP's and IPSP's may not be a local phenomenon spreading across the neurolemma, but it may employ complex metabolic changes deeply rooted in the neuroplasm.

"Non-classical" Neurotransmission: Framework for Multiple Local Controls

Me Geer et al. (1978) defined two modes of action of neurotransmitters: ionotropic ("classical"), where receptor sites are identical with, or tightly coupled to respective channels, and metabotropic, which employ second messenger systems. Usually ACh, aspartate, glutamate, GABA and glycine are quoted as examples of ionotropic neurotransmitters. However, the same neurotransmitter may exert both ionotropic and metabotropic actions on different types of receptors, even at the same neurone. For example, ACh acts ionotropically through nicotinic, but metabotropically through muscarinic receptors (cf. Eccles, 1986).

The above mentioned mechanisms form a framework for multiple controls of different ionic channels by a single neurotransmitter. The basis for control of one ionic channel by multiple chemical mediators, and also for a cooperation (synergistic or antagonistic) between different mediators is provided by the phenomenon of co-transmission. Probably in most of the neurons two, three, or more neurotransmitters coexist (Lundberg and Hokfelt, 1983). In majority of cases a nonpeptide neurotransmitter is accompanied by one or more neuropeptides, which are co-stored at the presynaptic terminal, and either co-secreted, or independently released on stimulation (Viveros et al., 1983). Recently Millhorn et al. (1987) reported the coexistence of two classical neurotransmitters, one excitatory (5-HT), the other inhibitory (GABA) in neurons of the ventral medulla, and speculated that these mediators may be differentially released, depending on the nature of the presynaptic excitation.

Co-transmission adds still more regulatory mechanisms to a synapse. First, one (or more) of co-transmitters may influence (modulate) presynaptic release. Usually it is a negative feedback, but positive presynaptic modulation has also been described (Jacket al., 1981). Second, a ionotropic "fast" neurotransmitter is usually accompanied by the other(s) acting metabotropically, which upon co-secretion may, for example, act synergistically and induce slow postsynaptic potential to follow the fast one. The metabotrope may not necessarily be a peptide. Ionotropic action of ATP accompanied by the more "typical" noradrenaline acting metabotropically has been described (Sneddon and Westfall, 1984). The interactions between co-secreted neurotransmitters may also occur through intramembrane receptor-receptor interactions (for refs. see Agnati et al., 1986).

The concept of electrometabolic integrative units (see Agnati et al., 1986) provides tentative explanation of the functional significance of these multiple and diverse controls involved in synaptic transmission. If we keep believing that a basic functional unit of CNS logic is a feed-

121

back (less frequently feedforward) loop circuit, presynaptic modulation and intramembrane receptor-receptor interactions may be interpreted as steps towards miniaturization and, at the same time, multiplication of the basic circuit(s). The result is redw1dancy of information processing, hence improved faultproofness of integrative capabilities of CNS.

Anatomically- and Chemically-Addressed Communications

Lets now return to the problem of definitions. The "classical" definition of neurotransmission falls too short to cover all the mechanisms described to date. Eccles (1986) suggests to use the Burnstock's definition, which (slightly changed) can be put as follows: a neurotransmitter is a substance synthetized and stored presynaptically, released to the synaptic cleft upon presynaptic excitation, and interacting with specific postsynaptic membrane receptors to influence postsynaptic excitability or excitation state. We may infer that the mechanism of neurotransmission may involve ionotropic, metabotropic, or intramembrane receptor-receptor interactions. It may not even by itself produce changes in membrane conductances. What matters is that the system is local and anatomically-addressed, the information being transmitted through a specialized synaptic junction. According to such a definition many phenomena traditionally referred to as neuromodulatory shall be classified as neurotransmittory. What then shall neuromodulation be referred to? Is such a term necessary at all?

Indeed it is. First, presynaptic action of neurotransmitters does not engage receptors at the postsynaptic membrane. Therefore I suggest to call it "local presynaptic modulation". Second, as pointed out by Iversen (1986), the precise anatomical connections between feeder and effector cells may not always be needed, or even desirable. Chemical mediators may be released in a more diffused manner, eg. from non-synaptic parts of neurons, to reach more than one addressee. Such a mechanism is sometimes called "neurohormonal communication", or (in invertebrates) - the "paracrine" effect. In such a system the precise mechanisms by which mediators act may also be very diverse. What matters is that such systems are not local and confined to synaptic boutons, but they are more or less diffuse, their responsive elements marked by appropriate receptors. Such a system is chemically addressed. This is the kind of chemical communication which shall be referred to as neuromodulation. Diffuse neuromodulatory systems may be "spillovers" of anatomically addressed intrasynaptic systems (Kuhar, 1985, as discussed by Agnati et al., 1986). They may represent specialized mechanisms of communications between groups of neurons. They may also be the links between peripheral circulating hormones and their CNS receptors.

Neurotransmission and neuromodulation, as presently defined, illustrate two principal types of information transfer within CNS, described by Agnati et al. (1986) as "wiring", ie. "neuron-linked", and "volume of transmission", ie. "humoral" electrochemical transmissions, respectively. Noticeably, the general concept of electrometabolic integrative units is open to incorporate also electrotonic coupling mechanisms, which may play a role in vertebrate and mammalian CNS (Vizi, 1984).

The unusual complication of synaptic phenomena, providing for almost infinite variety of controls over neuronal excitation, provokes the conclusion that it may be premature to formulate any general rules about the mode of action of neuromediators. Many of them may act both as neurotransmitters, as local presynaptic modulators, and as diffuse neuromodulators in different CNS areas, or even at the same local level. As Brown (1986) puts it, "each cell type has to be treated as an individual entity, whose responses to transmitters ... may vary with environ-

122

mental and developmental changes". Before considering these phenomena in a rather complex RPG lets consider a small and simple invertebrate pattern generator to find out how conclusions can be made concerning the role of chemical transmission in generation and final shaping of its output.

THE PYLORIC CIRCUIT - LESSONS FROM A SIMPLE CPG

One of the simplest and best understood CPG's is the pyloric circuit of the stomatogastric ganglion of the spiny lobster, Panulirus interruptus. The update of research on this network has recently been published by Harris-Warrick and Flamm (1986).

The circuit contains 14 neurons belonging to six major classes. The classification is functional, based on the type and "address" of synaptic connections. The neurons thus classified possess different intrinsic properties. For example, four are "conditional bursters", which can generate large oscillatory potentials underlying bursts of action potentials. The understanding of the diversified intrinsic properties of neurons in the circuit, and of the multiple, mutually inhibitory and of different relative strength, synaptic connections (some of which are electrical), makes it possible to explain qualitatively the mechanism of generation of a stereotyped motor pattern observed upon isolating the circuit from sensory inputs. In vivo, the pyloric circuit is capable of producing considerably variable rhythms, resulting from influences exerted on the CPG by sensory and modulatory inputs, and subserving adaptative behavior of the animal.

Some insight into the mechanisms of modulation of the basic pattern came from research concentrated on the modes of action of an array of chemical messengers employed in the circuit, namely "classical" neurotransmitters (ACh, GABA), four different monoamines, and two peptides. It appears that these mediators are capable to alter every aspect of the CPG functioning. Bath application of the messengers, assumed to mimic differential activation of modulatory inputs, have been shown to alter the actual composition of the active circuit, by switching some cell classes off the circuit. Furthermore, elimination of some modulatory inputs inactivates, whereas bath application of some messengers induces certain intrinsic properties of particular neurons, eg. bursting and plateau potential capabilities. Also "conditional burster" cells can be prevented from bursting by a tonic inhibitory conductance; at least one mechanism of bursts induction by removal of inhibitory conductance upon the action of a neuromodulator has been described. All these influences create a very complex system, which is able to integrate tonic modulatory inputs into the stereotyped basic rhythm.

Considering a fourteen-cell circuit as an analogy to the respiration-controlling network of mammals and humans may seem far-fetched. Nevertheless, it reflects one of the principal epistemologic rules of the contemporary neurophysiology: to select and study relatively simple "model" neuronal systems which are experimentally tractable in a sense that a fully detailed description of their mode of action seems conceptually at reach. In fact, the simple CPG mentioned above seems to involve quite a few mechanisms which are active in mammalian CNS, in more complex pattern generators including that of respiration. On the other hand, respiratory CPG is, indeed, too complicated to be tractable in the sense of identifying the whole diversity of interactions, neuron-byneuron, synapse-by-synapse. Therefore, extrapolations from simpler "model" systems may be of a great heuristic value for respiratory neurophysiology.

123

HOW COMPLICATED IS RESPIRATORY PATTERN GENERATOR?

To rationalize the intuition that a mammalian respiratory CPG indeed pretty complicated, let's consider its two basic features: architecture, and the diversity of chemical mediator systems which involved in the network.

Cells and their Connections

is neural may be

Neuronal substrate of mammalian respiration has recently been reviewed (Long and Duffin, 1986). Neurons discharging phasically in synchrony with phrenic neurogram are located in pons, medulla and upper cervical cord, mainly in two groups, called ventral (VRG) and dorsal (DRG) respiratory groups. These "respiratory neurons" (RN's) can be classified by their relation to the respiratory cycle, as eg. inspiratory, expiratory, phase-spanning, etc.

Apparently not all RN's are necessary for rhythm generation. The number of RN's recorded in a given region may vary with experimental conditions; in particular, general anesthesia turns off many of them (Hukuhara, 1973), while phrenic output is preserved, though its reactivity markedly depressed (see eg. Borison, 1981). Experiments with brainstem sectioning, which showed that even upon dissecting out large parts of central respiratory network phrenic output is preserved, led to the hypothesis that a certain "critical mass" of neurons is necessary for rhythm generation (Karczewski and Gromysz, 1980). Autonomous spinal generators of respiratory rhythm have also been described (Aoki et al., 1980; Viala and Freton, 1983). Thus, respiratory network seems to be redundant as far as the ability to generate rhythm is concerned. However, the reactivity and plasticity of the network appears to be roughly proportional to the number of intact and active RN's.

Under general anesthesia some "silent" cells located in VRG may be activated by local nanoliter injections of excitatory aminoacid transmitters (McCrimmon et al., 1985). Similar effect has been found in unanesthetized cats upon iontophoretic application of glutamate (Foutz et al., 1987). We may conclude that the actual composition of the active RPG varies with experimental as well as physiological conditions, presumably in relation to the degree of excitation of brainstem reticular formation - a noticeable analogy to the pyloric circuit case. At present we may only speculate on the chemistry underlying these "recruitment" (and "derecruitment") phenomena.

Although some axonal pathways connecting respiratory "centers" have been identified, little is known about their functional significance. While some are known to be excitatory or inhibitory, Long and Duffin (1986) describe majority of these connections as "unknown" or "unproven", "too many ... for the proposal of other than speculative models" of respiratory rhythm generation.

We probably may assume that aggregates of RN's (respiratory "nuclei" or "centers") group cells with similar properties, and that such aggregates act as functional units of the network (Dussardier, 1985; McCrimmon et al., 1985). If this is the case, one of the main problems awaiting explanation is the nature of synchronizing mechanisms operating within and between clusters of RN's, which may be the functional analogues of the integrative elecrometabolic units in the macro-scale. Such mechanisms may involve the "volume of transmission" type of the electrochemical coupling. Both spatial arrangements of dendrites (Quattrochi and Rho, 1985) and neuromodulatory influences may be of great significance

124

here. For example, Denauvit-Saubi~ et al. (1985) found increase of inspiratory duration upon microinjections of neurotensin to the nucleus tractus solitarii (NTS). Somatostatin given intracisternally has the oposite effect, it induces expiratory apneusis (Fuxe et al., 1985). However, to prove the role of a given mediator in respiratory regulations, it is also necessary to establish physiological conditions, in which the neuromediator release occurs.

Finally, it shall be pointed out that some yet unidentified neuronal groups may play an important role in the "functional architecture" of the respiratory CPG (Long and Duffin, 1986). It may even be, that RN's recognizable by extracellular recordings (as those of DRG and VRG) are in fact not necessary for respiratory rhythm generation (Feldman et al., 1985), serving only as relays to various respiratory motoneurons. True "rhythm producers" may not display respiratory-modulated action potentials at all; instead, they may be, for example, "silent" (nonspiking) neurons with intrinsic pacemaker properties, acting as "command" cells which drive the "relay" RN's by wave-form modulatory influences, electrotonic and/or chemical.

Chemical Functional Anatomy

There are two elementary approaches to the identification of mediator systems involved in the RPG. One extends from the requirement that to play a role in respiratory control, a given mediator system has to be present in its neuronal substrate. The other is a functional requirement, that a "respiratory" mediator has to influence respiration upon its (more or less precise) application (agonistic and antagonistic analogs are also used in such studies).

The nrunber of proven, putative and expected chemical mediators in CNS probably exceeds 100 by now. The list is expected to grow. Snyder's (1980) prediction, that the number of peptide mediators may exceed 200, came close to reality due to recent discoveries.

Detailed maps of the distribution of numerous receptor types and mediator-containing neurons have been published, based on histochemistry and histoimmunology (see Hokfelt et al., 1984, for review). It became evident from these studies that brainstem, including many locations believed to be a part of RPG, is very rich in various mediator systems. Noticeably, the NTS region contains unusual multitude of non-peptide and peptide transmitter-containing cells, terminals, and postsynaptic receptors. These include most of the "classical" mediators, all three endogenous opioid families (enkephalins, dynorphins and endorphins, Akil et al., 1984), peptide YY (Robertson et al., 1986), atriopeptin (Kurihara et al., 1987), etc. Several cases of coexistence of "classical" and peptide transmitters were also found in the brainstem, including noradrenaline-, adrenaline-neuropeptide Y, adrenaline-neurotensin, 5-HT-substance P, -TRH, and -enkephalin pairs (see Hokfelt et al., 1984, for refs.).

The existence of a given mediator system in respiration-related neuroanatomical location(s) does not prove its involvement in respiratory controls. However, from other studies it seems that but all of the putative mediators may influence respiration. The effects depend on many conditions, including: the route of administration and dose, anesthesia and/or decerebration, preservation of vagi, etc. Species differences have also been observed. More specific method is iontophoretic application of putative mediators on RN's in situ or in vitro (tissue slices). The amount of data collected to date is large and cannot be reviewed here (for reviews see eg. Mueller et al., 1982; Hedner, 1983).

125

PRINCIPLES OF CHEMICAL COMMUNICATIONS IN RPG

It may seem totally premature to formulate generalizations concerning chemical communications within the respiratory-controlling network, while its structure and chemistry is only vaguely known. However, such hypotheses may appear to be quite useful as guidelines for further research, even if they only will be disproved.

Physiological Correlations - Ultimate Proof of ~ "Respiratory" Mediator

Recognizing the immense variety of chemical messenger systems in brainstem "respiratory" locations, the great complication of RPG anatomical structure and largely unknown biophysical properties of its building blocks, does not help in understanding mechanisms of physiological control of respiration. The system is equipped with multiple and, perhaps, redundant regulatory mechanisms, but their mere existence and responsiveness to artificial conditions andjor stimuli neither prove that they are physiologically meaningful and operative, nor explain when, how, and why are they activated or suppressed. The case of opioids illustrates, that the system may be equipped with control means which rarely (if ever) go into action, just being there as an "architectural artifact".

Respiratory depressant action of exogenous opiates is known for ages. We are now a decade after the first identification of opiate receptors in CNS and discovery of a first endogenous opioid peptide, and assured by numerous studies that both the opioid peptides (Akil et al., 1984) and their receptors (Wamsley, 1983) are abundantly present in the "respiratory" regions of the medulla. Undoubtedly RN's are inhibited by opioids. Still there is almost no "hard-core" evidence for an involvement of endogenous opioids in physiological respiratory control (see reviews by McQueen, 1983; Santiago and Edelman, 1985), except, perhaps, in perinatal period (Grunstein et al., 1981; Moss and Scrapelli, 1984). In intact adults under no anesthesia the opioid system seems just "being there" (ie. in RPG). "Positive" results, evidencing tonic depression of respiration by endogenous opioids in acute experiments (eg. Pokorski et al., 1981) might have been a "by-product" of surgical stress or trauma. Respiratory depression results from various surgical procedures, eg. spinal cord transections (Holaday and Faden, 1980) and brainstem sectioning (Janczewski and Grieb, 1986), or from nociceptive stimulation in anesthetized animals (Kumazawa et al., 1980).

Interestingly, opiates do not inhibit all the phenomena of respiratory regulation with the same strength. Budzinska et al. (1985) found that a facilitatory effect of morphine on the inspiratory-inhibitory (Breuer-Hering) reflex in rabbits was selective in that it was sustained in animals chronically pretreated with morphine, which apparently developed tolerance to the opiate and failed to respond to the drug with the decrease in respiratory frequency. Yet, no physiological analogy of such a selective facilitation of the Breuer-Hering reflex has, to my knowledge, been demonstrated.

Altogether, the observations are in apparent disparity with the high relative concentration of endogenous opioid system markers in "respiratory" brainstem areas. We may speculate, that in physiological situations the opioid system of RPG is somehow suppressed, eg. by some yet unknown "endogenous naloxone" (an analogy to the recently discovered "endogenous ouabain", Cloix et al., 1987). The fact is, that to establish the physiological role of a given neuromediatory system in respiratory control we need an ultimate proof: to identify the physiological conditions in which such a system is activated or suppressed.

126

Intrinsic Properties of Neurons and Respiratory Pattern Formation

In one respect the RPG differs from many simpler invertebrate CPG's - it does require a minimal excitatory input to produce rhythmic activities. For this reason the "pacemaker" hypothesis, which assumed the existence of intrinsically active rhythm generating neurons as the primary source of respiratory rhythmicity has largely been abandoned (see Euler, 1986). However, spontaneously bursting pacemaker neurons exist in vertebrate and mammalian CNS. For example, an inferior olive cell seems to be equipped with all necessary and sufficient ionic conductances to be an active single cell oscillator (Llinas and Yarom, 1980); also spinal cord cells are able to produce rhythmic bursts of action potentials without any synaptic input (Legendre et al., 1985). These findings may revitalize the search for intrinsically oscillating neurons in RPG. Such "pacemakers" may be inhibited by hypocapnia, opioids, or other mechanisms; they also may, perhaps, be non-spiking "silent" cells.

The autonomic oscillatory behavior is but one example of intrinsic properties of neurons, which may be essential for respiratory pattern formation, especially in the (yet unidentified) process of incorporating multitude of tonic excitatory inputs into a phasic rhythm. By virtue of possessing a specific set of ionic conductances, some neurons involved in respiratory rhythm generation may be preprogrammed to respond to stimuli by generating organized, complex electrophysiological behaviors.

It is not known, to what extent the RN's firing patterns are derived from synaptic inputs, and to what extent they are set by intrinsic properties of neurons. Recently conflicting reports appeared on the in vitro behavior of neurons from ventral part of NTS. Dekin and Gettig (1985) found some of them displaying rhythmic bursting following depolarization, or stimulation by TRH. Champagnat et al. (1985) did not find spontaneous bursting in neurons from the similar area, but identified ionic currents responsible for bursting in response to membrane depolarization.

Intrinsic properties of neurons usually are evaluated by in vitro studies to assure no synaptic input. But we may hypothesize, that the pattern of organized neuron response may be mediator-specific. Chemical messengers, in addition to influencing voltage-independent channels may also modify the responsiveness of voltage-gated channels; their action cannot be considered as restricted to generation of IPSP's and EPSP's. Some respiratory rhythm variants may perhaps be induced by turning on appropriate mediator-specific intrinsic properties of RN's. This may be the mechanism for a so-called "patterned response", ie. the complex and time-delayed response of respiratory CPG, evoked eg. by electrical stimulation of mesencephalic structures (Gauthier et al., 1983).

Mediators Control Hierarchy of Modulatory Influences

The actual respiratory pattern of an intact, unanesthetized animal or human is the result of "integration of requirements". The principal requirement is for respiratory gas exchange to be coupled to body metabolism. "Basic" respiratory rhythm is frequently interrupted, overriden by, or synchronized with other behavioral patterns. In some cases an "alternative" behavior (eg. coughing, sneezing, swallowing, vomiting, etc.) may momentarily override even a high chemical drive. In extreme cases (eg. in submaximal exercise) blood chemistry and other signals related to tissue metabolism establish the ultimate and overwhelming respiratory drive, which may only be modified by life-threatening thermal stress. The actual balance between "basic" and "alternative" output patterns generated by RPG depends on the situation within the body,

127

"seen" by various receptors and signalled by afferent pathways.

We may hypothesize, that particular neuromediator systems must be engaged in breaking "normal" respiratory pattern in order to execute another coordinated pattern behavior generated by some other specialized CPG. Searching for nature of these systems, and establishing their relative "impact" on the conventional respiratory drive shall be a fascinating, and possibly very important line of research.

An example is the medullary swallowing reflex. Swallowing CPG has been located in the NTS region (Jean, 1984). Swallowing pattern may be activated by electrostimulation of superior laryngeal nerve (SLN), and upon activation it "overrides" respiratory CPG (this is a classical "reflex inhibition of respiration by SLN stimulation"). Kessler and Jean (1986) found that both 5-HT and noradrenaline microinjecions into the "swallowing" region of NTS, and electrostimulation of monoaminergic brainstem regions inhibit the swallowing reflex. Does such a mechanism suppress swallowing reflex, and at the same time preserve "conventional" respiratory rhythm during enhanced chemical respiratory drive? Does it suppress swallowing during sleep, when serotoninergic mechanisms in NTS are involved in modulation of blood pressure and some other vegetative functions (see eg. Laguzzi et al., 1984)?

Another example is peptide YY (PYY), which is a potent emetic; neurons containing PYY were found in area postrema, and PYY receptors also in NTS and dorsal nucleus of the vagus (Robertson et al., 1986). If PYY triggers the emetic CPG, does it at the same time "override" the respiratory CPG? The PYY case turns attention into the possible direct influences of blood-borne chemical messengers (hormones) on respiration through the area postrema - nucleus tractus solitarii circuit. However, this fascinating subject will not be discussed here.

CONCLUSION

Due to its respectful complication, mammalian RPG represents a great challenge to neurophysiologists. Because chemical communications between neurons form the basis of CNS signal processing, the problem of origin and purposeful modulation of respiratory rhythm will not be adequately solved without sufficient knowledge of neurotransmittory and neuromodulatory phenomena. Investigative efforts in this area shall employ an array of complementary approaches, including: (i) model studies of simpler circuits and networks, to establish general principles of CPG's functional architectonics; (ii) intracellular recordings and specific probes of intrinsic properties of RN's in situ, and in vitro (tissue slices, etc.); (iii) confronting various aspects of neuroanatomy and histochemistry with electrophysiological correlates of RPG functions; (iv) establishing physiological and pathophysiological correlates for activation and deactivation of various mediator systems, and their coordination.

Hopefully, more defined generalizations will emerge, which will make it possible to select the most important mechanisms from pretty complex "background" of possible neural controls.

REFERENCES

Agnati, L. F., Fuxe, K., Zoli, M., Merlo Pich, E., Benfenati, F., Zini, I., and Goldstein, M., 1986, Aspects of the information handling by

128

the central nervous system: focus on cotransmission in the aged rat brain, Progr. Brain Res. 68: 291.

Akil, H. S., Watson, S. J., Young, E., Lewis, M.E., Khachaturian, H. and Walker, J. M., 1984, Endogenous opioids: biology and function, Ann. Rev. Neurosci. 7: 223.

Aoki~., Mori, S., Kawahara, K., Watanabe, H., and Ebata. N., 1980, Generation of spontaneous respiratory rhythm in spinal cats, Brain Res. 202: 51.

Berridge, M. J., 1984, Inositol triphosphate and diacylglycerol as second messengers, Biochem. ~· 220: 345.

Borison, H. L., 1981, Central nervous respiratory depressantsanesthetics, hypnotics, sedatives and other respiratory depressants, in: "Respiratory Pharmacology", J. G. Widdicombe, ed., Pergamon Press, Oxford.

Brown, D. A., 1986, Synaptic mechanisms, Trends Neurosci. 9: 468. Budzinska, K., Grieb, P., and Romaniuk, J.R., 1985, Morphine selectively

facilitates the inspiratory-inhibitory reflex in rabbits, Experientia 41: 458.

Champagnat, J., Richter, D. W., Jacquin, T. and Denavit-Saubie, M., 1985, Voltage-dependent conductances in neurones of the ventrolateral NTS in rat brainstem slices, in: "Nobel Conference 1985 on Neurobiology of the Control of Breathing", Karolinska Institutet, Stockholm, no. 32.

Cloix, J. F., Grichois, M. L., Borrits, D., Sylvestre, D., Boudet, J., Guicheney, P. and Meyer, P., 1987, Comparison of biochemical properties of semi-purified Endogenous Digitalis-Like Compounds (EDLC) from human urine and hemofiltrat, Neurosci. 22: S392.

Dekin, M. S. and Getting, P. A., 1984, Firing pattern of neurons in the nucleus tractus solitarius: modulation by membrane hyperpolarization, Brain Res. 324: 180.

Dekin, M. S. and Getting, P. A., 1985, Pattern formation at the neuronal level: role of intrinsic membrane currents, in: "Nobel Conference 1985 on Neurobiology of the Controlof Breathing", Karolinska Institutet, Stockholm, no. 31.

Denavit-Saubie, M., Champagnat, J., and Morin-Surun. M.P., 1985, Chemical transmission and central respiratory mechanismis, in: "Neurogenesis of Central Respiratory Rhythm", A. L. Bianchi, M. Denavit-Saubie, eds., MTP Press Ltd., Lancaster.

Dussardier, M., 1985, An overview of backgrounds and themes for research in neurogenesis of respiratory rhythm, in: "Neurogenesis of Central Respiratory Rhythm", A. L. Bianchi, M. Denavit-Saubie, eds., MTP Press Ltd., Lancaster.

Eccles, J. C., 1986, Chemical transmission and Dale's principle, Progr. Brain Res. 68: 3.

Euler, C., von, 1986, Brain stem mechanisms for generation and control of breathing pattern, Handbook of Physiology, The Respiratory System II, American Physiological Scociety, Washington, D.C.

Feldman, J. L., McCrimmon, D. R., Speck, D. F., Smith, J. C., and Ellenberger, H. H., 1985, Generation of respiratory pattern in mammals, in: "Nobel Conference 1985 on Neurobiology of the Control of Breathing", Karolinska Institutet, Stockholm, no. 29.

Foutz, A. S., Boudinot, E., Morin-Surun, M.P., Champagnat, J., Gonsalves, S. F., Denavit-Saubi~, M., 1987, Excitability of "silent" respiratory neurons during sleep-walking states: an iontophoretic study in undrugged chronic cats, Brain Res. 404: 10.

Fuxe, K., Agnati, L. F., Harfstrand, A., Mutt, V., Anderson, K., Hokfelt, T., Vale, W., Brown, M., and River, J., 1982, Cardiovascular and respiratory actions of somatostatin peptides following intracisternal injection into the a-chloralose anesthetized rat, Neurosci. Lett., Suppl. 10: 189.

Gauthier, P., Manteau, R., and Dussardier, M., 1983, Inspiratory on-

129

switch evoked by stimulation of mesencephalic structures: a patterned response, Exp. Brain Res. 51: 261.

Grunstein, M. M., Hazinski, T. A. and Schlueter, M.A., 1981, Respiratory control during hypoxia in newborn rabbits: implied action of endorphins, ~· ~· Physiol. 51: 122.

Barris-Warrick, R. M., and Flamm, R. E., 1986, Chemical modulation of a small central pattern generator circuit, Trends Neurosci. 9: 432.

Redner, J., 1983, Neuropharmacological aspects of central respiratory regulation, Acta Physiol. Scand., Suppl. 524.

Hokfelt, T., Johansson, 0., and Goldstein, M., 1984, Chemical anatomy of the brain, Science 225: 1326.

Holaday, J. W., and Faden, A. I., 1980, Naloxone acts at central opiate receptors to reverse hypotension, hypothermia, and hypoventilation in spinal shock, Brain Res. 189: 295.

Hukuhara, T., Jr., 1973, Neuronal organization of the central respiratory mechanisms in the brain stem of the cat, Acta Neurobiol. ~· 33: 219.

Iversen, L. L., 1986, Chemical signalling in the nervous system, Progr. Brain Res. 68: 15.

Jack, J. J. B., Redman, S.J., and Wong, K., 1981, The components of synaptic potentials evoked in cat spinal motoneurones by impulses in single group Ia afferents, ~· Physiol. 321: 65.

Janczewski, W. A. and Grieb, P., 1986, Naloxone enhances respiratory output in rabbits with various brainstem sections, Bull. Europ. Physiopath. Resp. 22, Suppl.: 8.

Jean, A., 1984, Brainstem organization of the swallowing network, Brain Behavior Evol. 25: 109. -- --

Karczewski, W. A. and Gromysz, H., 1980, The "Split Respiratory Centre". Lessons from brainstem transections, Adv. Physiol. Sci., vol. 10 "Respiration", p. 587.

Kessler, J. P., and Jean, A., 1986, Inhibitory influence of monoamines and brainstem monoaminergic regions in the medullary swallowing reflex, Neurosci. Letters 65: 41.

Kuhar, M. J., 1985, The mismatch problem in receptor mapping studies, Trends Neurosci. 27: 190.

Kumazawa, T., Tadaki, E. and Kim, K., 1980, A possible participation of endogenous opiates in respiratory reflexes induced by thin-fiber muscular aferents, Brain Res. 199: 244.

Kurihara, M., Saavedra, J. M. and Shigematzu, K., 1987, Localization and characterization of atrial natriuretic peptide binding sites in discrete area of rat brain and pituitary gland by quantitative autoradiography, Brain Res. 408: 31.

Laguzzi, R., Reiss, D. J., and Talman, W. T., 1984, Modulation of cardiovascular and electrocortical activity through serotoninergic mechanisms in the nucleus tractus solitarius of the rat, Brain Res. 304: 321.

Legendre, P., McKenzie, J. S., Dupouy, B. and Vincent, J. D., 1985, Evidence for bursting pacemaker neurones in cultured spinal cord cells, Neurosci. 16: 753.

Llinas, R. and Yarom, Y., 1980, Electrophysiological properties of mammalian inferior olivary cells in vitro, in: "The Inferior Olivary Nucleus: Physiology and Anatomy", J. Courville et al., eds., Raven Press, New York.

Long, S., Duffin, J., 1986, The neuronal determinants of respiratory rhythm, Progr.Neurobiol. 27: 101.

Lundberg, J. M., and Hokfelt, T., 1983, Coexistence of peptides and classical neurotransmitters, Trends Neurosci. 6: 325.

McCrimmon, D. R., Feldman, J. L., Speck, D. F., Ellenberger, H.

130

H., Smith, J. C. and Weese-Meyer, D., 1985, Functional heterogeneity of dorsal, ventral and pontine respiratory groups revealed by micropharmacological techniques, in: "Nobel Conference

1985 on Neurobiology of the Control of Breathing", Karolinska Institutet, Stockholm, no. 25.

McGeer, P. L., Eccles, J. C., and McGeer, E. G., 1978, "Molecular Neurobiology of Mammalian Brain", Plenum Press, New York.

McQueen, D., S., 1983, Opioid peptide interactions with respiratory and circulatory systems, Br. Med. Bull. 39: 77.

Millhorn, D. E., Hokfelt, T., Seroogy, K., Oertel, W., and Verhofstad, A. A. J., 1987, Immunohistochemical evidence for colocalization of gamma-aminobutyric acid and serotonin in neurons of the ventral medulla oblongata projecting to the spinal cord, Brain Res. 410: 179.

Moss, I. R. and Scarpelli, E. M., 1984, C02 and naloxone modify sleepjwake state and activate breathing in the acute fetal lamb preparation, Respir. Physiol. 55: 325.

Mueller, R. A., Lundberg, D. B. A., Breese, G. R., Redner, J., Redner, T., and Jonason, J., 1982, The neuropharmacology of respiratory control, Pharmacal. Revs. 34: 255.

Pokorski, M., Grieb, P. and Wideman, J., 1981, Opiate system influences central respiratory chemoreceptors, Brain Res. 211: 221.

Quattrochi, J. J., Rho, J. H., 1985, Three-dimensional tissue reconstruction reveals integrative structural features among neurons within central respiratory centers of the brainstem, in: "Neurogenesis of Central Respiratory Rhythm", A. L. Bianchi, M. Denavit-Saubie, eds., MTP Press Ltd., Lancaster.

Robertson, H. A., Leslie, R. A., and McDonald, T. J., 1986, PYY receptors in the medulla - autoradiographic studies, in: "Neurobiology of the Nucleus of the Solitary Tract, Int. Satellite Symposium of X Annual Meeting of ENA", Marseillie.

Santiago, T. V. and Edelman, N. H., 1985, Opioids and breathing, J. ~· Physiol. 59: 1675.

Sneddon, P., and Westfall, D.P., 1984, Pharmacological evidence that adenosine triphosphate and noradrenaline are co-transmitters in the guinea-pig vas deferens, ~· Physiol. 347: 561.

Snyder, S., 1977, Opiate receptors and internal opiates, Sci. Am. 236: 44.

Viala, D. and Freton, E., 1983, Evidence for respiratory and locomotor pattern generators in the rabbit cervico-thoracic cord and for their interactions, ~· Brain Res. 49: 247.

Viveros, 0. H., Diliberto, E. J., and Daniels, A. J., 1983, Biochemical and functional evidence for the cosecretion of multiple messengers from single and multiple compartments, Fed. Proc. 42: 2923.

Vizi, E.S., 1984, "Non-Synaptic Interactions betwen Neurons: Modulation of Chemical Transmission", John Wiley, New York.

Wamsley, J. K., 1983, Opioid receptors: autoradiography, Pharmacal. Rev., 35: 69.

131

THE CONTROL OF BREATHING MOVEMENTS IN THE FETUS

G.S. Dawes

Charing Cross Sunely Research Centre, Lurgan Avenue London W6 8LW, England

In 1970 I and my colleagues in Oxford and Paris rediscovered fetal breathing (Dawes, 1984). For, as you will remember, maternal abdominal movements in man had been ascribed to fetal breathing by von Ahlfeld (1888), a view rejected contumaciously by his contemporaries. And Windle (1940) decided that the observations of Snyder and Rosenfeld (1937) on the breathing movements of fetal kittens and puppies in utero were likely to be experimental artefacts. Barcroft (1946) concluded that in the sheep breathing movements were not normally present antenatally, from 60 days gestation until term (147 days), when they suddenly began and became continuous, possibly as a result of cold exposure. So the misleading words, the "onset of breathing at birth" were perpetuated in the scientific literature and popular belief.

The work since 1970 has brought to light five phenomena, some with analogies to studies on postnatal respiratory control, but as yet no satisfactory explanation. They point to the conclusion that we do not know how breathing is controlled in the lower brain stem, pons and medulla.

INTERMITTENT FETAL BREATHING

The fact that, in man as in sheep, fetal breathing movements are intermittent suggests that the drive to the presumed respiratory centre is low. There is other evidence pointing in the same direction. Prenatally the systemic arterial chemoreceptors function over the normal fetal blood gas range, but neither acute isocapnic hypoxia nor the administration of drugs which stimulate the carotid bodies postnatally (cyanide or lobeline for example) cause anything more than a few gasps. The drugs which convincingly stimulate fetal breathing movements, doxapram (Bamford et al., 1986), pilocarpine (Moore et al., 1988 in the press) and apomorphine (Bamford et al., 1986) and were suspected of stimulating the carotid bodies, all have their effects by central actions. There is as yet no simple test of the integrity of fetal carotid body/respiratory stimulation by drug injection.

Raising the fetal plasma Paco2 above its normal value (-43 mm Hg) in the sheep causes deeper and more regular breathing movements (when present, i.e. in low voltage electrocortical activity only). So does carbonic anhydrase inhibition by acetazolamide (Hohimer et al., 1985) or

133

central acidosis caused by perfusion of the cerebral ventricular system with< 1mM HC03 or i.v. infusion of ammonium chloride (Hohimer et al., 1983; Koos, 1985). But the rise in PaC02 required to demonstrate a clear effect is relatively large (5-7 mmHg), perhaps because the fetus is then in a condition analogous to rapid-eye-movement (REM) sleep postnatally, which is known to reduce central C02 sensitivity. This does not explain the absence of respiratory movement in high voltage electrocortical activity (without REM). It is attributed to descending inhibition, since near term breathing movements become almost continuous after section of the brain stem at or just above the upper pons (Dawes et al., 1983).

In a recent review Millhorn and Eldridge (1986) gave cogent reasons for doubting whether the ventrolateral medulla is the only site for central chemoreception. Jansen et al. (1987) reported that destruction of the intermediate areas "S" in the ventrolateral medulla produces apnoea in anaesthetized lambs postnatally, as in adult cats. However in unanaesthetized fetal lambs in utero, previous ablation of areas "S" (by touching the surface with a probe cooled in liquid N2 to give lesions 0.5 - 1.5 mm deep and 3- 4 mm diam), did not change normal episodic breathing movements; it modified but did not abolish the response to hypercapnia. In some of these experiments the systemic arterial chemoreceptors also were denervated. So we need to look elsewhere for (perhaps additional) areas of central C02 chemoreception.

HYPOXIC ARREST OF FETAL BREATHING MOVEMENTS

Reduction of the fetal Pa02 in sheep from its normal value (about 23 mm Hg in the last third of gestation) by 8 - 10 mm Hg causes gross diminution or arrest of fetal breathing movements as recorded by changes in transthoracic pressure, by diaphragmatic and intercostal EMG activity (Dawes, 1984). The phenomenon persists after section of the carotid nerves and vago-sympathetic trunk (Dawes, 1986). It is abolished by section of the brain stem in the upper pons or at the inferior colliculus; it persists after more rostral transection and after destruction of the hypothalamus (Dawes et al., 1980) or pituitary.

Among the possible explanations the simplest is interruption of a descending pathway from an oxygen-sensitive area located at the level of the superior colliculus. An alternative hypothesis is that the lesions damaged the blood supply to or venous drainage from the medulla. This was unlikely since the size of the lesions after transection was variable, but the results (abolition of the apnoeic response to hypoxia) invariable. And we found that medullary blood flow and its % increase on hypoxic exposure was unaltered by brain stem transection, as judged by microsphere injections in unanaesthetized fetal lambs in utero (Jensen et al., 1985), even though arterial pressure did not rise in hypoxia after transection.

It is not necessary to cut the whole brain stem to abolish the hypoxic arrest of fetal breathing movements. Small bilateral lesions in the upper lateral pons, placed stereotactically just rostral to the nuclei of the trigeminal nerve, have the same effect (Johnston et al., 1985; Gluckman and Johnston, 1987). Similar results were obtained using less elegant methods with lesions placed by direct vision after removal of the cerebellum (Parkes et al., 1984). Gluckman and Johnston (1987) conclude that their observations do not discriminate between destruction of a neural pathway or of cell bodies sensitive to hypoxia.

Apomorphine (100 ~g i.a.) increases the amplitude of fetal breathing movements in intact lambs, by an action above the pons since section of the brain stem abolishes the effect (Bamford et al., 1986). The fact

134

that administration of apomorphine causes the reappearance of fetal breathing movements (and of activity in the limb muscles) in hypoxia is further evidence in support of the conclusion that this degree of hypoxia does not cause profound cerebral depression. It does not cause gross changes in electrocortical activity, nor does it affect two cranial reflexes (Walker and Harding, 1984).

Experiments to try and identify a putative neurotransmitter responsible for the arrest of fetal breathing movements in hypoxia have not been successful. Fetal breathing can be arrested by administration of the GABA-agonist muscimol, but the arrest of breathing by hypoxia is not reversed by picrotoxin (Johnston and Gluckman, 1983). Similarly fetal breathing is arrested by the a2-adrenergic agonist clonidine, but the hypoxic arrest of breathing is not reversed by the antagonist idazoxan (Bamford et al., 1986).

STIMULATION OF FETAL BREATHING BY HYPOXIA : CENTRAL ACTION

One of the unexpected results of fetal brain stem transection in sheep (Dawes et al., 1983) was that isocapnic hypoxia then caused an increase in the rate and depth of breathing movements, even when the carotid nerves also were cut. The effect has been seen after section of both the carotid nerves and the vago-sympathetic trunks (Hanson, M., personal communication), so cannot be attributed to stimulation of the systemic arterial chemoreceptors. And it has been observed after placing small lesions bilaterally in the upper pons as described above (Gluckman and Johnston, 1987); the systemic arterial chemoreceptors were not denervated in those experiments.

It was possible that the increase in breathing during hypoxia after brain stem section (or the placement of appropriate lesions) might have been due to the failure of medullary blood flow to increase as much as in intact fetal lambs, with the development of central acidosis. The experiments already mentioned (Jensen et al., 1985) do not support this hypothesis. But further experiments are desirable, with local measurement of pH also.

STIMULATION OF FETAL BREATHING BY PROSTAGLANDIN SYNTHETASE INHIBITORS

Intravenous administration of inhibitors of prostaglandin synthetase (indomethacin, meclofenamate or acetylsalicylic acid) to fetal lambs near term causes the onset of continuous fetal breathing movements (Kitterman et al., 1979; Koos, 1985). The effect is not altered by denervation of the systemic arterial chemoreceptors, nor by section of the brain stem in the upper pons. It can be elicited by small doses of meclofenamate infused into the cerebral ventricles. The continuous fetal breathing is arrested by isocapnic hypoxia. This phenomenon is reminiscent of an analogous effect in adult animals (e.g. Cameron and Semple, 1986; before the action of salicylates on prostaglandin synthetase had been discovered), associated with neither CSF acidosis nor hyperthermia. The observations raise two interesting questions. Is the stimulation of breathing attributable to a decrease in local blood flow in some critical region of the medulla? Or are prostaglandins involved more intimately, at the cellular level, with the central control of breathing?

Hohimer et al.(1985) reported a fall in medullary blood flow of 15% on administration of indomethacin to fetal lambs measured by the microsphere method. Heyman and Rudolph (1976) observed a small increase in cerebral blood flow on injection of acetylsalicylic acid. Neither

135

group discriminated between high or low voltage electrocortical activity. We now know that there is an increase by 19-281 in blood flow to the medulla in low voltage electrocortical activity (when breathing is normally present) as compared with high voltage activity, i.e. the "wrong" way round (Jensen et al., 1985, 1986). The suggestion that small decreases in medullary flow are responsible for central acidosis is not persuasive. It looks as if changes in metabolic rate, linked to local central nervous activity, may be of more importance. The argument on blood flow, unrelated to local metabolism, is naive.

DIURNAL VARIATION IN FETAL BREATHING MOVEMENTS

The diurnal variations in fetal electrocortical activity, breathing movements, muscular activity, heart rate and heart rate variation have been examined in the sheep and man (cf. Dawes and Robinson, 1976; Patrick et al., 1978; De Vries, 1987; Dawes, 1988 in the press). The diurnal variation in man has been identified as early as 20-22 weeks, well before the appearance of behavioural changes characteristic of sleep states postnatally (de Vries, 1987). There is a remarkable temporal coincidence between spontaneous diurnal changes in maternal plasma ACTH and cortisol concentrations and fetal activity, especially breathing movements. As reviewed elsewhere (Dawes, 1988 in the press) the evidence in man strongly suggests a causal relationship. Administration of corticosteroids to the mother in a variety of disease states or experimental conditions abolishes fetal diurnal activity. The problem is to explain how a change in a maternal hormone could influence fetal breathing movements. A possible mechanism could be that the concentration of a fetal hormone is altered by placental metabolism and/or transfer. A possible candidate effector agent is progesterone, since it is believed to vary diurnally in the fetus near term and modulates breathing in the adult.

POSTNATAL BREATHING

It has been suggested that the relative insensitivity of the central chemoreceptors in the fetus is due to the lack of sensory input, and that the increase in input postnatally accounts for the changes in breathing at birth. There is ample evidence that external cooling of exteriorized fetal lambs causes the onset of sustained respiratory movements (Barcroft, 1946; Dawes, 1968) even in a warm saline bath. Gluckman et al., (1983) found the same result on running cold water through plastic tubing wound round unanaesthetised fetal lambs in utero. Breathing movements then occured during both high and low voltage electrocortical activity and were accompanied by electromyographic (EMG) activity in the peripheral muscles attributed to shivering. However, breathing movements (and the EMG activity) were arrested by mild isocapnic hypoxia. Raising the fetal Pa02 to neonatal values by an extra-corporeal membrane oxygenator (Blanco et al., 1987) did not alter episodic breathing movements, varying with electrocortical activity in fetal lambs; there was no change in umbilical blood flow, although pulmonary blood flow increased. But tying the umbilical cord in a variety of conditions, even in a warm saline bath (Johnson et al., 1973), results in the onset of maintained regular breathing movements. This could be due either to the removal of a placentally generated or transferred hormone (for example), or to the cessation of placental clearance (of heat or otherwise). Adamson (1986) observed that breathing movements, initiated by temporary umbilical cord occlusion in unanaesthetised fetal lambs in utero (the trachea being supplied with 02 enriched air), were arrested on the resumption of umbilical flow. Regular breathing can thus be stimulated in the absence

136

of most of the thermal and sensory changes observed at birth. Clearance of the lung liquid and gaseous flow through the tracheo-bronchial tree does not excite breathing movements. More experiments are required to identify the substance(s) which might be implicated. They are unlikely to include natural opiates, since naloxone does not modify the normal pattern of fetal breathing movements. And it should be remembered that a rise in fetal temperature is accompanied by panting.

It seems likely that the onset of continuous breathing postnatally is dependent on more than one mechanism, including cold exposure and arrest of umbilical blood flow. It is independent of the systemic arterial chemoreceptors, and occurs after mid-collicular section of the brain stem.

I hope that enough has been said to whet the appetite of other respiratory physiologists, and to indicate the opportunities offered by studies on unanaesthetized fetuses in utero. In these conditions we may observe the natural pattern of breathing movements under central control free from respiratory fluctuations in the blood gases. Sensory input from the systemic arterial chemoreceptors, the baroreceptors, or from the lungs, tracheobronchial tree and upper airways can be eliminated by appropriate nerve section. There seem to be many divergences from the classical theories of central respiratory control, too many for comfort.

We need a better stereotactic atlas of the fetal brain stem, pons and medulla than that currently available, which was designed for access to the hypothalamus; localization in the areas of respiratory interest is still inexact. Comparisons between the neuro-anatomy of these areas in the sheep fetus, the rat and the cat are urgently required. There is a rich harvest to be gathered by direct attack on the central nervous control of fetal breathing movement~ are not discouraged by the failure of adult physiologists to unravel this complex mechanism. It would hardly be likely that a system so important to survival would be controlled by a few simple mechanisms, particularly when we take into account its remote phylogenetic development. I commend the project to young investigators who are not yet bound by the shackles of classical adult respiratory theory.

REFERENCE

Adamson, S.L., 1986, Respiratory control mechanisms before and after birth, IUPS Satellite Symposium on Fetal Physiology - Cellular and Systems Approach, Vancouver Island, Canada, 33.

Bamford, O.S., Dawes, G.S., Denny, R., and Ward, R.A., 1986, Effects of the ~2-adrenergic agonist clonidine and its antagonist idazoxan on the fetal lamb, ~ Physiol., 381: 29.

Bamford, O.S., Dawes, G.S., Hanson, M.A., and Ward, R.A., 1986, The effects of doxapram on breathing, heart rate and blood pressure in fetal lambs, Resp. Physiol., 66: 387.

Bamford, O.S., Dawes, G.S., and Ward, R.A., 1986, Effects of apomorphine and haloperidol in fetal lambs, ~ Physiol., 377: 37.

Barcroft, J., 1946, Researches on pre-natal life, Oxford: Blackwell Scientific Publications.

Blanco, C.E., Bamford, 0., Hawkings, R., and Chen, V., 1987, Responses in the chronically instrumented fetal sheep to changes in blood gases and temperature produced by an extracorporeal membrane oxygen system, Proc. Soc. Study Fetal Physiol. 14th Annual Meeting: Groningen, 23.

Blanco, C.F., Dawes, G.S., Hanson, M.A., and McCooke, H.B., 1984, The response to hypoxia of arterial chemoreceptors in fetal sheep and new-born lambs,~ Physiol., 351: 25.

137

Cameron, I.R., and Semple, S.J.G., 1968, The central respiratory stimulant action of salicylates, Clin. Sci., 35: 391.

Dawes, G.S., 1968, Foetal and Neonatal Physiology. Year Book Medical Publishers, Chicago.

Dawes, G.S., 1984, The central control of fetal breathing and skeletal muscle movements, ~ Physiol., 346: 1.

Dawes, G.S., 1986, The central nervous control of fetal behaviour, Eur. ~ Obstet. Gynaecol. Rep. Biol., 21: 341.

Dawes, G.S., 1988, Hormonal control of fetal breathing, The Endocrine Control of the Fetus. Kunzel, W., and Jensen, A., eds., SpringerVerlag, Berlin, (in press).

Dawes, G.S., Gardner, W.N., Johnston, B.M., and Walker, D.W., 1980, Breathing patterns in fetal lambs after midbrain transection, ~ Physiol., 308: 29P.

Dawes, G.S., Gardner, W.N., Johnston, B.M., and Walker, D •. W. 1983, Breathing in fetal lambs: the effects of brain stem section, ~ Physiol., 335: 535.

Dawes, G.S., and Robinson, J.S., 1976, Rhytmic phenomena in prenatal life, Prog. Brain Res., 45: 383.

De Vries, J.I.P., 1987, Development of specific movement patterns in the human fetus, Doctoral Thesis: University of Groningen, Chapter 8.

Gluckman, P.D., Gunn, T.R., and Johnston, B.M., 1983, The effect of cooling on breathing and shivering in unanaesthetised fetal lambs in utero, ~ Physiol., 343: 495.

Gluckman, P.D., and Johnston, B.M., 1987, Lesions in the upper lateral pons abolish the hypoxic depression of breathing in unanaesthetised fetal lambs in utero, J.Physiol., 382: 373.

Heyman, M.A., and Rudolph, A.M., 1976, Effects of acetylsalicylic acid on the ductus arteriosus and circulation of fetal lambs in utero, Circ. Res.,38: 418.

Hohimer, A.~Bissonette, J.M., Machida, C.M., and Horowitz, B. 1985, The effect of carbonic anhydrase inhibition on breathing movements and electrocortical activity in fetal sheep, Resp. Physiol., 61: 327.

Hohimer, A.R., Bissonette, J.M., Richardson, B.S., and Machida, C.M., 1983, Central chemical regulation of breathing movements in fetal lambs, Resp. Physiol., 52: 88.

Jansen, A.H., Ioffe, S., and Chernick, V., 1987, Breathing activity in peripheral and central chemoreceptor denervated fetal sheep. 14th Annual Meeting Society Study Fetal Physiol., Groningen, 3.

Jensen, A., Bamford, O.S., Dawes, G.S., Hofmeyr, G., and Parkes, M.J., 1986, Changes in organ blood flow between high and low voltage electrocortical activity in fetal sheep,~ De~ Physiol., 8: 187.

Jensen, A., Hofmeyr, O.S., Dawes, G.S., Hofmeyr, G.J., and Parkes, M.J., 1985, Changes in organ blood flow between high and low voltage electrocortical activity and during isocapnic hypoxia in intact and brain stem transected fetal lambs, in: "The Physiological development of the Fetus and Newborn", Jones, C.T., and Nathaniels, P.W., eds, Academic Press, London.

Johnson, P., Robinson, J.S., and Salisbury, D., 1973, The onset and control of breathing after birth, Foetal and Neonatal Physiology: Sir J. Barcroft Symposium, Cambridge: Cambridge University Press. 217.

Johnston, B.M., and Gluckman, P.D., 1983, GABA-mediated inhibition of breathing in the late gestation sheep fetus, ~ Dev. Physiol., 5: 353.

Johnston, B.M., Gluckman, P.D., and Parsons, Y., 1985, The effects of midbrain lesions on breathing in fetal lambs, in: "Physiological Development of Fetus and Newborn", Jones, C.T., and Nathanielsz, P.W., eds., Academic Press, London.

138

Kitterman, J.A., Liggins, G.C., Clements, J.A., and Tooley, W.H., 1979, Stimulation of breathing movements in fetal sheep by inhibitors of prostaglandin synthesis, ~ dev. Physiol., 1: 453.

Koos, B.J., 1985, Central stimulation of breathing movements in fetal lambs by prostaglandin synthetase inhibitors,~ Physiol., 362: 455.

Millhorn, D.E., and Eldridge, F.L., 1986, Role of ventrolateral medulla in regulation of respiratory and cardiovascular systems, ~ ~ Physiol., 61: 1249.

Moore, P.J., Nijhuis, J.G., Hanson, M.A., and Parker, M.J., 1988, The effect of pilocarpine on breathing movements in the chemodenervated fetal lamb in utero, in: "Fetal and Neonatal Development", Jones, C.T., ed, Perinatology Press, Ithaca, (in press).

Parkes, M.J., Bamford, O.S., Dawes, G.S., Gianopoulos, J.G., and Quail, A.W., 1984, The effects of removal of the cerebellum, brain stem transection and discrete brain stem lesions in fetal lambs. Proc. Soc. Study Fetal Physiol., Oxford, 8.

Patrick, J., Natale, R., and Richardson, B., 1978, Patterns of human fetal breathing activity at 34 to 35 weeks gestational age, Am. ~ Obstet. Gynaecol., 13: 507.

Snyder, F.F., and Rosenfeld, M., 1937, Direct observation of intrauterine respiratory movements of the fetus and the role of carbon dioxide and oxygen in their regulation, Am.~ Physiol., 119: 153.

Von Ahlfeld, F., 1888, Uber bisher noch nicht beschriebene intrauterine Bewegungen des Kindes. In Verhandlungen der deutschen Gesselschaft der Gynakologie, Breitkopf und Hartel, Leipzig.

Walker, D.W., and Harding, R., 1984, Effect of hypoxia on the excitability of two cranial reflexes in unanaesthetised fetal sheep, ~ Dev. Physiol., 6: 387.

Windle, W.F., 1940, Physiology of the Fetus. Philadelphia: Saunders.

139

PATHOPHYSIOLOGY OF THE SUDDEN INFANT DEATH SYNDROME

Cl. Gaultier

Laboratory of Physiology and CNRS UA 1159, Hospital Antoine Beclere, Clamart, France

For a number of years clinicians and investigators have had an active interest in the sudden infant death syndrome (SIDS). Despite ever-increasing new physiological and epidemiological data, however, pathophysiology of SIDS has yet to be clearly elucidated.

EPIDEMIOLOGY

The definition of SIDS is the "sudden death of any infant which is unexplained by history and in which a thorough postmortem examination fails to demonstrate an adequate cause of death" (Consensus statement ... , 1987). It is extremely important that a complete death scene investigation be performed. The label "SIDS" probably covers a variety of causes of death. For example Basset al., (1986) demonstrated that deaths considered as SIDS, were in fact clearly explainable by accidental factors.

The incidence of SIDS in the general population is around 2 per 1000 live births (Shannon and Kelly, 1982a). In certain infant categories the incidence of SIDS may be greater than in the general population. These categories concern sibling and twin of SIDS victims, premature infants, and infants with intra-uterine drug exposure (Shannon and Kelly, 1982a). Male infants are at slightly increased risk than females. A racial predisposition has been suggested: Black and Indian infants are at increased risk, in contrast Asian infants have a low risk (Shannon and Kelly, 1982a).

Environmental factors at the time of death have been incriminated (Shannon and Kelly, 1982a). A greater incidence of SIDS has been reported during the winter and thus upper airway infections and hyperthermia, favored by inapropriate home heating, have been suggested as possible triggering factors for SIDS. Associated disorders such as viral infection at the time of death have been also reported. Postmortem examinations have shown in some SIDS victims lesions of esophagitis (Imbert et al., 1987), thus raising the question of the role of a gastroesophageal reflux. Psychological stress preceding death, such as separation from the mother, has been also suspected as a triggering event. However, none of these various factors can be considered as a

141

clear explanation of SIDS. But all of them may be considered as additional factors which may trigger death, perhaps in predisposed infants.

The most striking and homogenous epidemiological data concern the peak incidence for SIDS between 2 and 4 months of age (Shannon and Kelly, 1982a). SIDS is rare in the neonatal period and after 6 months of age. In the present review, factors of vulnerability at the time of the SIDS peak incidence in sleep organization and cardiorespiratory control are discussed.

PATHOPHYSIOLOGY

The results of two decades of investigations on SIDS pathophysiology are still inconclusive. The most compelling hypothesis is that SIDS victims have brain stem abnormalities in cardiorespiratory control. But despite the pertinence of such a hypothesis, it is important to point out that there is still no test of cardiorespiratory control which is specific and sensitive enough to detect SIDS candidates either in the general population or in high risk infants.

There are two possible ways to investigate brain stem abnormalities in cardiorespiratory control. The first is to better understand the developmental pattern of sleep organization and cardiorespiratory control in normal infants, especially between 2 and 4 months of age, and to determine the factors that can elicit apnea and/or bradycardia. The second is to investigate infants who experienced an aborted SIDS, near-miss SIDS infants, as well as the siblings of SIDS victims. However near-miss SIDS infants are presumably a heterogenous group and data observed in such infants provide only possible indications for predisposion to SIDS.

Sleep Organization

In newborns, active sleep (or REM sleep) occupies nearly 50% of the total sleep time. From birth to 6 months the relative proportion of active sleep decreases whereas the relative proportion of quiet sleep (or NREM sleep) increases. From 3 weeks to 6 months quiet sleep modifications occur with the appearance of sleep spindles and differenciation of stage 2 and 3-4 (Coons and Guilleminault, 1982). Intervening wakefulness and quiet sleep changes appear simultaneously with the emergency of the sleep-wakefulness circadian rhythm (Navelet et al., 1982). It is intriguing to see that the critical period of change in sleep stage distribution coincides with the peak incidence of SIDS. In near-miss SIDS infants authors have looked at any dysregulation in sleep organization. Haddad et al. (1981) have shown that with advancing age near-miss SIDS infants have less quiet sleep and more active sleep than normals. This maturational abnormality or delay in sleep state distribution may be an argument for brain stem dysregulation predisposing for SIDS. Since active sleep is well known to favour respiratory and cardiac disturbances, more active sleep time may be a predisposing factor.

Respiratory Control

In normal sleeping infants, apneas of less than 10 s are common, especially during active sleep (Gaultier, 1985). In full term infants most apneas are central. With advancing postnatal age the total number of apneas decreases (Gaultier, 1985). Guilleminault et al. (1979) observed in normal infants a period of increase in obstructive apneas at around 6 weeks of age.

142

Several factors have been shown to elicit apnea: upper airway reflexes via pharyngeal (Fisher et al., 1985), laryngeal (Dowing and Lee, 1975; Boggs and Bartlett, 1982) afferences, increase in body temperature in healthy infants which increases the respiratory instability during sleep (Berterottiere et al., 1987 in the press) ; short time sleep deprivation favoring obstructive apneas (Canet et al., 1987).

Numerous respiratory abnormalities during sleep have been reported in near-miss SIDS infants. But none of them have sufficient specificity and sensitivity for the general screening of infants at risk for SIDS. Furthermore, respiratory abnormalities have been observed only in some of the near-miss SIDS infants. Among the respiratory abnormalities observed were increased frequency of short apnea (Southall et al., 1986), prolonged apnea (Steinschneider, 1977), excessive periodic breathing (Kelly and Shannon, 1979), and abnormal number of obstructive apnea (OA) (Guilleminault et al., 1979). Increased number of OA points out the interest of investigating upper airway patency in risk categories for SIDS. A possible relationship between SIDS and obstructive sleep apnea syndrome (OSAS) has been suggested. Some near-miss SIDS infants have developed OSAS (Guilleminault et al., 1984). In the same families association of SIDS, OSAS in children and adults, and narrowing of the upper airway in family members have been reported (Guilleminault et al., 1986).

Recently using spectral analysis of the respiratory frequency (RF), Gordon (Gordon et al., 1984) reported in near-miss SIDS infants a larger dispersion of FR than in control age-matched infants suggesting greater respiratory instability.

Several investigators have also studied chemical ventilatory control during sleep in near-miss SIDS infants. In some, diminished ventilatory sensitivity to hypercapnia and hypoxia has been observed (Hunt et al., 1981; Lindenberg and Newcomb, 1986). Arousal responsiveness to hypercapnia and hypoxia appear to be impaired in these infants (McCulloch et al., 1982; Vander Hal et al., 1985). The alveolar partial pressure of C02 (PAC02) during arousal was found to be significantly higher in near-miss SIDS infants (55 mmHg) than in controls (48 mmHg) (McCulloch et al., 1982). However a great overlaping of PAC02 values inducing arousal was observed between near-miss SIDS infants and controls (McCulloch et al., 1982). The arousal responsiveness to hypoxia was a more discriminative parameter: in all but one of the 11 near-miss SIDS infants studied, no hypoxic arousal was observed (McCulloch et al., 1982).

These data on respiratory control during sleep support the hypothesis of brain stem dysfunction (Hunt and Brouillette, 1987). Arousal deficit especially to hypoxia may be a factor necessary for SIDS. Arousal deficit induces failure to initiate inspiration and prolonged apnea. Other factors such as respiratory pattern abnormalities andjor chemoreceptor dysfunction may also contribute to SIDS risk.

Cardiac Control

At birth the two components of the autonomic cardiac control, the parasympathetic and the sympathetic, are not equally developed. Vagal innervation is believed to be complete, whereas sympathetic innervation appears immature (Schwartz, 1976). Furthermore there is an imbalance in the right and left sympathetic neural pathways (Schwartz, 1976). Newborn and infant heart rate (HR) appears to be under vagal dominance and susceptible to vagal stimuli.

Numerous factors elicit bradycardia in normal infants: apnea with

143

hypoxemia (Henderson-Smart et al., 1986), trigeminal afferences (Allen et al., 1979), stimuli originating from the esophagus (Schey et al., 1981) and the gastrointestinal tract (Vallbona et al., 1963). An interesting proposed mechanism to explain the occurrence of bradycardia is the fear paralysis reflex (FPR) (Kaada, 1987). The FPR is elicited by stimulus perceived as dangerous: unfamiliar noises, restraints of movements, separation fear. The FPR response includes motor paralysis, respiratory arrest and bradycardia. In animals the FRP is absent at birth and develops during early infancy. Thus in human infants the onset of the FPR is perhaps close to the peak incidence of SIDS. Several factors such as visceral stimuli, hypoxemia, smoke, etc ... , may lower the threshold of the FPR. Thus, the FPR may be a triggering mechanism for SIDS. However further investigations in human infants are necessary to confirm this hypothetical mechanism for SIDS.

In near-miss SIDS infants, dysfunction of autonomic cardiac control has been observed. Some results have suggested increased vagal reactivity (Kahn et al., 1983). Increased vagal reactivity can be tested by the response of a vagally mediated reflex, the oculocardiac reflex. Ocular compression induces slowing of the HR. In normal preterm infants the response of this reflex, i.e. the duration of the provoked asystole, decreases with postconceptional age (Ramet et al., 1987). Some near-miss SIDS infants have much longer provoked asystole than the upper limit of values for age-matched control infants, suggesting increased cardiac vagal reactivity (Kahn et al., 1983).

However, other data suggest increased sympathetic activity in cardiac control. In near-miss SIDS infants, Leisstner et al. (1980) reported a significantly shorter R-R interval and a decreased overall HR variability in these infants than in control age- matched infants. Looking at 24-H EKG recordings in SIDS victims just before their death with the use of a spectral analysis of HR, Gordon et al. (1984) observed greater power in the low frequencies (0.02 to 0.1 Hz) in near-miss SIDS infants compared to controls, whereas no significant difference was observed in the high frequencies (0.4 Hz). This increase in low frequency power suggests a a-adrenergic dominance in cardiac control. TI1us, greater low-frequency power of HR could be a marker of dysfunction of cardiac control. But, unfortunately, this marker failed to be sensitive and specific enough to detect subsequent SIDS victims in a large population (Gordon et al., 1986). The reason for such an increase in sympathetic activity is still unknown. It may be secondary to repeated hypoxemia before death. Accordingly, in some SIDS victims smooth muscle hyperplasia of pulmonary circulation and astrogliosis of the brain stem have been identified (Shannon and Kelly, 1982a,b), suggesting repeated hypoxemia. However, no direct proof of this is available.

SUMMARY

In 1987 the cause(s) of SIDS is still unknown. The label SIDS possibly covers a variety of causes of death. A complete death scene investigation and postmortem examination are required. Many of the deaths, considered to be due to SIDS, are possibly due to accidental factors or recognized pathologies.

Brain stem abnormalities in cardiorespiratory control is the most compelling present hypothesis, despite the fact that such abnormalities have only been observed in some near-miss SIDS infants. In addition to this possible predisposing situation, an occasional triggering event may be present at the time of death such as hyperthermia, upper airway

144

infection, stimuli eliciting the FPR, and gastroesophageal reflux. Despite the hypothesis of brain stem abnormalities, it is important to remember that SIDS can occur in apparently healthy full term infants without any familial cardiorespiratory history, and without any pathology or any abnormal environmental factor at the time of death. Thus the question of the unknown cause of vulnerability for SIDS, that is especially high during the period of peak incidence between 2 and 4 months of age, remains unresolved.

Now, no prospective assessment of cardiac and/or respiratory control has sufficient sensitivity and specificity to identify infants destined to die from SIDS. In asymptomatic infants at risk for SIDS, the controversy regarding home monitoring remains unresolved (Hunt and Brouillette, 1987; Robin, 1987; Ariagno and Glotzbach, 1987). Educational programs for parents and prevention of the suspected triggering events for death may be the most efficient attitude (Ariagno and Glotzbach, 1987).

REFERENCES

Allen, L.G., Howard, G., Smith, L.B., Me Cubbin, J.A., and Weaver, R.L., 1979, Infant heart rate response to trigeminal airstream stimulation: determination of normal and deviant values, Pediatr. Res., 13: 184.

Ariagno, R.L. and Glotzbach, S.F., 1987, Home monitoring of high-risk infants Chest, 91: 898.

Bass, M., Kravath, R.E., and Glass, L., 1986, Death-scene investigation in sudden infant death, ~Engl. ~ Med., 315: 100.

Berterottiere, D., D'Allest, A.M., Nedelco, H., Dehan, M., and Gaultier, Cl., 1987, Effects of increase in body temperature on breathing pattern during sleep in healthy infants, Pediatr. Pulmonol., 3: in press.

Boggs, D.F., and Bartlett, D., 1982, Chemical specificity of a laryngeal apneic reflex in puppies, ~ ~ Physiol., 53: 455.

Canet, E., Gaultier, Cl., Dehan, M., and D'Allest, A.M., 1987, Effect of sleep deprivation on respiratory adaptation during sleep in healthy infants, Fed. Proc., 46: 502.

Consensus statement National Institute of Health Consensus development conference on infantile apnea and home monitoring. Sept 29 to Oct 1 1986, 1987, Pediatrics, 79: 292.

Coons, S., and Guilleminault, C., 1982, Development of sleep wake patterns and non-rapid eye movements sleep stages during the first six months of life in normal infants, Pediatrics, 69: 793.

Dowing, S.E., and Lee, J.C., 1975, Laryngeal chemosensitivity: a possible mechanism for sudden infant death, Pediatrics, 55: 640.

Fisher, J.T., Mathew, W.P., Sant'Ambrogio, F.B., and Sant'Ambriogio, G., 1985, Reflex effects and receptor responses to upper airway pressure and flow stimuli in developing puppies,~~ Physiol., 58: 258.

Gaultier, Cl., 1985, Breathing and sleep during growth: physiology and pathology, Bull. Eur. Physiopathol. Respir., 21: 55.

Gordon, D., Cohen, R., Kelly, D., Axelrod, S., and Shannon, D., 1984, Sudden infant death syndrome abnormalities in shortterm fluctuations in heart rate and respiratory activity, Pediatr. Res. 18: 921.

Gordon, D., Southall, D.P., Kelly, D.H., Wilson, A., Akselrod, S., Richards, J., Kenet, B., Kenet, R., Cohen, R., and Shannon, D., 1986, Analysis of heart rate and respiratory patterns in sudden infant death syndrome victims and control infants, Pediatr. Res., 20: 680.

145

Guilleminault, C., Ariagno, D., Korobkin, R., Nagel, L., Baldwin, R., Coons, S., and Owen, M., 1979, Mixed and obstructive sleep apnea and near miss for sudden infant death syndrome: 2. Comparison of near miss and normal control infants by age, Pediatrics, 64: 882.

Guilleminault, C., Powell, N., Heldt, G., and Riley, R., 1986, Small upper airway in near-miss sudden infant death syndrome infants and their families, Lancet, 1: 402.

Guilleminault, C., Souquet, M., Ariagno, R.C., Korobkin, R., and Simmons, F.B., 1984, Five cases of near-miss sudden infant death syndrome, Pediatrics, 73: 71.

Haddad, G.G., Walsh, E.M., Lestner, H.L., Grodin, W.K., and Mellins, R.B., 1981, Abnormal maturation of sleep states in infants with aborted sudden infant death syndrome, Pediatr. Res., 15: 1055.

Henderson-Smart, D.J., Butcher-Puech, M.C., and Edwards, D.A., 1986, Incidence and mechanism of bradycardia during apnoea in preterm infants, Arch. Dis. Child., 61: 227.

Hunt, C.E., and Brouillette, R.T., 1987, Sudden infant death syndrome: 1987 perspective, ~ Pediatr., 110: 669.

Hunt, C.E., Me Culloch, K., and Brouillette, R.T., 1981, Diminished hypoxic ventilatory response in near-miss sudden infant death syndrome,~~ Physiol., 50: 1313.

Imbert, M.C., Roset, F., Guyot, H., Dehan, M., Benisvy, C., and Gautier, J.P., 1987, Presented au Club de pathologie pediatrique. Journees parisiennes de Pediatrie, Paris.

Kaada, B., 1987, Sudden infant death syndrome, Norwegian University Pr·ess, Oslo, pp 1-56.

Kahn, A., Riazi, J., and Blum, D., 1983, Oculocardiac reflex in Near Miss for sudden infant death syndrome infants, Pediatrics, 71: 49.

Kelly, D.H., and Shannon, D.C., 1979, Periodic breathing in infants with near-miss sudden infant death syndrome, Pediatrics, 63: 355.

Leisstner, H.L., Haddad, G.G., Epstein, R.A., and Mellins, M.A., 1980, Heart rate and heart rate variability during sleep in aborted sudden infant death syndrome, ~ Pediatr., 97: 51.

Lindenberg, J.A., and Newcomb, J.D., 1986, Hypercapnea and hypoxia challenge tests in infants at risk for sudden death syndrome, Am. J. Dis. Child., 140: 466.

Me Culloch, K., Brouillette, R.T., Guzzeta, A.J., and Hunt, C.E., 1982, Arousal response in near-miss sudden infant death syndrome and in normal infants,~ Pediatr., 101: 911.

Navelet, Y., Benoit, 0., and Bouard, G., 1982, Nocturnal sleep organization during the first month of life, Electroencephalogr. Clin. Neurophysiol., 54: 71.

Ramet, J., Gaultier, Cl., Praud, J.P., Magny, J.F., and Dehan, M., 1987, Relationship between vagally-mediated cardiac and respiratory reflexes in pret·erm infants, Am. Rev. Respir. Dis., 135: A 174.

Robin, E.D., 1987, SIDS and home monitoring, Chest, 91: 765. Schey, W.L., Meus, P., Levinsky, R.A., Campbell, C., and

Reproglie, R., 1981, Radiology, 140: 73. Schwartz, P.J., 1976, Cardiac sympathetic innervation and the sudden

infant death syndrome, Am. ~ Cardiol., 60: 167. Shannon, D.C., and Kelly, D.H., 1982a, SIDS and Near Miss SIDS (First

two parts),~ Engl.~ Med., 306: 959. Shannon, D.C., and Kelly, D.H., 1982b, SIDS and Near Miss SIDS (second of

two parts)~ Engl. ~ Med., 306: 1022. Southall, D.P., Richards, J.P., Stebbens, V., Wilson, A.J., Taylor, V.,

and Alexander J.R., 1986, Cardiorespiratory function in 16 full-term infants with sudden infant death syndrome, Pediatrics, 78: 707.

Steinschneider, A., 1977, Prolonged apnea and respiratory instability: a discriminative study, Pediatrics, 59: 962.

146

Vallbona, C., Desmond, M.M., Rudolph, A.J., Pap, L.F., Hill, R.M., Franklin, R.R., and Rush J.B., 1963, Cardiodynamic studies in the newborn II. Regulation of the heart rate, Bioi. Neonat. , 5: 159.

Vander Hal, A.L., Rodriguez, A.M., Sergent, C.W., Platzker, A.C., and Keens, T.G., 1985, Hypoxic and hypercapneic arousal responses and prediction of subsequent apnea in apnea of infancy, Pediatrics, 75: 848.

147

DEPRESSION OF DURING CARBON GLUTAMIC ACID

EXCITATORY AMINO DIOXIDE NARCOSIS, AND ASPARTIC ACID

ACID NEUROTRANSMITTERS IN ANESTHESIA, AND HYPOXIA:

BRAIN

Robert E. Dutton, Paul J. Feustel, Elizabeth H. Dutton, Anthony Szema, Vivian E. Shih, Paolo M. Renzi, and Gildo D. Renzi

Albany Medical College, Albany, NY, USA; Rensselaer Polytechnic Institute, Troy, New York, Massachusettss General Hospital, Boston, MA, USA and Notre Dame Hospital Montreal, Canada

Application of the excitatory amino acid neurotransmitters aspartic acid and glutamic acid to respiratory center neurons in the medulla causes an increase in neuron output (Toleikis et al., 1979). When neurotransmitter amino acids are delivered to the surface of the medulla by means of ventriculocisternal perfusion or application directly to the ventral surface, glutamic acid increases ventilation and gammaaminobutyric acid (GABA) depresses ventilation (Hoop et al., 1985; Systrom et al., 1985; Gatti et al., 1985). The present investigation was undertaken to determine whether hypercapnia at levels associated with C02 narcosis in man is accompanied by changes in the concentration of endogenous amino acid neurotransmitters in dorsal and ventral medulla of the brain. The aim of this study is to provide support for the hypothesis that reduction of glucose-derived endogenous excitatory amino acid neurotransmitters and elevation of depressant amino acid neurotransmitters contribute to impairment and even depression of the ventilatory response to carbon dioxide at higher inhaled C02 concentrations (Eisele et al., 1967).

The influence of hypoxia on changes induced by hypercapnia was also evaluated. A further goal of the study was to determine whether either plasma or CSF amino acid levels might serve as a useful index of amino acid neurotransmitter levels in the brain.

METHODS

Mongrel dogs were anesthetized with sodium pentobarbital (25-35 mgjkg). Initial surgical procedures consisted of insertion of femoral artery and vein catheters, tracheostomy, preparation of the lumbar spine for sampling lumbar cerebrospinal fluid and preparation of the calvarium for craniotomy. Following the surgical preparation, either a combination of hypercapnia-hyperoxia (high co2-high 02) or hypercapnia-hypoxia (high C02-low 02) was induced for a period of one to two hours. To achieve a more precise control of inspired gas mixtures, the high C02-low 02 animals were paralyzed with pancuronium Br (0.06 mgjkg) and were artficially ventilated. Thirteen spontaneously breathing high C02-high 02

149

dogs received inspired C02 (balance 02) that elevated PaC02 to 112-140 (mean 127) Torr and Pa02 to 240-372 (mean 309) Torr (7.5 Torr= 1 kPa). Mean pH was 6.93 ± 0.02. Ten mechanically ventilated, paralyzed high C02-low 02 dogs received mixtures of inspired C02, nitrogen ru1d air sufficient to raise PaC02 to 108-153 (mean 135) Torr and lower Pa02 to 30-50 (mean 41) Torr. Mean pH was 6.90 ± 0.02. Eleven control animals spontaneously breathed or were mechanically ventilated with room air to maintain normal blood gases for one to two hours. Mean PaC02 was 37 ± 1 Torr, mean Pa02 was 100 ± 3 Torr, and mean pH was 7.377 ± 0.002. Upon completion of the test period, arterial blood was obtained for analysis of blood gases, pH, and plasma amino acids. Lumbar cerebrospinal fluid was obtained by lumbar puncture for analysis of amino acids. Medullary brain tissue was obtained by dorsal craniotomy, and the medulla was divided into dorsal and ventral segments. Plasma and cerebrospinal fluid were prepared for amino acid analysis by deproteinization with 40 mgjml of sulfosalicylic acid. The samples were centrifuged at 10,000 g for 15 min at 4°C, and supernatant was frozen at -70°C until analyzed. Dorsal and ventral medulla tissue was homogenized with 9 ml of 3% sulfosalicylic acidjgram of brain tissue for 5 min using a Brinkman blender prechilled to 4°C, and was frozen at -70°C until analyzed. All samples were analyzed for amino acid content by ion exchange liquid chromatography in a Beckman 119CL or 6300 Amino Acid Analyzer. The samples were brought to pH 2.2 with saturated lithium hydroxide solution and were filtred through 0.2 ~m disposable filtres before the filtrate was placed in the amino acid analyzer.

Student's unpaired and paired t-tests were used for statistical analyses with significance accepted at the 95% confidence level (p<.05).

RESULTS

During carbon dioxide narcosis (anesthesia) the glucose-derived excitatory amino acid neurotransmitters, glutamic acid and aspartic acid, were significantly decreased to less than 55% of control values in dorsal medulla during both high C02-high 02 and high C02-low 02 (p<.05) (Fig.1). In ventral medulla, glutamic acid (GLU) and aspartic acid (ASP) levels were also significantly lower than control in both high C02 groups (p<.02). (Ventral GLU control = 5.49 ± 0.41 mMjkg, ventral GLU high C02-high 02 = 3.46 ± 0.26 mMjkg, and ventral GLU high C02-low 02 = 3.35 ± 0.21 mMjkg. Ventral ASP control = 2.15 ± 0.22 mMjkg, ventral ASP high C02-high 02 = 1.19 ± 0.19 mMjkg, and ventral ASP high C02-low 02 = 1.42 ± 0.42 mMjkg). There was a highly significant correlation between GLU and ASP in controls and high C02-low 02 animals (dorsal medulla r=.95, ventral medulla r=.88). However, there were no differences between the two high C02 groups in either dorsal or ventral medulla that would suggest a significant influence of hypoxia on GLU or ASP levels during hypercapnia.

In contrast to a highly significant reduction in both GLU and ASP transmitters during hypercapnia whether or not hypoxia was present, the depressant neurotransmitter GABA was diminished significantly from control in both dorsal and ventral medulla only during high C02-high 02 (p<.02) (Fig.1). These differences between GABA at high C02-high 02 and GABA at high C02-low 02 were highly significant (p<.OOl). (Ventral GABA control = 0.83 ± 0.13 mMjkg, high C02-high 02 = 0.44 ± 0.04 mMjkg, and high C02-low 02 = 0.89 ± 0.08 mMjkg). There were no significant differences in the depressant neurotransmitter, glycine, between control and either high C02 group, or between high C02-high 02 and high C02-low 02 at the levels of hypercapnia induced in this study. There were also no significant differences from control in the depressant neuroregulatory

150

8

CONTROL HIGH C02 HIGH C02 LOW 02


8

4


oL--L--~~--~~--~ CONTROL HIGH C02 HIGH C02

LOW 02

Fig. 1. Amino Acid Concentrations in Dorsal Medulla (mM/kg wet brain weight).

amino acid, taurine, during high C02. However, a tendency for taurine to rise during high C02-high 02 and fall during high C02-low 02 in both ventral and dorsal medulla led to significant differences between the two high C02 states (p<.02) (Fig.l). (Ventral taurine control = 5.14 ± 1.05 mM/kg, high C02-high 02 = 6.69 ± 1.13 mMjkg, and high C02-low 02 = 2.71 ± 0.63 mM/kg). Glutamine was not significantly changed from control at the levels of C02 administered in these studies.

Lower transmitter amino acid concentrations were present in ventral medulla than in dorsal medulla, except for taurine which showed no difference. When all experimental groups are combined for purposes of this analysis (n=30), dorsal minus ventral differences by paired t-test are as follows: GLU 1.04 ± 0.07 mM/kg (p<.05), ASP 0.64 ± 0.04 mM/kg (p<.01), GABA 0.87 ± 0.03 mM/kg (p<.001), and glycine 1.02 ± 0.03 mM/kg (p<.001). Glutamine was also lower in ventral medulla by 0.86 ± 0.05 mM/kg (p<.01) .

. Analysis of the correlation between arterial blood gases, plasma

amino acids, and medulla amino acids for all experimental groups combined (n=28) yielded the following results: The highest and most consistent correlation with brain GLU and ASP was PaC02 (p<.001). Dorsal GLU (r=.67), dorsal ASP (r=.62), ventral GLU (r=.75) and ventral ASP (r=.63) all decreased as PaC02 was raised. The essential amino acids, lysine, histidine, and arginine rose in plasma as brain GLU and ASP decreased. Correlations for plasma lysine versus GLU and ASP ranged from .47 to .56 (p<.01). Plasma histidine versus GLU and ASP ranged from r=.56 (p<.01) to r=.63 (p<.001). Plasma arginine had a higher correlation with dorsal GLU (r=.70, p<,001) and ASP (r=.62, p<.001), but ventral correlations were lower (r=.42 and .40, p<.05).

1 51

In contrast to the excitatory amino acids, the inhibitory amino acids, GABA, glycine, and taurine, showed no correlation with PaC02. Brain GABA and taurine did have low level but significant correlations with Pa02. Brain GABA increased as Pa02 was lowered (dorsal GABA r=.37, p<.05, and ventral GABA r=.52, p<.01), whereas brain taurine decreased as Pa02 was lowered (dorsal taurine r=.50, p<.01 and ventral taurine r=.47, p<.01). Brain GABA also increased as plasma GLU decreased (dorsal GABA r=.43, p<.05 and ventral GABA r=.57, p<.001); whereas, brain taurine decreased as plasma GLU decreased (dorsal taurine r=.60, p<.001, and ventral taurine r=.60, p<.001).

The cerebrospinal fluid (CSF) amino acids isoleucine and leucine rose as brain glutamic acid and aspartic acid decreased during C02 inhalation (n=24). The correlations are as follows: CSF isoleucine versus dorsal GLU r=.54 (p<.01), dorsal ASP r=.54 (p<.01), ventral GLU r=.48 (p<.02), and ventral ASP r=.42 (p<.05). CSF leucine showed a significant negative correlation with dorsal GLU (r=.54, p<.01) and with dorsal ASP (r=51, p<.02), but no with ventral GLU or ASP. CSF GLU also rose as brain GABA decreased during high C02-high 02 inhalation. CSF GLU versus dorsal GABA for all groups (n=24) was r=.58 (p<.01) and CSF GLU versus ventral GABA r=.46 (p<.02).

DISCUSSION

Failure of patients with chronic obstructive pulmonary disease to maintain normal ventilatory sensitivity to carbon dioxide may be caused by a reduction in excitatory neurotransmitters or an elevation of depressant neurotransmitters in the medulla or the brain. The results of the present study demonstrate a reduction in glucose-derived excitatory transmitters glutamic acid and aspartic acid in the dorsal medulla containing the region of the respiratory motor neurons, and in the ventral medulla containing the region of the ventrolateral superficial medullary chemoreceptive tissue. These findings support the hypothesis that blunting of the ventilatory sensitivity to carbon dioxide, in terms of ventilation per Torr increase in PaC02 during progressively increasing levels of hypercapnia, is caused by: 1) a suppression of the rate at which output of respiratory center motor neurons increases at higher levels of PaC02 in the dorsal medulla; 2) a declining rate of increase of input to the respiratory center from the ventrolateral superficial medullary chemoreceptors; and 3) an eventual reduction in neural output from peak levels resulting in a decline in ventilation toward room air values.

Depression of the ventilatory response to C02 accompanied by depletion of the excitatory neurotransmitter glutamic acid in cerebrospinal fluid during buffered ammonia infusion in the dog (Berkman et al., 1978), and during glycine infusion in man (Renzetti et al., 1974), are also consistent with the above hypothesis. The reaction of brain glutamic acid with ammonia to form glutamine during hyperammonemic syndromes is a well recognized cause of depletion of brain glutamic acid (Engelsen, 1986).

The most consistent correlation of blood with brain excitatory amino acid transmitters was PaC02. The depressant transmitters GABA and taurine had correlations with Pa02 but not PaC02. An increase in GABA and decrease in taurine are similar to the findings of Hitzig et al. (1985) in semiaquatic turtles. CSF leucine and isoleucine increased as medullary GLU fell in the present study. However, none of the present correlations are sufficiently high to accurately predict brain GLU during a given experiment. This individual variability may be the basis for the variability of PaC02 at which C02 narcosis becomes evident in man.

152

REFERENCES

Berkman, R.A., Meyer, K.T., Rosenberg, A.G., and Dutton, R.E., 1978, Depressant effect of ammonia on the ventilatory response to hypoxia and hypercapnia, in: "The Regulation of Respiration during Sleep and Anesthesia", R.S. Fitzgerald, H.Gautier, S. Lahiri, eds., Plenum Press, New York.

Eisele, J.H., Eger II, E.I., and Muallem, M., 1967, Narcotic properties of carbon dioxide in the dog, Anestehesiology, 28: 856.

Engelsen, B., 1986, Neurotransmitter glutamate: its clinical importance. Acta Neurol. Scand., 74: 337.

Gatti~J., Taveira da Silva, A.M., Hamosh, P., and Gillis, R.A., 1985, Cardiorespiratory effects produced by application of L-glutamic and kainic acid to the ventral surface of the cat hindbrain, Brain Res., 330: 21.

Hitzig, B.M., Kneussl, M.P., Shih, V., Brandstetter, R.D., and Kazemi, H., 1985, Brain amino acid concentrations during diving and acidbase stress in turtles, ~ ~ Physiol., 58: 1751.

Hoop, B., Shih, V.E., and Kazemi, H., 1985, Relationship between central nervous system hydrogen ion regulation and amino acid metabolism in hypercapnia. II, Am. Rev. Resp. Dis., 132: 248.

Renzetti A.D. Jr., Harris, B.A., and Bowen, J.F., 1974, Influence of ammonia on respiration,~~ Physiol., 16: 703.

Systrom, D., Hoop, B., Pappagianopoulos, P., and Kazemi, H., 1985, Interaction between GABA and glutamate in central control of cardiorespiratory function, Am. Rev. Resp. Dis., 131: A291.

Toleikis, J.R., Wang, L., and Boyarsky, L.L., 1979, Effects of excitatory and inhibitory amino acids on phasic respiratory neurons,~ Neurosci. Res., 4: 225.

153

RESPIRATORY ACTIONS OF CHOLECYSTOKININ AND ITS INTERACTION WITH

OPIOIDS AT THE BRAINSTEM LEVEL

INTRODUCTION

Maria A. Hurle1, MariaM. Dierssen1, Marie-Pierre Morinsurun2 and Jesus Florez1

lDept. of Pharmacology, Faculty of Medicine, University of Cantabria. Nat. Hosp. "Valdecilla", Santander, Spain

2c.N.R.S., Lab. of Nervous Physiology, Dept. of Applied Neurophysiology, Gif-sur-Yvette, France

CCK-8S may have an important role in the regulation of respiratory function. Immunohistochemical and autoradiographic studies have shown the presence of CCK-8-like immunoreactive terminals and CCK-8 receptors in several respiration related nuclei, such as the nucleus tractus solitarii (NTS) and the nucleus ambiguus (NA) (Zarbin et al.,1983). Iontophoretic application of CCK-8S onto respiratory neurons in the NTS of cats, produced a decrease in their spontaneous activity and CCK-8S microinjected in the same nucleus depressed the phrenic activity in a reversible fashion (Denavit-Saubie et al., 1985). In contrast, the administration of the peptide into the brain ventricles of cats induced a potent stimulation of tidal volume (Gillis et al., 1983). The i.v. administration in decerebrated ·cats reduced the respiratory amplitude when the vagi were intact, but this effect was reversed to stimulation when they were sectioned (Hurle et al., 1985). In order to assess the ability of CCK-8S to interact selectively with the brainstem respiratory structures, we have applied the peptide to the ventral medullary and to the dorsa-rostral pontine surfaces of cats. The specificity of the effects has been tested by analyzing the antagonistic properties of proglumide, a CCK-8S antagonist.

On the other hand, several findings suggest that CCK-8S may function as an endogenous anti-opioid substance. Both proglumide and CCK-8S antibodies strongly potentiate several actions induced by opioids, whereas they are antagonized by CCK-8S (Faris, 1985). We have tested this interaction, by analyzing the respiratory effects of CCK-8S and metenkephalin under the influence of naloxone or proglumide, respectively.

MATERIAL AND METHODS

The experiments were carried out on lightly anesthetized cats. The ventral medullary surface and the dorsa-rostral pontine surface were exposed as previously described (Hurl~ et al., 1985). For application of

155

Table 1. Respiratory effects of CCK-8S (0.09 nMol) before and after pretreatment with naloxone (1 mg/kg i.v.)

CCK-8S CCK-8S after naloxone

FREQUENCY 0.5 ± 2.9 0.8 ± 2.8 TIDAL VOLUME 15.3 ± 1.8 12.0 ± 4.3 MINUTE VOLUME 13.8 ± 3.7 12.9 ± 6.1

drugs in the medulla, two plastic chambers (10 mm longitudinal axis and 2.5 mm transverse axis) were lowered bilaterally covering the areas M, S and L described as chemosensitive (Schlaefke, 1981). In the pons, a single oval chamber (8 mm longitudinal axis and 5 mm transverse axis) was lowered, the anterior border remaining just behind the inferior colliculi. The chambers were filled with different drugs to be tested, diluted in a constant volume of saline (80 ul).

Respiration was monitored by pneumotachography under continuous 100% 02 breathing. End-tidal C02 and blood pressure were recorded. Rectal temperature was maintained at 37°C.

RESULTS

Respiratory Actions of CCK-8S

CCK-8S (0.09 lli~Ol, 0.44 nMol and 0.88 nMol) was applied for 30 min to the medullary surface of 27 cats. The peptide induced an immediate and dose-dependent increase in tidal volume. Respiratory frequency was slightly increased, but the values did not reach statistical significance. The increase in minute volume strictly followed that of tidal volume, and end-expiratory C02 was concurrently decreased. The specific antagonist proglumide (10 ug) antagonized the stimulatory effect of CCK-8S; significant diferences (paired t-test) were observed at all times for all doses (Fig. 1). CCK-8S (0.88 nMol and 8.8 nMol) was applied to the pontine surface of 10 cats for 30 min. It did modify neither respiratory frequency nor tidal volume. Therefore, minute volume and end-exspiratory C02 remained unchanged.

Interaction between CCK-8S and Opioids

The opiate antagonist naloxone (1 mg(kg, i.v. in 20 min) did not modify the respiratory actions induced by the subsequent application of CCK-8S (0.09 nMol) in the medullary ventral surface of 5 cats (Table 1).

The CCK-8 antagonist proglumide (10 ug) did not modify the respiratory depression induced by the opioid peptide met-enkephalin (400 ug and 800 ug) applied onto the ventral medullary surface of 8 cats (Table 2).

DISCUSSION

CCK-8S topically applied to the ventral medulla elicits a potent and dose dependent stimulation of respiration mostly accounted for by an increase in tidal volume. This action appears to be related to specific

156

20

· 10

•50

· 10

Minute Volume ·70

Frequency

T1dol Volume

~

End - Expiratory C02

0

· 10

-20

· 30

· 40

-50 0~1 ~3~5~--~10~--~15~~

TIME ! m1n)

I I

0 1 I I

3 5 I I .~ 10 15

Fig. 1.

TIME (mm)

Time courses of the respiratory effects induced by saline (0) and CCK-8S 0.09 nMol (.&.), 0.44 nMo 1 ( •) and 0. 88 nMo 1 ( e) , applied to the medullary surface in cats . Symbols represent mean values + S.E. * p<0.05, ** p<0.01, *** p<0.001 -(Student's t-test).

30

CCK-receptors, since it was antagonized by proglumide. The results suggest a specific involvement of CCK-8S in the mechanisms controlling respiratory amplitude, which appear most ly restricted to the medullary level (Hurle et al., 1985). In the pons CCK-8S (9 nMol) has no ettect on the respiratory activity. Such a lack of effect is in agreement with the absence of receptors for CCK in the pontine respiration related regions, the nucleus parabrachialis medial is and the Kolliker-Fuse nucleus, as evidenced by autoradiographic studies (Zarbin et al., 1983).

In contrast to analgesic and behavioural tests (Faris, 1985), in our experimental model naloxone did not modify the respiratory effects induced by CCK-8S, and the depression induced by met-enkephalin was not modified by proglumide. Thus, our results suggest no interaction between cholecystokinergic and opioid systems in the respiratory controlling function.

157

Table 2. Respiratory effects of met-enkephalin in the ventral medullary surface before and after the pretreatment with proglumide (10 ~g)

MET-ENKEPHALIN 400 ~

Frequency Tidal volume

MET-ENKEPHALIN 800 ~

Frequency Tidal volume

Met-enkephalin

-3.7 + 2.2 -15.6 + 2.5

-5.2 + 2.4 -21.3 + 3.9

Met-enkephalin after proglumide

-2.8 ± 2.9 -13.0 ± 1.5

-4.7 ± 1.3 -17.3 ± 1.9

REFERENCES

Denavit-Saubie, M., Hurle, M.A., Morin-Surun, M.P., Foutz, A.S. and Champagnat, J., 1985, The effects of cholecystokinin-a in the nucleus tractus solitarius, in: "Neuronal Cholecystokinin", J.J. Vanderhaeghen and J.N. Crawley, eds., Ann. New York Acad. Sci., New York.

Faris, P.L., 1985, Opiate antagonistic function of cholecystokinin in analgesia and energy balance systems, in: "Neuronal Cholecystokinin", J.J. Vanderhaeghen and J.M. Crawley, e~, Ann. New York Acad. Sci., New York.

Gillis, R.A., Quest, J.A., Pagani, F.D., Dias-Souza, J., Taveira da Silva, A.M., Jensen, R.T., Garvey, T.G., and Hamoshi, P., 1983, Activation of Central Nervous System cholecystokinin receptors stimulates respiration in the cat,~ Pharmacal. Exp. Ther., 224: 408.

Hurle, M.A., Mediavilla, A., and Florez, J., 1985, Differential respiratory patterns induced by opioids applied to the ventral medullary·and dorsal pontine surfaces of cats, Neuropharmacol., 24: 597. .

Hurle, M.A., Morin-Surun, M.P., Foutz, A.J., Boudinot, E., and DenavitSaubie, M., 1985, Different targets involved in the effect of cholecystokinin on respiration, Europ. ~Pharmacal., 118: 87.

Schlaefke, M.E., 1981, Central chemosensitivity: a respiratory drive, Rev. Physiol. Biochem. Pharmacal., 90: 171.

Zarbin, M.A., Innis, R.B., Wamsley, J.K., Snyder, S.H., and Kuhar, M.J., 1983, Autoradiographic localization of cholecystokinin receptors in rodent brain, ~ Neurosci., 3: 877.

158

INFLUENCE OF THE NEUROHORMONES: HUMAN CORTICOTROPIN RELEASING

FACTOR (hCRF), THYROTROPIN RELEASING HORMONE (TRH),

ADRENOCORTICOTROPIC HORMONE (ACTH) ON VENTILATION IN HUMANS

INTRODUCTION

M. Nink1, I. Huber 2, U. Krauze2, and H. Lehnert2

1Dept. of Pneumology and 2Dept. of Endocrinology Mainz, University Hospital, FRG

While exploring the biological effects of i.v. - administered hCRF in volunteers, we observed a marked ventilatory stimulation which was obviously provoked by hCRF (Oppermann et al., 1985). In a controlled study we demonstrated a highly significant ventilatory analeptic effect in 13 healthy male and female subjects. This effect was dose-dependent, minute volume and tidal volume rose markedly. Breathing frequency was not found to be altered (Nink et al., 1988 in press).

After pretreatment with dexamethasone (2 mg, orally) the analeptic effect of hCRF was significantly (p<0.01) potentiated (Huber et al., 1988). Bohmer et al. (1985) demonstrated a ventilatory stimulation under topical application of hCRF at brain stem level in rabbits. This finding is in good accordance with the discovery of a CRF-peptidergic neuronal system in medulla oblongata (Schipper et al., 1983), a neuronal complex where regulatory centers of ventilation are situated.

We thus believe that CRF acts as a physiological modifier of respiratory function in humans. In a follow up study we tested the effects of THR and ACTH on ventilation in humans after systemic application of both drugs.

METHODS

Ventilatory monitoring during drug testing of hCRF (synthetic hCRF, BACHEM AG, Bubendorf, Switzerland), TRH (synthetic TRH, THYROLIBERINR, MERCK, Darmstadt, FRG) and ACTH (synthetic ACTH_, SYNACTHENR, CIBA-GEIGY, Wehr, FRG) was performed in our pulmonary function laboratory while our volunteers were seated in a semi-recumbent position. Monitoring included continuous pneumotachographic registration of ventilatory flow, minute volume and furthermore end-expiratory partial pressures of C02 and 02 measured at the mouth by a mass-spectrometer.

A cubital vein was punctured with a "butterfly-cannula", intravenous bolus injections of the drugs were performed via a long connecting-

159

tubing. The drug applications were controlled by bolus placebo injections (saline). Standardized monitoring began, when steady state conditions of ventilation were reached and continued until drug effects had ceased and steady state was reached again.

STATISTICAL EVALUATION

WILCOXON-test for paired samples was used. Ventilatory response to drug application was compared to the corresponding parameters measured after placebo injections.

RESULTS

We observed a dose dependent rise of minute volume in 13 test subjects after bolus injection (30 s) of 33, 66, 100 ug hCRF. Minute volume rose from 7.0 ± 1.3 1 (saline) to 10.6 ± 2.1 1 (33 ug), 11.9 ± 2.4 1 (66 ug) and 12.9 ± 3.2 1 (100 ug) (Fig. 1.). This effect was highly significant when comparing these results with those after placebo administration (p<0.001). The rise of minute volume was essentially caused by an augmented tidal volume. Breathing frequency was not essentially affected. Our test persons predominantly felt a strong respiratory analeptic effect.

In a group of 15 female and male subjects ventilatory air-flow rose significantly after 200 ug TRH: 4.56 ± 1.3 1/s (saline) vs. 5.51 ± 1.29 1/s (200 ug TRH). This effect was also significant (p<0.01). Minute volume and tidal volume did not change as homogenously as after hCRFapplication; both parameters increased, but this rise was not significant (p>0.05) (Fig. 2.).

In a pilot study ACTH (25 I.U.) was given systemically to 6 test persons. Preliminary results showed no uniform effect on ventilation. In 4 subjects ventilation did not change noticeably yet two persons showed a marked depression of breathing with respiration pauses up to 8 s and a fall in minute volume down to 711 (!). This ventilatory depression obviously was a drug effect: it started abruptly after drug administration and was attributed to the medication by the test subjects.

DISCUSSION

Our results demonstrate an analeptic ventilatory effect after intravenous bolus injection of hCRF and TRH. hCRF provoked a rise of minute volume by an augmented tidal volume, thus dead space ventilation remains low and an economical pattern of breathing results. This respiratory analeptic effect was accompanied by effects on autonomic and visceral functions. Similar effects were also seen after TRHadministration.

Pretreatment with dexamethasone potentiates the respiratory analeptic effect of hCRF. Possibly a suppression of the hypothalamic CRF-secretion by dexamethasone causes an upregulation of central CRFreceptors thereby explaining this effect. Our findings on respiratory analepsy of both releasing hormones are consistent with evidence from the literature (animal studies) where a participation of CRF in the control of autonomic and visceral functions has been described (Brown et al., 1982; 1984). For TRH an influence on behavioral, cardiovascular and metabolic responses is well known (Horita et al., 1975). Besides these effects hCRF and TRH presumably exert a modulating or even a controlling

160

AMY Ill

n =13

15

13 l n.•· 12.9. !3,2 j 10.6·

!2.4 11

!2,1

9

7 17,0!1,3 •p<0.001 when compared to pkH:ebo

Placebo hCRF i.v. 33,AI!I ~ 100pg

Fig. 1. Minute volume (x ± SD) of 13 male subjects after placebo and intravenous bolus injection of human CRF (33, 66, 100 ug).

9 Ills I

n=15

8 p<0,01

6

4

2

I 4.56 I !1.3

I 5,51 I !1.29

TRH 200,ug

Fig. 2. Combined inpiratoryjexpiratory air-flow values (x + SD) of 7 female and 8 male subjects after placebo (saline) and after i.v.-TRH-injection (200 ug)

161

effect on ventilation. This effect seems to be a central one (Bohmer et al., 1985; Redner et al., 1983). Thus, for TRH and hCRF to exert centrally mediated physiological effects, there has to be a bypass of the blood-brain barrier to reach target brain structures. The only site where capillaries are physiologically permeable are the so-called circumventricular organs. These are highly vascularized organs with fenestrated endothelial cells. Passage through circumventricular organs and acting through area postrema and nucleus tractus solitarii might constitute the physiological basis for a ventilatory control by hCRF and TRH.

An influence of ACTH on ventilatory patterns also seems to exist, yet further research must be done in this field.

REFERENCES

Bohmer, G., Schmid, K., Oppermann, D., and Ramsbott, M., 1985, Effects of human corticotropin-releasing-factor on central respiratory activity in the rabbit, Pflligers Arch. Suppl. Sept.

Brown, M.R., Fisher, L.A., Spiess, J., Rivier, C., and Vale, M., 1982, Corticotropin releasing factor (CRF): actions on the sympathetic nervous system and metabolism, Endocrinology, 111: 928.

Brown, M.R.,and Fisher, L.A., 1984, Brain peptide regulation of adrenal epinephrine secretion, Am.~ Physiol., 147: 41.

Redner, J., Redner, T., Wessberg, P., Lundberg, D., and Jonason, J., 1983, Effects of TRH and TRH analogues on the central regulation of breathing in the rat, Acta Physiol. Scand., 117: 427.

Horita, A., and Carino, M.A., 1975, Thyrotropin-releasing hormone induced hyperthermia and behavioral excitation in the rat, Psychopharmacol. Commun., 1: 403.

Huber, J., Krause, U., Nink, M., Lehnert, H., and Beyer, J., 1988, Dexamethasone potentiates the respiratory analeptic effect of corticotropin releasing factor, ~ Clin. Endocrin. Metabol.- in preparation.

Nink, M., Huber, J., Krause, U., Schultz, V., Beyer, J., and Lehnert, H., 1988, Thyrotropin Releasing Hormone (TRH) und humaner Corticotropin Releasing Faktor (hCRF) als Modulatoren der Atmung beim Menschen. -Physiologie und Klinik- Prax. Klin. Pneumol., 41: in press.

Oppermann, D., Huber, I., Nink, M., Schulz, V., Beyer, J., and Ferlinz, R., 1985, Corticotropin-releasing-factor- ein neues sehr wirksames Atem-analeptikum, Klin. Wochenschr., 63: Suppl. VI, 555.

Schipper, J., Steinbusch, H.W.M., Vermes, I., and Tilders, F.J.H., 1983, Mapping of CRF-immunoreactive nerve fibers in the medulla oblongata and spinal cord of the rat, Brain Res., 267: 145.

162

TUBE BREATHING UNDER GENERAL ANAESTHESIA

V. Smejkal, F. Palecek, R. Havelka, and D. Miloschewsky

Inst. of Pathol. Physiology, Faculty of Pediatrics, Charles University and Department of Anaesthesiology, Hospital Bulovka, Prague, Czechoslovakia

During anaesthesia metabolism and production of C02 are decreased. Chemoregulation is impaired.

It is impossible, without measurements, to predict changes of PC02 in the organism during tube breathing, as the response can be iso- or hypercapnic, or mixed (Palecek, 1983). The aim of our study was to compare C02 during tube breathing in wake and anaesthetized animals and man.

MATERIALS AND METHODS

The animal experiments were carried out in 40 male rabbits and clinical studies in 7 women.

The chinchilla rabbits weighed in average 2.25 kg. An arteriovenous by-pass was prepared at the beginning of the experiment under 1.8% halothane anaesthesia. The time design of the experiments is on Fig. 1. After the preparation, the anaesthesia was stopped and a dead space in the form of a tube (volume 3, 6, or 10 ml, inner diameter 8 mm) was added between the animal and the tidal volume recording system. Ventilation was stimulated by tube breathing in wake rabbit and then under general ethylurethane anaesthesia (urethane dose 1.3 gjkg, one half s.c., the other i.p.). The tidal volume was evaluated by a laboratory gas-meter. The PaC02 was measured continuously in the by-pass (Smejkal et al., 1985).

In human experiments the subjects were injected at the beginning of each study with pethidine in the dose of 100 mg s.c. and atropine in the dose of 0.5 mg s.c. The women weighed in average 66 kg and were prepared to a minor gynecological surgical intervention (curettage). Ventilation was stimulated in the wake state after the premedication by added dead space. The added tube had a volume of 600 ml and inner diameter of 30 mm. For time design see Fig. 1. The same test was then carried out under general thiopental anaesthesia. Thiopental was slowly injected intravenously in the dose of 300 mg. Tidal volume was evaluated with a

163

164

10

0

2

0 J l . -1 · i'TW')

VE

Preparation of a.-v bypass

TIME DESIGN

IN RABBITS

xo measurements Halothane Woke Anaesthesia anoest . ,.......-/.,__r-_X~X~X~X~XX~X~XrX ___ ~X"iXrXXo:;X:.:;.X;;;XX;:..:;X:,__--,

0 30 40 50 60 min

IN WOMEN

Premedication GJI·vo I surgical treatment

Woke l Anaesthesia r-f

X X X

0 30 40

Fig. 1. Time design

-2 0 +2 +4 + 6 min

TIME

50 min

of experiments.

PaC~

kFI:J

6

5

4

l + "o 10 rri

p.. - -o- - -<> - .0..· -o.--o

~ <>- - - - 4

0-'--r---r--...---...---r--

- 2 0

TM:

Fig. 2. Examples of "iso-" and hypercapnic responses in rabbits. Solid circles represent individual studied values in wake animal ("isocapnic" response), open circles- in anaesthetized animal (hypercapnic response). Response was considered "isocapnic" if the increase of PaC02 did not exceed 0.67 kPa.

•;. 100

50

0

ADDED VD:

11 OGJ WAKE

3 6 10

9 D * 20

ANAESTI£SIA

3 6 10 rri

Fig. 3. The frequency of "isocapnic" response in anaesthetized and wake rabbits. The columns indicate quantal responses. For isocapnic the same criterion was used as in Fig. 2. The numbers in columns show the numbers of experiments for each value. Asterisk indicates the start. Significant difference in comparison with the wake state.

gas-meter (Volumometer GDR type No. 44101). The arterialized ear lobe PC02 was measured intermittently with a microelectrode.

The results were evaluated by a paired-test, p i 0.05.

RESULTS

The mean tidal volume in rabbits was 15 ml. Tube breathing increased tidal volume, minute ventilation and the PaC02.

Fig. 2 shows examples of an "isocapnic" response to the added dead space of 10 ml in a wake rabbit and hypercapnic response in an anaesthetized rabbit.

Fig. 3 indicates the frequency of "isocapnic" responses in rabbits depending on volume of added dead space. This frequency is significantly lower under general anaesthesia with the large space of 10 ml.

The table shows that in anaesthetized women hypercapnia during tube breathing is significantly higher than in the wake state.

DISCUSSION

In rats under urethane anaesthesia an analogous responses to tube breathing as in rabbits was seen (Palecek, 1983).

The insertion of added dead space decreases the effectiveness of ventilation and results in a retention of C02 in relation to the volume

165

Table. Mean values and S.D. of some respiratory indices in wake and anaesthetized women without, or with added Vn in the form of a tube volume 600 ml.

Value units

Tidal volume [ml]

Resp. frequency [c .min-1]

Minute ventilation [l.min-1]

PC02 [kPa]

P02 [kPa]

I: Wake no Vn

499 + 191

15 ± 3.2

7.5 ± 2.4

4.8 ± 0.8

11.0 ± 1.5

II: Wake + Vn

919 ± 167*

17 ± 4.4

15.4 ± 3.8*

5.3 ± 0.6

8.0 ± 1.6*

III :Anaesthesia + vn

787.5 ± 61.6-.x

17 ± 2

13.4 ± 1. 7*

5.6 ± 0.6*

7.9 ± 1.4*

* = stat. significant difference in comparison with I x = stat. significant difference in comparison with II

of added dead space to the resting tidal volume. The same phenomenon as in rabbits was seen in man (Smejkal et al., 1985). Stimulating ventilation by added dead space (tube breathing) can be u~~ful a~ a clinical functional test and for evaluating the effect of drugs on respiration.

CONCLUSION

Hypercapnia during tube breathing is more pronounced in anaesthetized animals and men than in wake ones.

REFERENCES

Pale~ek, F., 1983, Regulation of breathing in pulmonary diseases, (in Czech), Avicenum, Praha.

Smejkal, V., Palecek, F., and Strnad, P., 1985, Test of tube breathing in healthy man, Acta Facult. Med. Univ. Brunensis, 92: 119.

166

EFFECTS OF ENFLURANE ON THE VENTILATORY RESPONSE TO INCREASED

CARBON DIOXIDE AND METABOLIC RATE IN DOGS

INTRODUCTION

D.S. Ward, R. Ginsburg, I.H. Abdul-Rasool and K. Aqleh

Department of Anesthesiology, UCLA Medical School Los Angeles, CA 90024, USA

Inhaled anesthetic agents cause a dose related ventilatory depression as evidenced by an increase in arterial PC02 (PaC02) and blunting of the ventilatory response to hypercapnia and hypoxia (Hirschman et al., 1977). Ventilation is also closely coupled to metabolic rate, resulting in regulation of PaC02 in spite of changes in metabolic rate (Wasserman et al., 1986). This study investigates the effects of inhalational anesthetics on this coupling.

METHODS

Six mongrel dogs (18.5 to 23.5 kg) were given 10 mgfkg ketamine, i.m., followed by enflurane in air and the dogs were allowed to breathe spontaneously through an endotracheal tube. Blood gas analysis was performed on an IL system 1303. Airway gas concentrations were continuously measured by a calibrated mass spectrometer (Perkin Elmer 1100 MGA). Expired and inspired airflow were measured with an impeller flowmeter (Alpha Technologies, VMM110).

The upper portion of each hindlimb was shaved and conductive gel with 0.75% bupivicaine applied. Thin steel strip electrodes were sutured around the upper thigh and a brass rod electrode was inserted into the rectum. The exercise apparatus consisted of a frame with horizontal steel bar and traveller, to which a spring was attached. During exercise, the legs, which had been bound to the traveller, were extended along the steel bar against the force of the spring. Rhythmical muscular contractions were produced with an electronic stimulator (Grass Instruments S48). Stimulating voltages were chosen to produce maximal leg movement during 2.5% end-tidal enflurane.

The inspired enflurane concentration was adjusted to obtain endtidal concentrations of 2.5, 3.0 and 3.5% and equilibrated for at least 60 minutes prior to each experimental run. The experiments consisted of two five-minute periods; rest and exercise. An attempt was made to perform each experiment twice at each level of end-tidal enflurane in every dog (see results). A stabilization period of at least 30 minutes was allowed between each experimental run. The hypercapnic response was

167

Table 1. Values (mean ± SD) for one minute breath by breath averages for the last minute for each of the 5 minute intervals. See text. " p < .05 compared to Rest values, # p < 0.05 as compared to corresponding values measured at 2.5% enflurane.

ENFLURANE 2.5% 3.0% 3.5%

REST EXER REST EXER REST EXER

VE 4.75 7. 81"' 3.9 5. 1"' 3.9 5.05*# +1.09 +1.65 +1.02 +0.85 +1.0 +0.8

f 14.4 20.6" 11.3 13.7* 10.5 13.1* +6.6 +9.8 +5.4 +6.7 +5.8 +5.8

vT 0.37 0.45" 0.40 0.45 0.43 0.45 +0.16 +0.20 +0.19 +0.18 +0.18 +0.21

V(C02) 147 265* 114 170" 116 159*# +67 +83 +22 +30 +25 +12

P£TC02 45 47 47.0 47.7 45.7 46.0 +3.3 +2.8 +4.1 +3.4 +2.9 +2.7

PaC02 47.6 45.7 47.7 48.0 48.3 46.2 +3.1 +4.6 +4.2 +8.2 +4.1 +2.9

VD/VT 0.43 0.34* 0.45 0.41 0.43 0.39 +.19 +.18 +.14 +.21 +.12 +.13

obtained using the Read rebreathing technique at each enflurane concentration in each dog.

Data was collected on-line using a DEC LSI 11/23 microcomputer and included breath-by-breath respiratory and anesthesic gas concentrations, volume measurements and leg movements. For the data analysis of each experiment, the values of the respiratory variables were averaged over the last minute of each interval. The corresponding interval means for each run were then averaged. The relationship between VE and end-tidal PC02 were obtained by least-squares linear regression for each experiment. Unpaired Student t-test, analysis of variance and Bonferroni's correction for multiple comparisons were used as appropriate.

RESULTS

At rest, minute ventilation (VE, ljmin) decreased only slightly through a decrease in breathing frequency (f, breaths/min); carbon dioxide production (V(C02), mljmin), end-tidal PC02 (PErC02, Torr), PaC02 (Torr) and dead space to tidal volume ratio (VDfVT) at different levels of end-tidal enflurane concentration (Table 1). Because of the seizure activity at the higher enflurane concentration, 11 experiments (6 dogs) were performed at 2.5% enflurane, 9 (6 dogs) were completed at 3.0% and 7 (5 dogs) at 3.5%. Because of problems with blood gas analysis these results are for one less dog and experiment at each enflurane level. As the enflurane was increased the amount of leg movement (stimulus parame-

168

ters were unchanged) decreased, this resulted in a lower steady-state exercise V(C02).

The ventilatory response to inhaled C02, as measured by the slope of VE to PErC02 relationship, decreased significantly with increasing enflurane (0.64 + 0.42, 0.36 + 0.13 and 0.20 + 0.08 l.min-1.Torr-1, at 2.5, 3.0 and 3.5~ respectively). The ventilation response occurred mainly by an increase in tidal volume with insignificant changes in breathing frequency.

The VE response to exercise consisted of an increase in Vr and f and a decrease in VnfVT· The maintenance of constant PaC02 requires a coordinated response in all three of these variables. The relative contribution of each of these components can be assessed by calculating from equation (Hirschman et al., 1977) how much the PaC02 would have increased if there had been no change in that variable while the other variables (including V(C02) changed by the measured amount.

Paco2 = 863 • vcco2) 1 {Vr • f • (1 - Vn/Vr)} [1]

Fig. 1 shows the results of these calculations. If all the variables had remained at their resting values the increase in PaC02 would have been as shown. Because the increase in metabolic rate was less at the higher enflurane concentrations there was a smaller increase in PaC02 calculated for the higher concentrations. Note that the effects are not additive because of the multiplicative form of equation [1]. The response in the breathing frequency was the predominate effect at each level of enflurane.

40 !Z2J 2.5% E..,_

J 30 IZ;zj 3.5% E~

8 l. 20 iii

!!I z g 10

0

-10 VT VD/ VT r AU None

Carnponent Fb.ed

Fig. 1. The amount that the PaC02 would increase in response to the increase in V(C02) if the Vr, f or VnfVT remained fixed at the rest value . The increase if they were all fixed and the actual change (none fixed) is also shown.

169

DISCUSSION

The mechanisms for the apparent tight coupling of ventilation to metabolic rate are not well defined. Since anesthetic drugs have marked effects on the peripheral and central chemoreflexes, it might be expected that the metabolic rate to ventilation relationship would also be modified. Available data are conflicting, but apparently anesthetic drugs have less effect on the response to exercise than they do on the chemoreflex loops.

In dogs one MAC of enflurane (2.2%) causes no changes in VE and PETC02 as compared to the awake values (Hirschman et al., 1977). In this study, increasing the concentration of enflurane from 2.5 to 3.5% caused no significant further changes (Table 1). In contrast, enflurane causes a dose related depression of the C02 response (Hirschman et al., 1977). In this study, the decrease in the slope of the C02 response curve suggests that the central chemoreflex is increasingly depressed as the concentration of enflurane is increased. This depression occurred despite the fact that resting ventilation and PaC02 remained unchanged.

Chonan and Hida and their coworkers (Chonan et al., 1984; Hida et al., 1986) reported extensive experiments in dogs with chloralose-urethan anesthesia. In both of their studies, the dogs became hypercapnic in response to exercise only at deep levels of anesthesia. At light and moderate levels of anesthesia, the dogs were either hypocapnic or isocapnic. The steady state ventilatory response was depressed by increasing anesthetic depth but unaffected by rhizotomy.

There are apparently differences between the respiratory effects of enflurane anesthesia and chloralose-urethan anesthesia in dog. As judged by the depression in the C02 response slope, our deepest level of anesthesia is equivalent to their moderate levels. For the amount of increase in resting C02 and the apparent depth of anesthesia, enflurane is less disruptive of metabolic rate to ventilation coupling than is chloralose-urethan. The isocapnic response to exercise still occurs despite the C02 response slope being reduced to one third of the value found under light anesthesia. Chonan and Hida and their associates (Chonan et al., 1984; Hida et al., 1986) stimulated the sciatic and femoral nerves which were cut proximal to the site of stimulation while we used external electrical stimulation. It is possible, therefore, that this difference in the experimental model contributed to the discrepancies of the results.

Enflurane has pronounced effects on the ventilatory timing mechanisms as well as effects on total ventilation. At rest the ventilatory frequency was reduced by enflurane but still the increase in frequency was the primary mode for the ventilatory response to exercise. In contrast, tidal volume was the primary mode for the response to hypercapnia and this was greatly effected by the increasing concentrations of enflurane. The role of VnfVT in the exercise response is less clear, but as Figure 1. illustrates, substantial hypercapnia would occur with exercise if Vn/Vr did not decrease. Enflurane could potentially affect this response by its effects on bronchial smooth muscle.

Both neural afferents from the exercising limbs and humoral components have been proposed to have roles in the hyperpnea of exercise, although their relative importance is still unknown (Wasserman et al., 1986). This study can not determine the mechanisms of exercise hyperpnea, but our data indicate that enflurane has little effect on these mechanisms. This may be related to changes in three variables (Vr,f,Vn/Vr) being responsible for the isocapnic exercise response, while

170

only one variable (Vr) is responsible for the response to exogenous C02. Our data support the hypothesis that the central and peripheral chemoreflexes are not obligatory mechanisms responsible for exercise hyperpnea.

ACKNOWLEDGEMENTS

D.S. Ward acknowledges funding by a Senior International Fellowship (1F06TW01276-01) from the Fogarty International Center of the NIH.

REFERENCES

Hirschman, C.A., McCullough, R.E., Cohen, J.P., and Weil, J.V., 1977, Depression of hypoxic ventilatory response by halothane, enflurane and isoflurane in dogs, Br. Jr. Anaesth., 49: 957.

Wasserman, K., Whipp, B.J., and Cassaburi, R., 1986, Respiratory control during exercise. in: "Handbook of Physiology, Section 3, The Respiratory System, Vol. II, Control of Breathing, Part 2", N.S. Cherniack and J.G.Widdicombe, eds., American Physiological Society, Washington, D.C.

Chonan, T., Kikuchi, Y., Hida, W., Shindoh, C., Inoue, H., Sasaki, H., and Takishima, T., 1984, Response to hypercapnia and exercise hyperpnea in graded anesthesia,~~ Physiol., 57: 1796.

Hida, W., Shindoh, C., Kikuchi, Y., Chonan, T., Inoue, H., Sasaki, H., and Takishima, T., 1986, Ventilatory response to phasic contraction and passive movement in graded anesthesia,~~ Physiol., 61: 91.

171

EFFECTS OF CHANGES IN INSPIRATION VOLUME AND

FLOW RATE ON RESPIRATORY ACTIVITY

INTRODUCTION

Murray D. Altose, Alfred F. Connors, Jr., and Anthony F. DiMarco

Department of Medicine, Case Western Reserve University Cleveland, Ohio 44109, USA

Changes in inspired volume and inspiratory flow rate alter the activity of vagal receptors in animals and this, in turn, causes reflex changes in respiratory neural activity (Bartoli et al., 1975). However, little is known about the effects of such changes on respiratory activity in humans. The present study, in normal subjects, examined the effects of mechanically controlled changes in inspired volume and inspiratory flow rate on the level and pattern of respiratory motor output.

METHODS

Ten normal subjects were studied in the supine position. The apparatus consisted of a mouthpiece connected to a directional valve. Mouth pressure was measured with a pressure transducer and end-tidal PC02 was recorded with an infrared analyzer. Airflow was measured with a pneumotachograph and the flow signal was electrically integrated to obtain tidal volume.

The breathing circuit was connected to a volume-cycled mechanical respirator (Bennet MA-l) to provide assisted ventilation. In two separate trials in each subject, the end-tidal PC02 was adjusted to a level of 40 and 46 mmHg. Following a period of equilibration, several combinations of inspired volume and inspiratory flow rate were set for 2 minutes each, in a random order. In one series, inspiratory flow rate was held constant at 75 1/min while inspired volume was set at 12, 18 and 24 ml/kg body weight. In a second series, inspired volume was held constant at 18 ml/kg with inspiratory flow rate settings of 50, 75 and 100 1/min. In the third series both inspired volume and inspiratory flow rate were varied. Three combinations of flow and volume included 12 mljkg at 50 1/min, 18 ml/kg at 75 1/min, and 24 ml/kg at 100 1/min so as to maintain a constant inspiratory duration.

Occlusion pressure was determined during mechanical ventilation on a breath-by-breath basis. The sensitivity setting on the assist-trigger of the respirator was adjusted to allow at least 100 msec between the onset of inspiratory effort and the initiation of flow from the respirator.

173

Since the airway was effectively occluded during this time, occlusion pressure could be ascertained.

In a separate series of trials of mechanical ventilation at PC02 42 mmHg, an aerosol of normal saline was delivered using a jet nebulizer. This was followed by control measurements of the respiratory responses to simultaneous changes in inspired volume and flow rate. During a subsequent period of mechanical ventilation an aerosol of 41 lidocaine was administered and thereafter, the effects of simultaneous changes in inspired volume and inspiratory flow rate were again assessed.

RESULTS

At all respirator volume and flow rate settings, expiratory duration was significantly shorter at PC02 46 mmHg as compared to PC02 40 mmHg. At both PC02 levels, progressive reductions in inspiratory flow rate alone resulted in no significant changes in expiratory duration. However, with reductions in inspired volume, expiratory duration progressively shortened both at PC02 40 mmHg and 46 mmHg. Similarly, as both inspiratory flow rate and inspired volume were simultaneously reduced, expiratory duration progressively fell.

The occlusion pressure during mechanical ventilation was significantly higher at PC02 46 mmHg than at PC02 40 mmHg, reflecting the respiratory motor output response to hypercapnia. With reductions in inspiratory flow rate, there was a tendency for the occlusion pressure to increase at PC02 46 mmHg. This tendency was not observed at PC02 40 mmHg and the occlusion pressure values at the high and low flow settings were the same. At a constant inspiratory flow rate, changing from the high to the medium volume setting did not affect the occlusion pressure. However, the occlusion pressure increased significantly with further reductions in inspired volume to the low setting. Combined reductions in both inspiratory flow rate and inspired volume resulted in significant increases in occlusion pressure at PC02 40 mmHg and at PC02 46 mg.

Airway anesthesia with inhaled aerosolized lidocaine had no effect on the changes in expiratory duration resulting from combined reductions in inspired flow and volume at PC02 46 mmHg. At the high flow and volume settings, the occlusion pressure was also not affected by lidocaine inhalation. However, the increase in occlusion pressure with reductions in inspiratory flow and volume was blocked by lidocaine airway anesthesia.

DISCUSSION

This study demonstrated that reductions in inspired volume and inspiratory flow rate during mechanical ventilation in normal people result in a shortening of expiratory duration and an increase in respiratory motor output as measured by the occlusion pressure. The occlusion pressure response to reductions in inspired volume and inspiratory flow rate, however, is blunted after the inhalation of aerosolized lidocaine. This suggests that the increases in inspiratory motor activity may be mediated by airway receptors.

Local anesthetic aerosol inhalation has been previously utilized to investigate the role of airway receptors in the control of breathing in humans. In general, lidocaine aerosols produce only small changes in the pattern of spontaneous resting breathing. These are characterized by slight increases in tidal volume consequent to prolongations of

174

inspiratory duration (Easton et al., 1985). The conditions of the present study, however, were considerably different from those in previous investigations. Assisted ventilation was utilized in our subjects and both inspired volume and inspiratory flow rate were regulated by the mechanical respirator.

It is not clear which receptors are involved in mediating the respiratory responses to changes in inspired volume and flow rate. Slowly adapting pulmonary stretch receptors seem to have very little effect on the pattern of breathing in humans until a certain volume threshold is exceeded (Clark and von Euler, 1972). Even then, stretch receptor feedback only serves to shorten inspiratory duration without influencing respiratory drive (Polachek et al., 1980). Consideration should also be given to upper airway reflexes. It has been suggested that pressure and flow changes in the upper airway can influence respiratory activity. Also, in conscious humans, airflow through the larynx is known to inhibit respiratory activity (Mathew and Farber, 1983; McBride and Whitelaw, 1981).

Regardless of the precise mechanisms, it is clear from the observations of this study that respiratory activity, even at a constant chemical drive, varies with the level and pattern of changes in thoracic volume. Thus, without changing PC02 and P02, intrinsic respiratory activity and work of breathing can be minimized during mechanical ventilation by increasing inspired volume and inspiratory flow rate.

REFERENCES

Bartoli, A., Gross, B.A., Guz, A., Huszczuk, A., and Jefferies, R. 1975, The effect of varying tidal volume of the associated phrenic motoneurone output: studies of vagal and chemical feedback, Respir. Physiol., 25 : 135.

Clark, F.J., and von Euler, C., 1972, On the regulation of depth and rate of breathing, ~ Physiol., 222: 267.

Easton, P.A., Jadue, C., Arnup, M.E., Meatherall, R.C., and Anthonisen, N.R., 1985, Effects of upper and lower airway anesthesia on hypercapnic ventilation in humans,~~ Physiol., 59: 1090.

Mathew, O.P., and Farber, J.P., 1983, Effect of upper airway negative pressure on respiratory timing, Respir. Physiol., 54: 259.

McBride, B., and Whitelaw, W.A., 1981, A physiological stimulus to upper airway receptors in humans, ~ ~ Physiol., 51: 1189.

Polachek, J., Strang, R., Arens, J., Davies, C., Metcalf, I., and Younes, M., 1980, Phasic vagal influences on inspiratory motor output in anesthetized human subjects, ~ ~ Physiol., 49: 609.

175

EXCITATORY EFFECTS OF ELECTRICAL AND CHEMICAL STIMULATION OF THE

BOTZINGER COMPLEX ON EXPIRATORY ACTIVITY IN THE CAT

INTRODUCTION

F. Bongianni, G. Fontana and T. Pantaleo

Dipartimento di Scienze Fisiologiche, Universita di Firenze, Firenze, Italy

Lipski and Merrill (1980) have described a population of expirationrelated (ER) neurons -the so-called Botzinger complex (Bot. c.) - in the region of the retrofacial nucleus, at the rostral pole of the ventral respiratory group (VRG). The presence of bilateral axonal projections of these neurons to the almost purely ER region of the nucleus retroambigualis (nRA) in the caudal VRG suggests that Bot. c. neurons have a role in shaping the firing patterns of nRA neurons (Fedorko and Merrill, 1984).

In the present research we studied the effects of electrical stimulation of the Bot. c. region on the activity of the expiratory muscles and ER neurons of the caudal VRG. Microdoses of potent excitatory amino acids were injected into the same area to insure that the observed responses were due to the activation of perikarya and not of axons of passage (Goodchild et al., 1982).

METHODS

Experiments were performed on vagotomized cats under pentobarbitone anesthesia (35 mgjkg i.p.). In some experiments the animals were also paralyzed (gallamine thriethiodide, 5 mgjkgjh i.v.) or bilaterally thoracotomized, and artificially ventilated. Efferent phrenic nerve activity was recorded from desheated C5 phrenic roots. Electromyograms (EMGs) were derived from abdominal, external and internal intercostal muscles. The phrenic and EMG activities were amplified, full-wave rectified and passed through a "leaky" integrator (time constant 100 ms) to obtain a "moving average" of the activities. Extracellular neuronal activity was recorded from ER neurons of the Bot. c. and of the caudal VRG with tungsten microelectrodes and processed in the same way. Both phrenic and neuronal activities were displayed as "raw" signals on an oscilloscope and photographed. "Moving averages" of phrenic, EMG and neuronal activities were displayed on a multichannel pen recorder together with the signals of arterial blood pressure, tracheal pressure, tidal volume and end-tidal C02. Monopolar cathodal microstimulation of the Bot. c. (0.05-0.5 ms current pulses at intensities of 5-60 AA) was effected through the same electrode used for recording neuronal activity. Micro-

177

Phr. contra

E.l. ipsi

Abd. ipsi

B.P. mmHg

200

[ ... 100

•• u • -stim.

u 0 • 10s

Fig. 1. Activation of the expiratory muscles during the apneic response induced by Bot. c. tetanic stimulation (80 Hz, 20 uA) in one vagotomized spontaneously breathing cat. Records from top to bottom: contralateral phrenic nerve activity (Phr. contra); ipsilateral external intercostal muscle activity (E.I. ipsi); ipsilateral abdominal muscle activity (Abd . ipsi), arterial blood pressure (B.P.). A black bar indicates the stimulation period.

injections (25-100 nl) of L-glutamate monosodium salt (GLU) 0.5 M and DLhomocysteic acid (DLH) 0.16 M (Goodchild et al., 1982) were performed via a glass micropipette fused to the microelectrode in order to apply electrical and chemical stimulation to the same area from which ER activity was recorded. Serial frozen sections (50 urn tick) of the medulla were used to define stimulation and recording sites.

RESULTS

Sustained tetanic stimulation of the Bot. c. at 20-100 Hz (5-15 s duration) caused strong depression in the inspiratory activity up to complete apnea. This depression was a graded effect which depended upon the intensity, and especially the frequency of stimulation . This effect was consistently accompanied by the activation of the expiratory muscles, which displayed tonic activity during the apneic response (Fig. 1). Such tonic activation was induced only at stimulation frequencies above a certain level (usually about 40Hz); augmentations of tonic expiratory activity were generated by further increasing the frequency of stimulation.

Single shocks applied to the Bot. c. region elicited orthodromic activation in most of the ER neurons tested in both the ipsi- and the contralateral nRA. A prolonged silent interval usually followed the period of activation. These neuronal responses were, as a rule, confined to the expiratory phase. Some of these neurons showed an increase in the number of evoked spikes at shorter initial latencies, and even excitatory responses during the inspiratory phase, as the stimulus intensity was raised (Fig. 2). Such changes probably implied activation through disy-

178

A

c

B

0

'-' 5 ms

Fig. 2. Orthodromic activation of one ER neuron of the nRA elicited by single shocks at two different intensities applied to the contralateral Bot. c.: in A and B 10 ~; inC and D 40 ~. Traces triggered at the onset of the stimulation (filled triangles). Vagotomized, paralyzed and artificially ventilated animal. At the lower level of intensity, no response was obtained during the inspiratory phase (A) while excitatory responses were evoked during expiration (B). Stimulation at the higher strength elicited excitatory responses also during the inspiratory phase (C) and caused an increase in the number of evoked action potentials, accompanied by a decrease in the initial latency, during the expiratory phase (D).

naptic or polysynaptic pathways. The majority of ER neurons of the nRA were able to follow stimulation frequencies greater than 40 Hz, and in many instances up to 80-100 Hz.

Microinjections of GLU and DLH into the Bot. c. region caused effects which were consistent with those obtained by electrical stimulation, i.e., depression of the inspiratory activity or even apneic responses, consistently associated with the activation of the expiratory muscles (Fig. 3). Corresponding excitatory effects were observed in the activity of ER neurons of the caudal VRG.

CONCLUSIONS

These results support the hypothesis that Bot. c. neurons are involved in the control of the breathing pattern (Budzinska et al., 1985; Euler, 1986). The depressant effects observed on inspiratory activity appear to be coherent with previous observations suggesting or providing evidence that Bot. c . neurons have inhibitory projections to medullary inspirat ion-related neurons and to phrenic motoneurons (Fedorko and Merrill, 1984; Lipski and Merrill, 1980; Merrill and Fedorko, 1984; Merri ll et al ., 1983). Furthermore , in agreement wit h previous suggestions (Fedorko and Merrill. 1984), this investigation revealed that Bot.

179

Phr. contra

E.l. ipsi

A 8 c

10s

Fig. 3. Depressant effects on inspiratory activity associated with activation of the expiratory musc l es in response to 50 nl GLU (0.5 M) injected into the Bot . c . r egion. Vagotomized, thoracotomized and artificially ventilated preparation. In each panel from above: contralateral phrenic nerve activity (Phr. contra); ipsilateral external intercostal muscle activity (E. I. ipsi); ipsilateral abdominal muscle activity (Abd. ipsi); tracheal pressure (PTr)· A: control; Band C: 30 sand 10 min after the end of the injection, r espectively.

c. neurons can exert excitatory influences on the expiratory motor output through the activation of ER neurons in the caudal VRG.

ACKNOWLEDGMENTS

This study was supported by grants from the Ministero della Pubblica Istruzione of Italy.

REFERENCES

Budzinska, K., von Euler, C., Kao, F.F., Pantaleo, T., and Yamamoto, Y., 1985, Effects of graded focal cold block in rostral areas of the medulla, Acta Physiol. Scand., 124: 329 .

Euler, C. von, 1986, Brain st em mechanisms f or gener ation and control of breathing pattern, in: "Handbook of Physiology , Section 3, The Respiratory System, Vol. II, Control of Breat hing", N.S.Cher niack and J.G. Widdicombe. eds., American Phys i ological Soc i ety , Bethesda.

Fedorko, L. , and Merrill, E.G., 1984, Axonal pr ojections from the rostral expiratory neurones of the Botzinger complex to medulla and spinal cord in the cat, ~ Physiol. (London), 350: 487.

Goodchild, A.K., Dampney, R.A.L., and Bandler, R., 1982, A method for evoking physiological response by stimulat ion of cell bodies, but not axons of passage, within l ocalized regions of t he cent r a l n ervous system, ~ Neurosci. Met h., 6: 351.

Lipski, J., and Merrill, E.G., 1980, Electrophys iological demonst ration of the projections from expirator y neurones in rostral medulla to contralateral dorsal respiratory group, Brain Res., 197: 521.

Merrill, E.G., and Fedorko, L., 1984, Monosynaptic inhibition of phrenic motoneurons : a long descending projection from Botzinger neurons, ~ Neurosci., 4: 2350.

180

Merrill, E.G., Lipski, J., Kubin, L., and Fedorko, L., 1983, Origin of the expiratory inhibition of nucleus tractus solitarius inspiratory neurones, Brain Res., 263: 43.

181

TRIGEMINAL NERVE, BREATHING AND SLEEP APNEA

H. Gromysz, A. Kukwa, U. Jernajczyk, and W.A. Karczewski

Medical Research Centre, Polish Academy of Sciences Warsaw, Dworkowa 3, Poland

Earlier studies from this laboratory suggested that neurones of the Vth's nerve motor nucleus might be involved not only in their "routine" functions (including control of muscles responsible for maintaining upper airway patency) but also in the control of the respiratory pattern (Gromysz et al., in the press).

This view emerged from the following observations: i) the mylohyoid nerve (MN) which supplies some of the main muscles attached to the hyoid bone, exhibits phasic expiratory activity, (with an additional inspiratory component); ii) the expiratory activity is enhanced by lung inflation or inspiration-inhibiting vagal stimulation (100Hz). Lowfrequency vagal stimulation (30 Hz) which typically increased the central respiratory frequency, elicited tonic-modulated, low-amplitude activity in the MN: iii) after vagotomy, the MN activity becomes biphasic, an additional inspiratory volley firing in step with the phrenic and facial nerve activity.

We hypothesized that since the mylohyoid nerve's activity appears as a "mirror image" of the central inspiratory activity, its expiratory subvolley being precisely synchronized with the expiratory phase of the central respiratory cycle, its neurones might be directly or indirectly involved in the control of the respiratory pattern. Although the activities of the trigeminal nerve were already studied by several authors (see, e.g., St.John and Bledsoe, 1985) such a possibility wasas far as we know - never considered.

The present study was designed with the purpose of getting more insight into the role of the Vth nerve motor nucleus in controlling the pattern of breathing.

METHODS

Seventeen male rabbits were anaesthetized with halothane, paralyzed with d-tubocurarine and artificially ventilated at eucapnic level. Pa02 was kept apove 100 nwHg. Blood samples were regularly taken to check arterial blood gases and pH. End-expiratory C02 concentration was

183

continuously monitored. Arterial blood pressure was recorded from the femoral artery. Both vagus nerves were dissected in the neck and prepared for cutting and stimulation. The mylohyoid nerve , the C4 phrenic root and in some experiments also the facial ner ve were dissected, placed on bipolar silver electrodes and covered with warm mineral oil. The activities were amplified and recorded in a conventional way and as a "moving average".

The motor nucleus of the Vth nerve (NVmt) was stereotaxically penetrated with a tungsten-in-glass microelectrode for recording and stimulation. The penetration was histologically verified post mortem (Fig. 1).

RESULTS

Under control conditions, the phrenic and facial nerves discharged synchronously in inspiration. The NVmt units and the MN exhibited a similar, phase-spanning pattern of activity, with maximal frequency in expiration. After vagotomy this pattern usually changed to whole expiratory. Single-pulse stimulation at the recording site produced a short-latency (2.5 ms) powerful inhibition of phrenic nerve activity (Fig. 2). Afferent stimulation of the vagus nerve elicited a delayed, strong excitation of both NVmt units and MN, with a concomitant inhibition of phrenic nerve activity (Fig. 3). The behaviour of the facial and mylohyoid nerve activity was described elsewhere (Gromysz et al. in the press).

Fig. 1. Histological section of t he brainstem. Arrow indicates the point of recording and stimulation in NVmt.

184

A B

Fig. 2. Effects of a single-pulse stimulation of NVmt on phrenic nerve discharge. A: Five sweeps superimposed. B: Single sweep . Time scale: one div. = 5 ms.

B

1s

Fig. 3. Effects of afferent vagal stimulation (40 impjs , 20 uV) on a NVmt unit and phrenic nerve activity . B - 0.5 s later; upper traces -phreni c nerve, lower traces - NVmt unit. Horizontal bar -stimulat ion marker.

185

DISCUSSION

The short-latency inhibition of phrenic nerve discharges in response to NVmt stimulation has been the most striking finding of this study. Phenomenologically, this effect is similar to those elicited by stimulation of the NPBM-KF region ("pneumotaxic centre" - see Bertrand and Hugelin, 1971) or retrofacial nucleus ("Botzinger complex" - see Merrill et al., 1983). There are, however, obvious differences. The NVmt is a well-defined, easily recognizable brain stem structure which is quite distant from the retrofacial nucleus. The NPBM-KF system, although very close to the NVmt, forms also an obviously separate set of nuclei and the site of stimulation cannot be mistaken provided that histological examination is - as in our case - carefully performed after each experiment. Moreover, the latency of response is entirely different. Cohen (1971) demonstrated an expiratory-facilitatory effect of stimulation at pontine ventrolateral points with a latency of 4 to 7 ms; in our experiments the latency never exceded 2.5 ms indicating an oligosynaptic pathway between NVmt and phrenic nerve nucleus. On the other hand, the vagal input to NVmt is obviously polysynaptic, since we were not able to demonstrate any effect of a single stimulus applied to the vagus nerve on the activity of NVmt unit and longer stimulation (over one or two respiratory cycles) was necessary to elicit changes in its pattern of firing. This might be regarded as another indication of NVmt involvement in the respiratory neuronal network. Mechanisms and pathways of this vagal-to-trigeminal connections remain to be elucidated. Another question we cannot answer at present is whether the NVmt-to-phrenic pathway is direct or involves also the premotor medullary neurones. Judging from the latency alone, one might suspect that there is a direct connection bypassing the medullary respiratory neurones. In this case, the NVmt could be regarded as a relay station of the "behavioural" (cortical) drive. Theoretically, however, one can also assume that these units may play the role of inspiratory "off-switch" neurones, being part of the basic (metabolic) pattern generator (see von Euler, 1986). Their response to vagal stimulation might support this view. One cannot exclude the possibility that structures of the rostral pons (including the NVmt) are integrating suprapontine and metabolic/reflex influences on the respiratory pattern generator. Further experiments shall elucidate this dilemma.

REFERENCES

Bertrand, F., and Hugelin, A., 1971, Respiratory synchronizing function of nucleus parabrachialis: pneumotaxic mechanisms, -I.:_ Neurophysiol., 34: 189.

Merrill, E.G., Lipski, J., Kubin, L., and Fedorko, L., 1983, Origin of the Expiratory Inhibition of Nucleus Tractus Solitarius Inspiratory Neurones, Brain Res., 263: 43.

Cohen, M.I., 1971, Switching of the respiratory phases and evoked phrenic responses produced by rostral pontine electrical stimulation, -I.:_ Physiol., (London) 217: 133.

Euler, C. von, 1986, Brain stem mechanisms for generation and control of breathing pattern, in:"Handbook of Physiology- the Respiratory System, Vol. 2, Control of Breathing", Cherniack and Widdicombe, eds. Chapter 1, Amer. Physiol. Soc., Washington, D.C.

St.John, W.M., and Bledsoe, T.A., 1985, Comparison of respiratoryrelated trigeminal, hypoglossal and phrenic activities, Resp. Physiol., 62: 61.

Gromysz, H., Kukwa, A., Jernajczyk, U., Karczewski, W.A., 1988, Studies on the mechanisms of obstructive sleep apnea.- in the press.

186

LATERALIZED RESPONSE OF THE HYPOGLOSSAL, FACIAL AND PHRENIC

NERVES TO LUNG INFLATION

Wiktor A. Janczewski

Department of Neurophysiology, Medical Research Centre Polish Academy of Sciences, 3 Dworkowa str., Warsaw, Poland

INTRODUCTION

For several animal species it is well known that, after the lungs are prevented from collapsing at the end of inspiration or while they are inflated by positive pressure, pattern of breathing markedly changes -inspiratory activity diminishes and expiratory time increases. Magnitude of these changes rises with increasing tidal volume. Volume of air in the lungs is sensed by the slowly adapting pulmonary stretch receptors (PSR). The information arising from each lung is transmitted to the brainstem solely via the vagus nerve ipsilateral to the lung ( Guz et al., 1966; Cross et al., 1981). It is not known how volume information arising from each lung individually and reaching the brainstem by two separate afferent pathways summates and converges onto a common output. The aim of this study was to assess whether, after lateralization of the PSR inputs by denervation of one lung, activity of respiratory nerves on both sides of the body will be equally inhibited during maintained lung inflation or whether lateralized input will result in a lateralized response.

METHODS

Animal Preparation

Twelve cats and fourteen rabbits were anesthetized with halothane (0.9-1.5 vol%), paralyzed and mechanically ventilated. Endotracheal pressure, arterial blood pressure, body temperature and end-tidal C02 were monitored. Arterial blood gases and pH were estimated using a Corning 175. Phrenic nerve (PhN), superior and inferior bucca-labial branches of the facial nerve (VIIN), and hypoglossal nerve (XIIN) were dissected bilaterally, cut and prepared for recording. Recurrent laryngeal and superior laryngeal nerves were cut bilaterally. One vagus nerve was blocked by 1% lignocaine or cut. The other vagus nerve was left intact. The electrical activities of PhN, XIIN, VIIN on both body sides were processed to obtain a moving average (time constant 100 ms) and simultaneously recorded andjor sampled on-line every 10 ms by a special purpose computer Anops 101. In six additional rabbits only PhN activity was processed and bilaterally recorded as described above. PSR afferents were activated unilaterally by: a) maintained inflation (Itest) produced by tracheal occlusion at peak inflation, or by applying

187

Left vagus cut

TP l cmH.p

84/. 100% 55 % • • •

Left l X liN

Right] X liN

97 % • 100% 86%

• •

s. Fig. 1. Asymmetrica l changes of the XII'\ activity in the

response to lung inflation (I-test) after section of the left vagus nerve. Note that during inflation amplitude of the left XIIN drops to 55'1. of the control value while amplitude of the right XIIN only to 86'1. of the control. Asymmetry index amounts 86'1.-55'1. = 31'1.. Traces from top to bottom: endotracheal pressure, integrated activities of the left XIIN, right XIIN, and right PhN.

constant positive pressure (8-15 em H20) at t he end of the inspiratory phase in animals with one vagus nerve blocked by lignocaine or cut ; b) electrical stimulation (100 Hz , 0.5 ms square pulse , 17-25 ,uA) of one vagus nerve (IE-test ) in bilaterally vagotomi zed an imals . Stimulator (NL-300, NL-510, . NL-800- Digitimer Ltd.) was triggered at the beginning of the expiratory phase. Lungs were kept expanded or stimulation continued for 1-4 central respiratory cycles.

Data Analysis

The activity of all nerves was quantified by measuring the peak amplitude of the integrated signal s . The nerve activity during lung inflation (or electrical stimulation) was expressed as percent of t he control activity. This was calculated separate ly for both sides of the body and subtracted. The r esult was named asymmetry index (AI). It expressed the difference in the suppression of the nerve activity during I-, IE-test on the ipsilateral (to the intact vagus) and contralateral

188

side of the body. AI was equal 0% for totally sy~netrical responses. For animals with one vagus nerve intact the control amplitude was a mean of 2-3 amplitudes measured while lungs were kept at the functional residual capacity level. In the bivagotomized animals control amplitude was a mean of 2-3 amplitudes measured just prior to the stimulation. To investigate the interaction of chemical drive with the effects of volume related feedback I- and IE-tests were repeated 8 to 15 times at different levels of chemical drive. The magnitude of suppression of the nerve activity and AI were related to the levels of arterial PaC02 and Pa02. Data were analysed in two groups: a) at low chemical drive - PaC02 of 30-45 mmHg in cats, and 20-35 mmHg in rabbits, Pa02>90 mmHg; b) at high chemical drive PaC02 of 45-65 mmHg in cats and 35-55 mmHg in rabbits, Pa02 of 60-80 mmHg. Pressure for inflation and current for stimulation was chosen to produce maximal asymmetrical effects - maximum AI. They were ty~ically just above threshold for suppression of the activity of the studied nerves and amounted 8-15 em H20 and 17-25 uA respectively. Such thresholds are known to be different for XIIN, VIIN and PhN (van Lunteren et al., 1986); in this study it was observed that they varied also with the level of chemical drive.

RESULTS

XIIN and VIIN exhibited biphasic respiratory modulated activity which peaked during inspiration. I-, IE-tests resulted in transient inhibition (Breuer-Hering inspiration-inhibiting reflex) of the inspiratory activity of all nerves. After a period exceeding control expiratory time, activity of the nerves "escaped" from total inhibition and inspiratory activity reappeared. Its amplitude, duration and frequency were reduced in comparison with control inspiratory volleys. The first expiratory period after onset of the stimulation was usually the longest one and the first volley had the smallest amplitude. Then, although stimulation of PSR afferents was constant, both amplitude and time of expiration were gradually returning toward control values. Comparison of neurograms recorded simultaneously from the left and right side during I- or IE-test revealed that activity on the side ipsilateral to the cut or blocked vagus was depressed significantly more than on the side of active or stimulated vagus (Fig. 1). This as~etry was seen in the shape and peak amplitude of neurograms but not in the inspiratory and expiratory time (Figs. 1, 2, 3).

Asymmetry indexes were calculated from 154 I-, IE-tests separately for cats and rabbits at low and high chemical drive (Table 1). Results suggested that: 1) asy1runetry in the amplitude response of the XIIN, VIIN and PhN increased with decreasing chemical drive - for all nerves AI at

Table 1. Asymmetry indexes (means ± SD (%))

XIIN XIIN VIIN VIIN PhN CAT RABBIT CAT RABBIT RABBIT

LOW CHEMICAL 15.8 29.7 21.6 34.9 9.2 DRIVE +4.3 +4.9 +3.1 +14.2 +6.0

HIGH CHEMICAL 9.8 19.8 13.3 22.9 close DRIVE +6.0 +4.9 +5.2 12.3 to 0%

189

:.'N-:'. A4 •1111 •

.,..,~· ... __ ..... - . •, .·· . .. .... , ' _, ~ •y • /: ·, -:: -.-.,.,,. .\·. . . '• .. " .... ..

-~·· .. ~ IE-TEST 0.25

15

Fig. 2. Asymmetrical effect of unilateral stimulation of one vagus nerve (IE-test) on VIIN activity in a rabbit. Left side upper trace; superimposed moving averages of t he left and right VIIN. Left side lower trace; an ar ithmetic difference of both moving averages. Arithmetic difference which is close to zero before stimulation increases considerably during stimulation. Right panel presents superimposed asymmetrical volleys photographed at faster time base.

high and low chemical drive differed significantly (p<0.05 , t-test); 2) asymmetry was more pronounced in rabbits than in cats; 3) asymmetry was more pronounced for VIIN than for XIIN. Finding that lowered inspiratory drive promoted lateralized responses prompted reinvestigation of the problem with respect to PhN under such experimental conditions. This was done on 6 additional rabbits. EI-test was repeated 20-30 times in each animal at different levels of chemica l drive. Asymmetrical responses, the same as those observed for XIIN and VIIN were seen only in hypocapnia (Fig. 3). Asymmetry on PhN as well as on XIIN and VIIN neurograms was more pronounced after an increase of the level of anesthesia with i. v . small single dose of: propanidid 15 mgjkg, ketamine 1-2 mgjkg, nembutal 10 mgjkg. Propanidid was found to be the most appropriate since it eliminated tonic PhN activity often observed in hypocapnia, and influenced respiration no l onger than 4-8 min.

DISCUSSION

This study demonstrated that during unilateral activation of the PSR afferents, the activity of the XIIN and VIIN on the stimulated side was significantly less diminished than on the contralateral side . The difference in the magnitude of the response between t he two sides increased with decreasing arteri al C02 l evel . In hypocapnia asymmetrical responses could also be seen in the PhN act ivity in rabbits.

Vagally mediated volume feedback is known to inf luence respiratory activity by two mechanisms (Younes et al., 1978, Euler, 1986 for review). The first operates in an all - or-none fashion ("off- switch") and when activated terminates inspiration. The second inhibits inspiratory activity in a graded and reversible manner. The first mechanism could not be responsible for the difference in the amplitudes reported in this study , since the onset and termination of inspi ration was perfectly synchronous on both body sides. What differed was the rate of rise of t he inspiratory activity. It is lik~ly that this difference was due t o a greater magnitude of graded inhibition on the side contralater a l to the intact vagus nerve . This study confirmed t he findings of Younes et al. (cf. Kuna, 1986) that the activity of the inspiratory network may be

190

a b a b

IE-TEST 15

15

Fig. 3. Asymmetrical effect of unilateral stimulation of one vagus nerve (IE-test) on PhN activity. Left side: upper trace - superimposed moving averages of the left and right PhN (activity of the nerve contralateral to stimulated vagus is suppressed more); lower trace - result of digital subtraction of the above presented signals. Note that before IE-test signals were nearly equal, so that the result of subtraction is close to zero. During IE-test the difference is considerably bigger. Right panel - volleys from the left panel photographed at faster time base.

inhibited in a graded manner before activation of the "off-switch" mechanism and additionally demonstrated that: a) this "direct" inhibitory influence was state-dependent, being most pronounced when inspiratory tone was low and PSR activity high; b) graded inhibition usually depressed XIIN and VIIN activity from the very beginning of inspiration (van Lunteren et al., 1984; Kuna, 1986) (Fig. 2). This was also true with respect to PhN in hypocapnia (Fig. 3).

Factors promoting observation of the "direct" component were those which considerably diminished respiratory activity (i.e. anesthesia, drugs, hypocapnia) but had relatively smaller effect on PSR activity. Possibly that was why in hypocapnia the asymmetry was more pronounced and suppression of the XIIN, VIIN and PhN activity during I- and IE-tests was stronger than in normocapnia.

Previous studies on lateralization of the response to stimulation of the respiratory afferents (Berger and Mitchell, 1976; Bruce et al., 1982) dealt only with PhN activity and employed signal-averaging technique. Berger and Mitchell (1976), observed short-latency contralateral PhN excitation following electrical stimulation of a pharyngeal branch of the glossophar~1geal nerve and an internal branch of the superior laryngeal nerve. On the basis of these findings they suggested paucisynaptic (likely disynaptic) neural circuit between cranial nerve afferents and contral ateral spinal respiratory motoneurons.

It is concluded that the over-all response in PhN activity to lateralized PSRs-originating afferent information is typically symmetrical, presumably because information from PSRs from both lungs is highly integrated centrally . PhN amplitudes on each side of the body differ only under certain experimental conditions, i.e. hypercapnia and deeper anesthesia. These results, together with the findings of Cross et al. (1981) suggest that unilateral PhN recordings are sufficient to assess the bilateral PhN response in the experiments in which PSRs information is asymmetrical. However, similar approach for studying XIIN and VIIN activity may lead to substantial quantitative errors, since the drop in the activity on the stimulated side is smaller than on the other side.

191

REFERENCES

Berger, A.J., and Mitchell, R.A., 1976, Lateralized phrenic nerve responses to stimulating respiratory afferents in the cat, Am. ~ Physiol., 230(5): 1314.

Bruce, E.N., Euler, C., Romaniuk, J.R., and Yamashiro, S.M., 1982, Bilateral reflex effects on phrenic nerve activity in response to single-shock vagal stimulation, Acta Physiol. Scan., 116: 351.

Cross, B.A., Guz, A., and Jones, P.W., 1981, The summation of left and right lung volume information in the control of breathing in dogs, ~ Physiol., 321: 449.

Euler, C. von, 1986, Brain stem mechanisms for generation and control of breathing pattern, in: "Handbook of Physiology, The Respiratory System, Control of Breathing", Vol. II, chapt. 1, N.S. Cherniack and J.G. Widdicombe, eds., Am. Physiol. Soc., Bethesda, MD.

Guz, A., Noble, M.I.M., Trenchard, D., Smith, A.J., and Makey, A.R., 1966, The Hering-Breuer inflation reflex in mru1: studies of unilateral lung inflation and vagus nerve block, Respir. Physiol., 1: 382.

Kuna, S.T., 1986, Inhibition of inspiratory upper airway motoneuron activity by phasic volume feedback, ~ ~ Physiol., 60: 1373.

Lunteren van, E., Strohl, K.P., Parker, D.M., Bruce, E.N., van de Graaff, W.B., and Cherniack, N.S., 1984, Phasic volume- related feedback on upper airway muscle activity,~~ Physiol., 56: 730.

Younes, M.K., Re~ners, J.E., and Baker, J., 1978, Characteristics of inspiratory inhibition by phasic volume feedback in cats, ~ ~ Physiol., 45: 80.

192

LARYNGEAL CONTRIBUTION TO RESPIRATORY PATTERN IN

ANESTHETIZED RABBITS

Beata Kamosinska and HaJgorzata Szereda-Przestaszewska

Department of Neurophysiology, Polish Academy of Sciences t-ledical Research Centre, Warsaw, 3 Dworkowa str., Poland

There have been a number of animal and human studies regarding the larynx as an effector organ of ventilatory control (Brancatisano et al., 1983; England et al. 1982; Gautier et al., 1973; Remmers and Bartlett, 1977). These experiments were based on the normal route of airflow through the upper airways. The purpose of the present study was to delineate the role of the larynx as the flow limiting segment and presumable respiratory timing controler. We have compared the effect of breathing via tracheostomy or via larynx (supralaryngeal airway being removed from the airflow path) on respiratory airflows and timing in spontaneously breathing rabbits. The measurements were made in control conditions and after section of the laryngeal nerves and cervical vagi.

METHODS

Studies were performed in fourteen adult rabbits, anaesthetized with a mixture of urethane and chloralose. Femoral arterial pressure was measured by means of strain gauge. Body temperature was maintained at 37 + 1°C. The superior laryngeal nerves (SLNs), the recurrent laryngeal nerves (RLNs) and the cervical vagi were encircled with loose ligatures. Tracheostomy was performed and T-shaped cannula was inserted. A second T-shaped cannula was fixed within the oropharynx, after drawing aside the epiglottis. The rostral end of this T-tube was connected to the endotracheal tube, which was clamped, so the airways from the oropharynx up to the mouth were bypassed and the animal breathed through the side arm of the oropharyngeal tube (Fig. 1). The latter was connected to a Fleisch pneumotachograph head and pneumotachometer (Medipan type 351) and to a differential micromanometer (Hilgier IRD). The other side of the micromanometer was connected to the side-arm of the tracheal T-tube, which was otherwise closed. This created the "larynx in" condition. During breathing with omission of the larynx ("larynx out"), the rostral end of the tracheal tube was clamped and its side-port was connected to the pneumotachograph and micromanometer. Action potentials from a C3 phrenic root were amplified and integrated. Recordings of the respiratory variables with the larynx in and out of the circuit were made in control conditions, following laryngeal denervations (SLNs and RLNs cut) and after subsequent midcervical vagotomy, in the sequence in each rabbit. All records were taken 20 min after section of each pair of nerves. Amplified recordings of airflow and integrated phrenic electroneurograms were subjected to computer analysis. From

193

t-he averaged phrenic neurogram inspiratory and expiratory durations (Tr and TE), breathing frequency and peak phrenic amplitude were computed. All results were calculated as means and standard deviations. Data were analysed by a paired t-test, p<0.05 was deemed significant.

Fig. 1. Schematic representation of the experimental setup. Clamping the rostral end of the lower T-tube created the condition of the larynx out of circuit.

RESULTS

Respiratory Airflows

The results are presented in Fig. 2. In contro l conditions the presence of the larynx compared with its omission from the circuit significantly decreased peak inspiratory and expiratory airflows. Section of the superior laryngeal nerves did not significantly modify control values or the effects of putting larynx in and out of circuit. Paralysis of the vocal cords due to RLNs section significantly lowered the airflows when the larynx was in circuit. Midcervi cal vagotomy nonsignificantly increased respiratory airflow at tracheostomy and subsequent connection of the larynx reduced them to a lesser degree than at the previous experimental step (RLNs section).

Respiratory Timing

The results are shown in Fig. 3. There was no significant difference in inspiratory and expiratory durations or in breathing frequency between t he two routes of respiration in cont rol conditions and following sensory denervation of the l arynx. Breathing t hrough the deeferented larynx significantly prolonged inspiratory and expiratory times, so that breathing frequency decreased. This e ffect was due to mechanical obstruction of the vocal tract, as i t was absent on tracheostomy breathing. The respiratory timing was apparently lengthened by midcervical vagotomy and the presence of the inactive larynx did not affect either inspiratory or expiratory durations.

194

DISCUSSION

V (1/min) 1n *** *** 2.50 *

2.00

150

1.00

ot

2.50

2.00

1.50

1.00

ol.

Fig. 2.

V (1/min) ex

**

control

c:=J larynx OUT

f::2:l lory nx IN

** **II- ***

Vagi-cut

*** p<0.005 * * p< 0.05

* P< 0.02

Effect of airways' denervations on inspiratory (Vin) and expiratory (Vex) airflow during tracheostomy (OUT) and laryngeal breathing (IN). Bars represent mean values± SD.

Present experiments, as initially reported (Kamosinska and SzeredaPrzestaszewska, 1986) brought the corroborative evidence for laryngeal limitation of respiratory airflows. This is in line with the results of Citterio et al.(1985) and Brancatisano et al. (1987) who described the decrease in mean inspiratory and expiratory airflows on transition from tracheostomy to upper airways breathing in rabbits and dogs. Our results show that in this respect the laryngeal valve acts as a throttle, barely but significantly restraining airflows when patent and substantially more reducing them when inactive due to motor denervation.

Diversion of the flow from tracheostomy to the larynx did not change the duration of the respiratory phases and the frequency of breathing. This is consistent with the results of Sant'Ambrogio et al.(1985) described in the dogs breathing through the larynx and intubated upper airways. In cont rast, all studies, in which the airflow was diverted onto the upper airways indi cated a slowing down of respiration (Brancatisano et al., 1987; Citterio et al., 1985; Gautier et al., 1973). It depends likewise upon highly resistive nasal and oral compartments (Ferris et al., 1964; Ohnishi and Ogura, 1969). In our experimental model only breathing through the paralysed larynx in animals with spared vagi affect ed the breathing rate.

195

REFERENCES

1.00

080

0.60

0.40

ol

2.00

1. 80

1.60

1.40

1. 20

o l

T I (sec)

[] [] TE (sec)

cont r o l SL N5-cut

c::J larynx OUT

E:Z.I la r ynx IN

*

cl *

RLN5 cul Vag i-cut

* p < 0.005

Fig. 3. Changes in inspiratory (Tr) and expiratory (TE) durations during tracheostomy (OUT) and laryngeal (IN) breathing in the course of successive denervation of airways.

Brancatisano, T.P., Collet, P.W., and Engel, A., 1983, Respiratory movements of the vocal cords,~~ Physiol., 54: 1629.

Brancatisano, A., Kelly, S.M., Tully, A., and Engel, L.A., 1987, Effect of respiratory glottic constriction on lung volume and pattern of breathing in adult dogs, Respir. Physiol., 67: 53.

Citterio, P.C., Mortola, J.P., and Agostoni, E., 1985, Reflex effects on breathing of laryngeal denervation, negative pressure and so2 i n upper airways, Respir. Physiol., 62: 203.

England, S.J., Bartlett, D.,Jr., and Daubenspeck, J.A., 1982, Influences of vocal cord movements on airflow and resistance during eupnea, ~ ~ Physiol., 52: 773.

Ferris, B.G., Mead, J., and Opie, L.H., 1964, Partitioning of respiratory flow resistance in man, ~ ~ Physiol., 19: 653.

Gautier, H., Remmers, J.E., and Bartlett, D.,Jr., 1973, Control of the duration of expiration, Respir. Physiol., 17: 205.

Kamosinska, B., and Szereda-Przestaszewska, M., 1986, Laryngeal patency and its ventilatory effects in anaesthetized rabbits, ~ Physiol. (Loud.), 380: 44P.

Ohnishi, T., and Ogura, J.H., 1969, Partitioning of pulmonary resistance in the dog, Laryngoscope, 79: 1847.

Remmers, J.E., and Bartlett, D.,Jr., 1977, Reflex control of expiratory airflow and duration,~~ Physiol., 42: 80.

Sant'Ambrogio, F.B., Mathew, O.P., Clark, W.D., and Sant'Ambrogio, G., 1985, Laryngeal influences on breathing pattern and posterior cricoarytenoid activity, ~ ~ Physiol., 58: 1298.

196

PROPRIOSPINAL INSPIRATORY NEURONS IN THE UPPER CERVICAL SPINAL

CORD OF THE RABBIT: LOCATION AND EFFERENT SPINAL PROJECTIONS

INTRODUCTION

Leszek Kubin and JarosYaw R. Romaniuk

Department of Neurophysiology, Medical Research Center Polish Academy of Sciences, Warsaw, Poland

A new group of inspiratory neurons was recently discovered in the upper cervical spinal cord of the cat (Aoki et al., 1980; Aoki, 1982). The group became a subject of several further studies as it could have represented an important relay and processing site interposed between the respiratory neurons of the brain stem and spinal respiratory motoneurons. It has been established that the cervical neurons form a well defined longitudinal column which extends from C1 to C3 cervical segments and send descending projections which may reach lower thoracic segments (Aoki et al., 1984; Lipski and Duffin, 1986). Although a monosynaptic connections of these neurons with either phrenic or external intercostal motoneurons were rather uncommon (Lipski and Duffin, 1986), the neurons still seemed potentially capable of providing indirectly a substantial drive to spinal respiratory motoneurons.

The studies done so far on the upper cervical inspiratory neurons were performed in the cat. However, experiments in spinal cats and rabbits show that a rhythmic activity may be obtained in spinal respiratory motoneurons of both preparations (Aoki et al., 1980; Viala and Freton, 1983). Moreover, sections and lesions in the medulla have a moderate effect on the respiratory output at the spinal cord level in the rabbit as compared to the cat (Gromysz and Karczewski, 1981; Speck and Feldman, 1982; Budzinska and Romaniuk, 1986; Kubin et al., 1987). These findings suggest that inspiratory neurons of the upper cervical spinal cord may be an important component of a spinal respiratory generator. Thus, the present study was undertaken in order to search for respiratory neurons in the upper cervical spinal cord of the rabbit which would exhibit properties similar to those of the neuronal group described earlier in the cat.

METHODS

Experiments were performed on 13 rabbits weighting 2.5-3.7 kg, anesthetized with sodium pentobarbital (35 mgjkg, i.v., supplemented every hour by 3-5 mgjkg injections), thoracotomized, paralyzed and artificially ventilated.

197

The left C5 phrenic root was prepared for recording. The animals were fixed in a stereotaxic head holder and a vertebral clamp at Th3 level and the whole dorsal surface of the cervical spinal cord was exposed. Activity of single neurons was recorded on the left side with tungsten microelectrodes (NL-05, Neurolog). A second tungsten microelectrode was introduced into the lower cervical segments, and, in one experiment into the three lowest thoracic segments, for monopolar microstimulation (NL-800 stimulus isolation unit, Neurolog). Current pulses of 0.1 ms duration and up to 120 uA strength were applied. Phrenic nerve activity was integrated with a time constant of 0.1 s while the unit activity was first fed to a discriminator (NL-200, Neurolog) and it's impulses were integrated to obtain a firing rate record.

Blood pressure, end tidal C02, and body temperature were continuously monitored and kept within physiological l imits. Selected recording sites were marked by small e l ectrolytic les ions . The tissue was cut and processed by standard methods.

RESULTS

Location of Inspiratory Neurons

Guided by earlier studies in the cat, we searched for inspiratory neurons in the three uppermost cervical segments. Closely spaced penetrations were distributed along a 1.2 mm wide column centered around the dorsal roots entry line. Twenty nine penetrations were made in the C1 segment in two animals, 120 penetrations were placed in the C2 segment in 11 animals, and 40 in the C3 segment in five animals .

Inspiratory ce lls were found cons istent ly only in the C2 segment. Thirty six such single cells wer e recorded . They wer e locat ed just medial to dorsal roots, at depths from 1.3 to 2.0 mm from the dorsal surface of the cord. Recordings were made from cell bodies rather than from fibers as suggested by a triphasic shape of action potentials, stability of recordings, and a phasic shortening of antidromic latency in inspiration by as much as 0.25 ms (see below). The ce lls displayed a phasic, increasing in a ramp-like fashion , discharge t hat commenced just prior or just after the phreni c nerve activity and ended synchronously with the off-switch of the phrenic nerve burst. No obvious modulation

INT. PHRENIC

UNIT FIRING RATE

A

TRACHEAL PRESSUREL/1

I ~OHz s

'-----~Jl/V~::. I 0 cmH-:20

5 sec.

Fig. 1. Example of t he firing pattern (A) and location within the C2 segment of propriospinal inspiratory neurons (B). INT. PHRENICintegr ated a ctivity of the phrenic nerve.

198

A

~ 6 +---------~~._-------.~--------_,

% 5 +----------~~~-------------------4 => 0 4 t----a-u ] +-------11!1!

2

17.0 24 .0 J 1.0 Jll 0

CONDUCTION VELOCITY (m/s)

B

___. 1mm

c

"" c

Collision lesl

i: fj ga t _., :: S l U1,

t

450

Fig. 2. Properties of spinal axons of the cervical inspiratory neurons. A - histogram of conduction velocities. B- location of five descending axons. C - collision test for one cell. Collision occurs in bottom traces, when the stimulus is applied with a shorter delay after a spontaneous triggering spike. Time calibration- 1 ms .

of the cell firing with pump-induced lung inflations was observed. When the respirator was t emporarily stoped, the ce ll firing increased in parallel to the phrenic nerve activity (Fig. l.A). Figure 1 B shows the location of inspiratory ce lls in the C2 segment. The ce lls were found just lateral to the boundary of the gray matter at sites where usually a few scattered, medium size, cell bodies were present on serial sections.

In the C1 segment, activity of phasic expiratory cells was dominant while inspiratory units, although occasionally encountered, had in most cases small, short lasting, and monophasic spikes which suggested their origin in fibers rather than ce ll bodies. No inspiratory ce ll s wer e found in the c3 segment.

Spinal Pro jection of Cervical Inspiratory Neurons

Eleven inspiratory ce ll s (all tested) could be excited antidromically frQm t he ipsi lateral white matter of the C6 segment or below (including one activated from the Th12 segment). Axons of five of these ce lls and

199

of 13 others were also stimulated in c4-cs segments where many penetrations were made at different medio-lateral coordinates in an attempt to detect the presence of some branches of the main axon or its crossing towards the opposite of the cord. In no instance a distinct ramification of the main axon could be detected within the C4 or Cs segments. The mean conduction velocity of the axons was 23.0 m/s + 7.0 (SD) (Fig. 2.A). Descending axons of the cervical inspiratory cells were localized within the vetro-lateral or ventral quadrants of C4-C6 segments (Fig. 2.B). No positive cross-correlations were found between the neuronal spikes (triggers) and the phrenic nerve activity for five neurons submitted to such tests.

DISCUSSION

The present study shows that, similarly to the cat, inspiratory propriospinal neurons are present in the upper cervical spinal cord of the rabbit and that their axons descend at least to the thoracic spinal cord. The location of these neurons in the rabbit is, however, limited to the c2 segment.

As in the cat, descending axons of the rabbit's propriospinal neurons do not seem to affect directly (monosynaptically) phrenic motoneurons nor is likely their action on interneurons of the C3-C5 segments which in the rabbit contain phrenic motoneurons (Rikard-Bell and Bystrzycka, 1980). Thus, one may speculate that their spinal action is limited to thoracic respiratory motoneurons andjor even lower levels. That may mean that the cervical neurons have a restricted and specific target of action rather than a widespread effect on spinal network of respiratory neurons which, in turn, would cast doubts on their importance for generation of a respiratory rhythm in spinal preparations.

Propriospinal inspiratory neurons, at least those in the cat, do not seem to act monosynaptically on external intercostal motoneurons (Lipski and Duffin, 1986). Their possible action at the thoracic level may be important for control of respiratory interneurons (cf., Kirkwood et al., 1987). They may, however, be also involved in synchronization of respiratory and locomotor activities (Viala and Freton, 1983) or even in relaying an inspiratory drive to sympathetic preganglionic neurons. Although a relatively high conduction velocity of their axons makes the latter hypothesis unlikely, the anatomical location of the cell bodies, analogous to the location of the sympathetic preganglionic neurons, favours such a possibility.

We think, that determination of the target of the cervical propriospinal neurons is most important for understanding their functional significance.

ACKNOWLEDGEMENTS

The authors thank Professor W.A.Karczewski for making the study possible and for the supportive atmosphere. The technical assistance of Ms. K. Sroczynska is greatefully acknowledged. The study was supported by grant No. 06.02.III.1.1. from the Polish Academy of Sciences.

200

REFERENCES

Aoki, M., Mori, S., Kawahara, K., Watanabe, H., and Ebata, N., 1980, Generation of spontaneous respiratory rhythm in high spinal cats, Brain. Res., 202: 51.

Aoki, M., 1982, Respiratory related neuron activities in the cervical cord of the cat, in: "Proceedings of the International Symposium. Central Neural Production of Periodic Respiratory Movements", J.L. Fe 1 dman and A. J. Berger, eds. , Lake Bluff, Illinois.

Aoki, M., Kasaba, T., Kurosawa, Y., Ohtsuka, K., and Satomi, H., 1984, The projection of cervical respiratory neurons to the phrenic nucleus in the cat, Neurosci. Lett., Suppl. 13: S9.

Budzinska, K., and Romaniuk, J.R., 1986, Phrenic reflexes in the decerebrate and spinal rabbit, Bull. Eur. Physiopathol. Respir., 22: 65.

Gromysz, H., and Karczewski, W.A., 1981, Respiratory activity generated by a split brainstem preparation of the rabbit, Acta Neurobiol. Exp., 41: 237.

Kirkwood, P.A., Munson, J.B., Westgaard, R.H., and Sears, T.A., 1987, The organization of the respiratory input to intercostal motoneurones: the contribution from interneurones? in: "Respiratory Muscles and Their Neuromotor Control", G.Sieck, ed., A.R. Liss Inc., New York.

Kubin, L., Lipski, J., and Trzebski, A., 1987, Respiratory rhythmicity in a split medulla preparation of the cat, Exp. Neural., 96: 720.

Lipski, J., and Duffin, J., 1986, An electrophysiological investigation of propriospinal inspiratory neurons in the upper cervical cord of the cat,~ Brain Res., 61: 625.

Rikard-Bell, G.C., and Bystrzycka, E.K., 1980, Location of phrenic motor nucleus in the cat and rabbit studied with horseradish peroxidase, Brain Res., 194: 479.

Speck, D.F., and Feldman, J.L., 1982, The effects of microstimulation and microlesions in the ventral and dorsal respiratory groups in the medulla of the cat,~ Neurosci., 2: 744.

Viala, D., and Freton, E., 1983, Evidence for respiratory and locomotor pattern generators in the rabbit cervico-thoracic cord and for their interaction, Exp. Brain Res., 49: 247.

201

SPECTRAL ANALYSIS OF BREATHING PATTERN IN MAN DURING EXERCISE

INTRODUCTION

Jarmila SiegelovA and Sylvie FeitovA

IIIrd Department of Medicine, Medical Faculty, University J.E. Purkyne, Department of Functional Investigations KUNZ, Brno, Czechoslovakia

In recent years analysis of the breathing pattern parameters in various conditions was used to study respiratory control in man. The findings of a significant positive correlation between tidal volume and inspiratory time and the weaker positive correlation between inspiratory time and expiratory time, as well, as between tidal volume and expiratory time were described as a result of a mechanism which controls the ventilation in every single breath (Newsom Davis and Staag, 1975). These correlations could be alternatively explained by slow synchroneous oscillations of breathing pattern parameters studied in our laboratory (SiegelovA and Kopecny, 1985). The aim of the present study was to follow the slow waves of breathing pattern parameters during exercise and also to evaluate the role of both control mechanisms - a single breath regulatory mechanism and slow oscillations of breathing pattern.

METHODS

We examined 10 healthy men at rest and during bicycling (work load ranging from 50 to 75 Wand from 100 to 200 W). We analysed 4 minutes records (Siregnost) breath by breath in steady state at both levels of exercise and at rest, and we computed autocorrelation functions of inspiratory time, expiratory time, tidal volume and crosscorrelation functions. We also computed power spectral density of the various breathing pattern parameters. Power spectral density gives a more exact description of the periodicity or periodicities in an analysed signal. We computed it from autocorrelation functions modified by a Hanning spectral window (Blac~nann and Turkey, 1959). We computed the crosscorrelation functions because we can, from the curve of crosscorrelation function, determine if the crosscorrelation function between breathing pattern pareameters is the result of a single breath regulatory mechanism - in that case the crosscorrelation function has a spike at zero time unit (lag), the other crosscorrelation coefficients are very small, or if it is the result of synchroneous breathing pattern oscillations - in which case the crosscorrelation function is positive but lower and flatter.

203

RESULTS

The auto~orrelation functions of inspiratory time, expiratory time and tidal volume showed rhythmical changes in various parameters at rest and during both levels of exercise. Rhythmic changes in inspiratory time, expiratory time and tidal volume in the same subject differed at rest and during bicycling, and lasted several respiratory cycles.

n•10

J ,...._ I'· Ti/TE

,,..· .. J • ..... _____ _,.,

.-'/ '·· I -.

J I v.Jrl \ -t ; . _,/ .

~ ... ··· /· .····· .. --

J "rfTE ! ......

....-.-::::-::-

·20 0 20s~ 0 20s~O 0 20s

ow 50·75 w 100·200W

Fig.l. Mean crosscorrelation functions between inspiratory time and expiratory time (TriTE) in the upper part; between tidal volume and inspiratory time (Vr/Tr) in the middle part; between tidal volume and expiratory time (Vr/TE) in the lower part - at rest (OW), and during exercise. The abscissa shows the lag in seconds, the ordinate - the value of crosscorrelation coefficients.

The crosscorrelation functions between inspiratory time and expiratory time, between tidal volume and expiratory time, between tidal volume and inspiratory time were different in every individual. Therefore we calculated the mean values of crosscorrelation functions between various breathing pattern parameteres as shown in Fig.l.

At rest we can see that only mean curve of the crosscorrelation function between tidal volume and inspiratory time has a peak at zero lag, which could be explained by the single breath regulatory mechanism, but the curves of crosscorrelation functions between inspiratory time and expiratory time and between tidal volume and expiratory time could be the result of slow oscillations of breathing pattern parameters of both studied values. The crosscorrelation functions are changed at both levels of exercise, at zero lag we can observe the spikes in every studied crosscorrelation functions. We suppose that these spikes could

204

be caused by a single breath regulatory mechanism, which plays a role in all studied interrelationships during exercise.

Power spectral densities of inspiratory time, expiratory time and tidal volume showed the presence of different frequencies in the same individual at rest and on both levels of exercise. The frequency peaks

no10

0.5 ~ ~

........ ··. 0~-·--_:·~-:~-"~ .. ~·==~ ~--··--~ .. :~-- -~~--~· .. ~·--=·=--~

o.s

• .. 0 .... '· .. , .. . ......... .

_ ... __ . -~-- ... -:-...... __ __ ,._

0.5 VI

\ 0 ·····,,. :. __ ...... ..

\' "···· "· .. : ... : ... ::: __ ; __ ':' __ :_

o.s

\' ·--·-·· -.. - .

·•;,,•• - •• ~ ••• T ••.• -;-,,_

o'-----:.::0;:::.1 =~H:-:<z 0 • , ,,,; __ ,,•,J,oo O. o

0.1 SOW ow

·-·~ .. ~ ... : ...... : ... : ... ~. - -

·.

I

\:

Hz 0

· ··•• •• . ,: .•••. •. ; ,,, , .:- • ,r

---·· ·-... ;-... =- - ·~· -·=-

0.1 100W

Hz

Fig.2. Mean power spectra of inspiratory time (Tr), expiratory time (TE), inspiratory tidal volume (Vr), expiratory tidal volume (VE) at rest and both levels of exercise. Abscissa, frequency (Hz), ordinate, relative power. The lines show the standard error of the mean.

were not stable in the same parameter of the same subject at rest and during exercise. The spectrograms of different subjects differed markedly. We, therefore, constructed a mean spectral density of all studied subjects at rest and at both levels of exercise, for all studied parameters as shown in Fig.2.

DISCUSSION

25 years ago Priban (1963) put forward the hypothesis that shortterm changes of breathing pattern parameters presented a picture of the behavior of the central nervous control mechanism actively regulating the depth and rate of breathing for every given ventilation level. Since that time the slow oscillations have been studied in different laboratories with different techniques (Hathorn, 1978; Patil et al., 1986; Modarreszadeh et al., 1987). These oscillations of breathing pattern could not be explained simply by a feedback mechanism, they are different in different breathing pattern parameters in the same individual, they are not stable at rest by repeated measurements and they

205

change in REM and non-REM sleep in newborn infants. Contrary to the unchanged character of oscillations of breathing pattern parameters at rest and during exercise, the interrelationships between volume and time components are changed at both levels of exercise and our results demonstrate that a single breath regulatory mechanism plays an important role in all studied interrelationships.

CONCLUSION

Mean power spectra of inspiratory time, expiratory time and tidal volume had a power which fell at values of 0 to 0.1 Hz and then remained approximately constant at rest and on both levels of exercise. The mean values of crosscorrelation functions between inspiratory time and expiratory time, between tidal volume and inspiratory time, between tidal volume and expiratory time have at both levels of exercise marked correlations between all studied parameters of breathing pattern.

REFERENCES

Blackmann, R.B., and Turkey, J.W., 1959, "The measurement of power spectra", Dover, New York.

Hathorn, M.K.S., 1978, Analysis of periodic changes in ventilation in new-born infants, ~ Physiol., 285: 85.

Modarreszadeh, M., Elhefnawy, A., and Bruce, E., 1987, Components of breath-to-breath variability of respiratory cycle parameters in awake man, Fed. Proc., 46: 658.

Newsom Davis, J., and Staag, D., 1975, Interrelationships of the volume and time components of individual breath in resting man, ~ Physiol., 245: 481.

Patil, C.P., Saunders, K.B., and Sayers, B., Me., 1986, An analysis of the irregularity of breathing at rest and during light exercise in man, I.R.C.S. Med. Sci., 14: 644.

Priban, I.P., 1963, An analysis of some short-term pattern of breathing in man at rest,~ Physiol., 166: 425.

Siegelova, J., and Kopecny, J., 1985, Spectral analysis of breathing pattern in man, Physiol. Bohemoslov., 34: 321.

206

DISTURBED PATENCY OF THE UPPER AIRWAY AND ITS CONSEQUENCES

Henryk Skar±ynskil, Wojciech Jeglinskil, Andrzej Kukwal, Grzegorz Opolski2, Krzysztof SJomka2, MirosJaw Ryba3, and Ret Krauzel

lDepartment of Otolaryngology, 2Department of Cardiology Medical Academy in Warsaw, and 3Department of Neurophysiology, Medical Research Centre Polish Academy of Sciences, Warsaw, Poland

Every ENT (Ear, Nose and Throat) surgeon meets many patients, both adults and children, suffering from disturbances of upper airway patency. Usually the obstruction is at the level of the nose, nasopharynx or pharynx (Guilleminault et al.,1976). Disturbed patency of the upper airway due to the hypertrophy of tonsills or adenoid is discussed in another paper (Skar±ynski et al., this volume). Separate discussion is needed, concerning the problem of nasal airway compromise in patients with sleep apnea syndrome. In this paper we present the consequences of chronic upper airway compromise. A typical patient with mild obstruction of the airflow in the upper respiratory routes presents himself to an otolaryngologist after many years of unsuccessful treatment performed by cardiologists, neurologists or psychiatrists. Not before the authors working on the problem of disturbances of breathing during sleep have proposed a new syndrome in otorhinolaryngology, such patients were fully diagnosed and treated (Simmons and Hill, 1974). The basic examination is an all-night monitoring of many parameters of the blood and functions of the organism- a polysomnographic study (PSG). In patients with upper airway compromise PSG reveals many pathologies in circulatory system function (Orr and Shappell, 1975; Lugaresi et al., 1980) namely the increase in blood pressure and lung circulation, disturbances of the heart rhythm and significant changes in ST - T in ECG examination. Retrospective epidemiologic studies confirmed more frequent pathology in circulatory system in the patients with apneas during sleep. It is possible that sleep apnea may be one of the etiologic factors of the hypertension and of some of the heart diseases (Lugaresi et al., 1980; Miller, 1982).

The aim of this paper is to present the results of investigations performed on patients hospitalized in an intensive cardiologic care unit, who, besides their heart problems, suffered from an upper airway compromise.

METHODS

The material was based on 26 patients treated in an intensive

207

cardiologic care unit. All the patients, besides cardiologic problems had a disturbed patency of the upper airway. In all cases ENT examination revealed the cause of obstruction. In 11 cases it was a deviated nasal septum, in 4 - a hypertrophy of the nasal conchae inferior, in 8 -nasal polyps and in 3 - large tonsills. All the patients underwent a PSG examination and after cardiologic treatment they were submitted to surgery eliminating the obstruction in the airflow. Six months after the operation once again a PSG study was performed on all of them.

RESULTS

The cardiologic problems of our patients were: in 4 cases bradycardia, in 16 cases - disturbances of the heart rhythm (atrioventricular second degree block, ventricular extrasystole), in 7 cases -arterial hypertension. After the acute phase of these pathologies all patients were controlled and the reason of airway obstruction revealed. Each patient underwent PSG studies. In all cases PSG exhibited apneic episodes and in 14 cases the apneas occured in a number justyfing the diagnosis of sleep apnea syndrome (53.8%). After surgery all the patients reported that they "feel better" and especially "emotionally stronger". Six months after surgery all patients underwent another study and a complete cardiologic and laryngologic examination. In all cases a significant improvement in their heart and lung function was noted. None of the patients suffered from hypertension. In four cases some disturbances in the heart rhythm were still present , but only trivial. From the group of 14 patients with the sleep apnea in 4 cases there was no improvement in PSG, but in the remaining 10 there were no apneas at all, or only 5 - 7 per hour. In all cases from the group of 12 patients in whom we did not diagnose sleep apnea there was a significant improvement in the second PSG examination. So, to sum up, no improvement was observed in 15.3% of the whole material and in 29.1% of the group with the diagnosis of sleep apnea.

CONCLUSIONS

In many cases of the chronic upper airway compromise a sleep apnea syndrome develops. Some of these patients are after many years hospitalized in cardiology boards for acute disturbances of the circulatory system. The physical examination is not always capable of revealing a real etiologic factor of their problems. The analysis of our investigations suggests that in such cases an ENT examination and a PSG study may be of great help. In our inconspicuous material the disturbances of the circulatory system were in 84,7% caused by upper airway compromise.

REFERENCES

Guilleminault, C., Tilkian, A.G., and Dement, W.C., 1976, The sleep apnea syndromes, Ann. Rev. Med., 27: 465.

Lugaresi, E., Coccagna, G., and Crignotta, F., 1980, Some epidemiological data on snoring cardiocirculatory disturbances,Sleep, 3: 221.

Miller, W.P., 1982, Cardiac arrhythmias and conduction disturbances in the sleep apnea syndrome, Ann. ~ Med., 73: 317.

Orr, W.C., and Shappell, S.D., 1975, REM sleep and cardiac arrhythmias, Circulation, 52: 519.

Simmons, F.B., and Hill, M.W., 1974, Hypersomnia caused by upper airway obstructions: A new syndrome in Otolaryngology, ~nn. Otol., 83: 670.

208

THE VAGUS NERVE AND THE CONTROL OF BREATHING DURING

POSTNATAL DEVELOPMENT

Daniel Marlot

Laboratoire de Neurophysiologie, Faculte de Medecine AMIENS Cedex, France

In 1930, Coombs and Pike reported that double cervical vagotomy in the kittens affects respiration more than in the adult. They noted that, as in adult, respiratory frequency decreased, but sometimes, the amplitude of the diaphragmatic respiratory movements was less in kittens, often taking on a sobbing form. Moreover, kittens usually survived double vagotomy for only few hours. This observation suggested that vagal afferent information was more important for the control of breathing in the newborn than in the adult. After this original observation, numerous studies have been performed to precise the relative importance of the vagus nerve in newborns of several species. Reviews on this subject, have been recently published (Bruce, 1981; Trippenbach, 1981; Jansen and Chernick, 1983; Fisher an Sant'Ambrogio, 1985; Steele, 1986).

The present report presents histological and physiological data from literature and from my laboratory.

HISTOLOGICAL DATA

At birth, the number of myelinated fibers (MF) in the cervical vagus nerve (CVN) is considerably less than in adult. The MFs represent about 8% (275 fibers) of the adult value in the kitten (Marlot and Duron, 1979) and 18% (653 fibers) in the rabbit pup (De Neef et al., 1982). In the opossum, a marsupial mammal, MFs are missing during the first month of postnatal life and represent only 2% (54 fibers) on the adult value at 35 days (Kraus et al. 1984). During postnatal development, the number of MFs increases at different rates in these 3 species. In rabbit, there is a fast increase of the number of MFs which represents 60% (2137 fibers) of the adult value in 15 day-old animals. In the kitten, at the same postnatal age, the MFs represent only 25% (920 fibers) of the adult value reaching 55% (2017 fibers) at 2 months. In the opossum, the number of MFs is always small in 50 day-old animals (364 fibers = 11% of the adult value). A study performed on human babies (Sachis et al., 1982) shows that counts of total MFs of CVN in term infants are comparable to those in an adolescent group, whereas preterm infants show significantly fewer MFs than full-term babies or adolescents.

209

In newborn kittens, MFs from inferior lary-ngeal nerve represent about 40% of the total number of MFs (300 fibers) of the CVN. This indicates that there are about 180 broncho-pulmonary fibers in the CVN (Marlot, 1982). If we compare these data with adult ones (1300 fibers: Agostoni et al., 1957; 292 fibers: Schwieler, 1968 and 560 fibers: Jammes et al., 1982), the number of broncho-pulmonary MFs varies at birth between 14% and 60% of the adult value.

At birth, in the kitten, the diameter (axon + myelin thickness) of the majority of MFs ranges from 0.5 to 2 pm. The diameter of the largest fibers does not exceed 4 .urn. The growth velocity is about 0.2 .urn daily during the first month of postnatal life. In 2 month-old kittens, all adult diameters are represented in the CVN. This results differs from those obtained in the phrenic or intercostal nerves (Marlot and Duron, 1979a) and in the superior laryngeal nerve (Miller and Dunmire, 1976). Moreover, in the CVN, for a given axon diameter the myelin sheath area is similar in cat (Marlot, 1982) and in rabbit (De Neff et al., 1982), whatever the postnatal age. This differs from the phrenic nerve in which, for a given axon diameter, the myelin sheath area is larger in older animals. As yet, no satisfactory explanation has been found for this discrepancy between these two nerves.

These histological data show that there exists a quantitative and qualitative immaturity of the vagal information in newborn.

PHYSIOLOGICAL DATA

The physiological properties of CVN include the receptor discharge modalities and the reflex effects analysis.

Receptor Discharge

The broncho-pulmonary receptors are divided in 3 main groups: the slowly adapting mechanoreceptors (SAR), the rapidly adapting mechanoreceptors (RAR) and C-fibers including J-receptors. Only the activity of SAR and RAR is modified by respiration and has been recorded in newborn. Data on activity of C-fibers are missing. The actual stimulus for both SARs and RARs is transpulmonary pressure which is lower in newborn than in adult (Fisher and Mortola, 1980).

Newborn has fewer RARs spontaneously active than the adult: 4% vs 15% in the dog (Fisher and Sant'Ambrogio, 1982) and 5% vs 18% in the opossum (Farber et al., 1984). The distribution of RARs in the tracheobronchial tree of the newborn is still unknown. These receptors may play a role in the initiation of the first deep breath after delivery and their activity may be one of the many sources of the irregular pattern of breathing in the neonates. They are also important in the generation of augmented breaths.

Activity of SARs has been well studied in newborn of different species: sheep (Ponte and Purves, 1973), cat (Schwieler, 1968; Marlot and Duron, 1979a; Marlot et al., 1982) and dog (Fisher and Sant'Ambrogio, 1982). Results from these studies show that the percentage of SARs active at functional residual capacity (FRC) is lower in the newborn (O -- 12%) than in the adult (27- 60%). Moreover, the mean transpulmonary pressure threshold required for receptor discharge is higher in the newborn than in the adult and, the discharge frequency of SARs is lower in the newborn than in the adult at any given transpulmonary pressure. In the tracheobronchial tree, the localization of SARs in kittnens and puppies is roughly similar to that of the adult. Briefly, SARs are

210

concentrated in the larger airways and those active at FRC are mainly located in the trachea.

The properties of SARs in the newborn (low discharge frequency, high transpulmonary pressure threshold, few SARs active at FRC) may have several explanations: 1 - immaturity of the receptors andjor fibers (most of the fibers are unmyelinated at birth); 2 -mechanical factor (tension at the site of the receptor and U-shaped cartilage) and finally 3 -transpulmonary pressure at FRC (small in newborn). It appears therefore that in the newborn, at the beginning of inspiration and during expiration the pulmonary feedback is almost negligible and the quality of vagal information during inspiration is less in the newborn than in the adult.

Vagal Reflexes

Vagal information modifies the activities of medullary respiratory neurons and then the breathing pattern. We have to keep in mind that the strentgh of vagal reflexes depends on the sensitivity of medullary respiratory neurons to vagal input and on the vagal information itself. We have previously shown that this last parameter is qualitatively and quantitatively less important in the newborn than in the adult. Unfortunately, very few data are available concerning direct or indirect recordings of medullary respiratory neurons activities in newborn (Bystrzycka et al., 1975; Lucier. et al., 1979; Marlot and Duron, 1981). These results show that medullary inspiratory neurons are inhibited by various peripheral stimulations which produce excitations in adult. For example, in the neonate kitten, only 23'1. rhythmic medullary respiratory neurons receive a short-latency excitatory input from vagal or superior laryngeal nerve stimulation (Lucier et al., 1979). In contrast, in the adult cat, 60'1. of these neurons are excited by similar stimulations (Sessle et al., 1978). Moreover, the newborn kitten has few early inspiratory motoneurons indicating that during the initial portion of inspiration, inspiratory activity is weaker and thus, could be more easily inhibited in the newborn than in the adult (Marlot and Duron, 1981). Recently, Sica et al. (1987) have recorded extracellular activities of dorsal medullary inspiratory neurons in newborn pigs. Their results show that projections from these neurons to phrenic motoneurons may not be as abundant in the piglet as they are in the adult cat and that a large proportion of dorsal medullary inspiratory neurons have an inspiratory decrementing pattern.

The Breuer-Hering inflation reflex due to activation of SARs has been analyzed in newborn using several manoeuvres: increase of tidal volume with C02 inhalation, tracheal occlusion at FRC, lung inflation, resistive and elastic loads, continuous positive pressure or continuous negative pressure applied around the thorax. In spite of discrepancies between results, the Breuer-Hering inflation reflex was shown to be already mature at birth in different species. Is this reflex stronger in the newborn than in the adult? The answer remains unclear. Results have shown that for the same transpulmonary pressure, apnea following inflation is shorter in newborn rats and rabbits and similar in newborn dogs than in adult (Gaultier and Mortola, 1981). In contrast, during C02 breathing, shortening of inspiratory time is greater in young kittens for the same increase of phrenic activity, suggesting that the strength of the reflex decreases during maturation (Trippenbach, 1981). Finally, in babies, a response similar to that observed in adults is described during C02 inhalation (Bodegard, 1975). Recently, Kosch et al. (1986) applied resistive and elastic loads and total airway occlusions to single inspiration in full-term sleeping newborn infants. They showed that the relationship between inspiratory time and inspired volume is curvilinear

211

for all loads when inspiratory time is measured from diaphragm electromyogram. When inspiratory time is measured from the pattern of airflow, the effect of loading results in different inspired volume vs. inspiratory time relationships for elastic and resistive loads. This result indicates that the use of neural timing is a more direct indicator of respiratory control than mechanical timing.

The variations observed in newborn may have several explanations related to the animal species and experimental conditions (anesthetized or unanesthetized preparation, awake or sleeping animals). For example, in newborn rats, the inspiratory changes due to negative pressure applied arow1d the body (increasing FRC) are always more marked after than before anesthesia (Marlot and Mortola, 1984). Indeed, in several occasions, negative pressure leads to apnea in anesthetized animals; this was never observed before anesthesia. The differences observed between the two preparations are not due to different stimulation of pulmonary stretch receptors. The compliance of the whole respiratory system is similar before and after anesthesia. The fact that no difference could be detected despite the well-known decrease of muscle tone with anesthesia indicates that the compliance of the lung in newborn rats is so small compared to the chest-wall compliance that it is effectively the only determinant of the compliance of the whole inspiratory system. This also indicates that the changes in transpulmonary pressure were very close to the measured changes in transrespiratory system pressure, hence similar before and after anesthesia. A different pressure stimulation of the pulmonary stretch receptors before and after anesthesia seems therefore very unlikely. It seems reasonable to conclude that the most likely explanation for the more marked effect of negative pressure breathing after anesthesia is an effect of anesthetic on the central integration of the vagal information. In their recent study on anesthetized, paralyzed and ventilated newborn pigs, Sica et al. (1987) show that when CVN activity is removed by withholding lung inflation, dorsal medullary inspiratory neurons have a response similar to that of inspiratory neurons in adult cat. This result suggests that connections between vagal afferent fibers and medullary inspiratory neurons are fully developed at birth, at least in the pig.

The Breuer-Hering deflation reflex results in an increase of inspiratory frequency due to deflation of the lw1gs. This reflex is most likely mediated by RARs even if deflation reduces also activity of SARs. This reflex has been found to be present in newborn rabbits, kittens and monkeys (Dawes and Matt, 1959; Trippenbach et al., 1979, 1985) or to be absent in newborn kittens and babies (Fleming et al., 1978; Marlot and Duron, 1979a; Marlot and Mortola, 1984). This reflex is very sensitive to anesthesia and like the Breuer-Hering inflation reflex, depends on the integrity of the larYllgeal muscle activity (Johnson, 1979). In contrast to the adult, positive pressure around the body (collapsing pressure) decreasing FRC does not change significantly the breathing pattern in newborn rats (Marlot and Mortola, 1984). Nevertheless, the peak of expiratory flow is delayed and there is a tendency to maintain the lung volume elevated. Both phenomena could be the result of a more marked post-inspiratory muscle activity orjand laryngeal narrowing in expiration.

Data on respiratory reflexes due to activation of pulmonary C-fibers are very scarce in the literature. Stimulation of J-receptors by injection of phenyl diguanide into the right atrium in one to six daysold kittens does not produce any respiratory effect (Kalia, 1976). Respiratory reflexes are fully developed by the lOth day of postnatal life. This result suggests that the role of these receptors is minimal in the early neonatal period.

212

STIMULATION AND SECTION OF THE CERVICAL VAGUS NERVE

In newborn kittens, high frequency (50 to 100 Hz) stimulation of the CVN produces an inhibition of inspiration similar to that observed in adult. Low frequency (15 to 20 Hz) stimulation of the CVN elicits only an inconsistent reinforcement of inspiration without change in respiratory frequency (Marlot and Duron, 1979b). The increase in respiratory frequency obtained in adult is observed after the 3rd week of postnatal life. When applied during expiration, vagal stimulation increases expiratory time according to the stimulus delay (Trippenbach and Kelly, 1985).

After bilateral vagotomy, respiration becomes deeper and slower with a prolongation of both inspiratory and expiratory times. The percent changes in expiratory and inspiratory times are of the same order in anesthetized newborn rabbits (Mortola et al., 1987). In decerebrated newborn kittens, expiratory time increases more than inspiratory time after bilateral vagotomy (Marlot and Duron, 1979b). This discrepancy could be due to different experimental conditions or to a species difference.

Recently, Clement et al. (1986) have shown that vagotomy does not affect respiratory system compliance and resistance in piglets. In contrast, resistance increases in the adult suggesting that efferent vagal control of bronchomotor tone is more pronounced in the adult. It should also be noted that vagotomy does not affect the postnatal development of the lung (Mortola et al., 1987).

SUMMARY

In summary, the cervical vagus nerve of newborn is more or less immature according to its number of myelinated fibers. Moreover, activities from rapidly and slowly adapting mechanoreceptors suggest that vagal information during the respiratory cycle is weaker in the newborn than in the adult. Breuer-Hering inflation and deflation reflex are mature at birth but their relative strength are difficult to estimate because several factors interact in the final respiratory response. It will be useful to have additional information on the connections between vagal afferent fibers and medullary inspiratory neurons to improve our knowledge on the role of the vagus nerve in the control of breathing in newborn mammals.

REFERENCES

Agostoni, E., Chinnock, J.E., Deburgh-Daly, M., 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, ~ Physiol. (London), 135: 182.

Bodegard, G., 1975, Control of respiration in newborn babies, Acta Paediat. Scand., 64: 684.

Bruce, E.N., 1981, Control of breathing in the newborn, Ann. Biomed. ~. 9: 425.

Bystrzycka, E., Nail, B.S., and Purves, M.J., 1975, Central and peripheral neural respiratory activity in the mature sheep foetus and newborn lamb, Respir. Physiol., 25: 199.

Clement, M.G., Mortola, J.P., Albertini, M., and Aguggini, G., 1986, Effects of vagotomy on respiratory mechanics in newborn and adult pigs, ~ ~ Physiol., 60: 1992.

213

Coombs, H.C., and Pike, F.H., 1930, The nervous control of respiration in kittens, Am. ~ Physiol., 95: 681.

Dawes, G.S., and Matt, J.C., 1959, Reflex respiratory activity in the newborn rabbit, ~ Physiol. (London), 152: 271.

De Neef, K.J., Jansen, J.R.C., and Vesprille, A., 1982, Developmental morphometry and physiology of the rabbit vagus nerve, Develop. Brain Res., 4: 265.

Farber, J.P., Fisher, J.T., and Sant'Ambrogio, G., 1984, Airway receptor activity in the developing opossum, Am. ~ Physiol., 246: 756R.

Fisher, J.T., and Mortola, J.P., 1980, Statics of the respiratory system in newborn mammals, Respir. Physiol., 41: 155.

Fisher, J.T., and Sant'Ambrogio, G., 1982, Location and discharge properties of respiratory vagal afferents in the newborn dog, Respir. Physiol., 50: 209.

Fisher, J.T., and Sant'Ambrogio, G., 1985, Airway and lung receptors and their reflex effects in the newborn, Pediatrie Pulmonology, 1: 112.

Fleming, P.J., Bryan, A.C., and Bryan, M.H., 1978, Functional immaturity of pulmonary irritant receptors and apnea in newborn preterm infants, Pediatrics, 61: 515.

Gaultier, C., and Mortola, J.P., 1981, Hering-Breuer inflation reflex in young and adult mammals, Can. ~ Physiol. Pharmacal., 59: 837.

Jammes, J., Fornaris, E., Mei, N., and Barrat, E., 1982, Afferent and efferent components of the bronchial vagal branches in cat, ~ Auton. Nerv. Syst. , 5: 165.

Jansen, A.H., and Chernick, V., 1983, Development of respiratory control, Physiol. Rev., 63: 437.

Johnson, P., 1979, Comparative aspects of control of breathing during development, in: "Central nervous mechanisms in breathing", C. von Euler, and H. Lagercrantz, eds., Pergamon Press, Oxford.

Kalia, M., 1976, Visceral and somatic reflexes produced by J-pulmonary receptors in newborn kittens,~~ Physiol., 41: 1.

Kosch, P.C., Davenport, P.W., Wozniak, J.A., and Stark, A.R., 1986, Reflex control of inspiratory duration in newborn infants, ~ ~ Physiol., 60: 2007.

Kraus, H.F., Jordan, J., Wen, J., and Farber, J.P., 1984, Developmental morphology of the opossum vagus nerve, Fed. Proc., 43: 505.

Lucier, G.E., Storey, A.T., and Sessle, B.J.~9~Effects of upper respiratory tract stimuli on neonatal respiration: reflex and single neurons analyses in the kitten, Biol. Neonate, 35: 82.

Marlat, D., 1982, Recherches sur le contrOle nerveux de la respiration chez le chaton nouveau-ne et au cours de la periode post-natale, These Doctorat es-Sciences, Universite de Picardie, Amiens.

Harlot, D., and Duron, B., 1979a, Postnatal maturation of phrenic, vagus and intercostal nerves in the kitten, Biol. Neonate, 36: 264.

Harlot, D., and Duron, B., 1979b, Postnatal development of vagal control of breathing in the kitten, ~ Physiol. (Paris), 75: 891.

Marlot, D., and Duron, B., 1981, Postnatal development of the discharge pattern of phrenic motor units in the kitten, Respir. Physiol., 46: 125.

Harlot, D., Mortola, J.P., and Duron, B., 1982, Functional localization of pulmonary stretch receptors in the tracheo-bronchial tree of the kitten, Can.~ Physiol. Pharmacol., 60: 1073.

Harlot, D., and Mortola, J.P., 1984, Positive- and negative-pressure breathing in newborn rat before and after anesthesia, ~ ~ Physiol., 57: 1454.

Miller, A.J., and Dunmire, C.R., 1976, Characterization of the postnatal development of superior laryngeal nerve fibers in the postnatal kitten, ~ Neurobiol., 7: 483.

Mortola, J.P., Saetta, M., and Bartlett, Jr. D., 1987, Postnatal development of the lung following denervation, Respir. Physiol., 67: 137.

214

Ponte, J., and Purves, M.J., 1973, Types of afferent nervous activity which may be measured in the vagus nerve of the sheep foetus, ~ Physiol. (London), 229: 51.

Sachis, P.N., Armstrong, D.L., Becker, L.E., and Bryan, A.C., 1982, Myelination of the human vagus nerve from 24 weeks post-conceptional age to adolescence, ~Neuropath. Exp. Neural., 41: 466.

Schwieler, G.H., 1968, Respiratory regulation during postnatal development in cats and rabbits and some of its morphological substrate, Acta Physiol. Scand., suppl. 304.

Sessle, B.J., Greenwood, L.F., Lund, J.P., and Lucier, G.E., 1978, Effect of upper respiratory tract stimuli on respiration and single respiratory neurons in the adult cat, Exp. Neural., 61: 245.

Sica, A.L., Donnelly, D.F., Steele, A.M., and Gandhi, M.R., 1987, Discharge properties of dorsal medullary inspiratory neurons in newborn pigs, Brain Res., 408: 222.

Steele, A.M., 1986, Developmental changes in neural control of respiration. in: "Developmental neurobiology of the autonomic nervous system", P.M. Gootman, ed., Humana Press Inc., Clifton, N.J.

Trippenbach, T., 1981, Laryngeal, vagal, and intercostal reflexes during the early postnatal period, ~Develop. Physiol., 3: 133.

Trippenbach, T., Zinman, R., Mazes, R., and Murphy, L., 1979, Differences and similarities in the control of breathing pattern in the adult and neonate, in: ''Central nervous control mechanisms in breathing", C. von Euler and H. Lagercrantz, eds., Pergamon Press, Oxford.

Trippenbach, T., Kelly, G., and Marlot, D., 1985, Effects of tonic vagal input on breathing pattern in newborn rabbit,~~ Physiol., 59: 223.

Trippenbach, T., and Kelly, G., 1985, Expiratory effects of vagal stimulation in newborn kittens, ~ ~ Physiol., 59: 218.

215

NEURAL MECHANISMS THAT LEAD TO APNEA

A.I. Pack, L.R. Kline, J.C. Hendricks, and M.F. Cola

Cardiovascular-Pulmonary Division, Department of Medicine School of Medicine and School of Veterinary Medicine University of Pennsylvania, Philadelphia, USA

During apneas or hypopneas in sleep there are intermittent reductions in the activity of the upper airway muscles (obstructive apnea) or of the respiratory pump muscles such as the diaphragm (central apnea). Although anatomical abnormalities in the upper airway can predispose subjects to development of apnea, presumably by magnifying the mechanical effects of reduced activity of the upper airway muscles, important to our understanding of the pathogenesis of apnea is determining the mechanisms that produce intermittent inhibition or disfacilitation of the respiratory muscles during sleep. In this chapter we discuss our current understanding of the mechanisms that produce this reduction in neural activity in both non-rapid eye movement sleep (nonREM) and rapid eye movement sleep (REM). It is our thesis that the neural mechanisms that lead to apnea are different in these states.

NEURAL MECHANISMS PRODUCING APNEA IN NON-REM SLEEP

In Stage 1-2 sleep regular periodicities of ventilation occur even in subjects who do not exhibit frank apneas (Bulow, 1963; Webb, 1974; Shore et al., 1985). Periodicities are observed in the neural activity of the upper airway muscles and of the diaphragm (Onal and Lopata, 1982). Although these oscillations in neural activity are now well characterized the mechanisms producing them are less certain. One theory proposes that they result from unstable operation of the chemical feedback system that controls ventilation (Cherniack, 1981; Longobardo et al., 1982; Cherniack, 1984; Khoo et al., 1982). Since this is a closed-loop negative feedback system, control systems theory provides a theoretical basis for predicting when such instability will occur resulting in selfsustaining oscillations. Indeed this system has been analyzed theoretically and instability criteria have been proposed (Mackey and Glass, 1977; Elhefnawy et al., 1987). In general, such systems will exhibit self-sustaining oscillations when the overall loop gain is sufficiently high and there is a sufficient phase-lag between the output of the controlled plant and of the controller (for more extensive discussion, see Khoo et al., 1982). The frequency of oscillation of the system will be determined by the parameters of the system and will be such that there is 180° phase-lag between the output of the controlled plant and of the controller (Khoo et al., 1982).

217

Instability of the chemical feedback system controlling ventilation seems to explain the periodic breathing that occurs during non-REM sleep at high altitude (Brusil et al., 1980; Waggener et al., 1984; Lahiri et al., 1983; White et al., 1987) or during induced hypoxia (Berssenbrugge et al., 1983). In this case hypoxia increases the gain of the peripheral chemoreflex thereby destabilizing ventilatory control. The oscillations that occur have cycle-times of the order of 20 seconds, i.e. intervals between successive zeniths or nadirs in ventilation (Brusil et al., 1980; Waggener et al., 1984). The cycle-time is a function of the degree of hypoxia being shorter during more intense hypoxic stimulation (Waggener et al., 1984). Not unexpectedly the degree of periodic breathing in individual subjects is related to the magnitude of their peripheral chemoreflex (Lahiri et al.,1983; White et al., 1987). It is absent in the subjects, such as high altitude dwellers, who have a blunted ventilatory response to hypoxia (Lahiri et al., 1983). That this type of periodic ventilation is produced by unstable operation of the ventilatory control system is supported by theoretical analyses (Khoo et al., 1982). Mathematical models of this feedback system (Khoo et al., 1982) correctly predict that oscillations produced by increasing the gain of the peripheral chemoreceptor will have cycle times of the order of 20 s and the cycle time will decrease with larger increases in gain (more intense hypoxia). As discussed above this is that is found experimentally.

These data supporting the concept that periodic breathing produced by hypoxia is related to unstable operation of this control system cannot, however, be extrapolated to other types of periodic breathing. Indeed there is some evidence that instability to feedback control is not the explanation for oscillations in neural output to the respiratory muscles in subjects with sleep apnea syndrome. Both in obese middle-aged subjects with obstructive sleep apnea (Garay et al., 1981) and elderly subjects with sleep-disorders breathing (Littner et al., 1984) reduced ventilatory responses to hypoxia and hypercapnia are found. Such reductions in response whould tend to stabilize the ventilatory control system. Moreover, apneas can occur even in the virtual absence of any ventilatory response to chemical stimuli such as in subjects with hypothyroidism (Millman, et al., 1983). While these data cast doubt on the instability hypothesis they do not by themselves refute it. Instability of operation is not determined by the value of one single parameter

218

STATEDEPENDENT

INPUT

Fig. 1. Block diagram for the system controlling ventilation. It is envisaged that the input to the central respiratory pattern generator is an integration of inputs from the peripheral and central chemoreceptors as well as an input that is dependent on state. This latter input has been called the "wakefulness stimulus" (Pack et al., 1988 in press, with permission).

(ventilatory response) but rather by the values of a number of parameters. Among the other parameters particularly important is the delay between the lung and central chemoreceptor in the brainstem (Mackey and Glass, 1977; Elhefnawy et al., 1987) but there is, to date, no method to measure this.

Nevertheless, the arguments just presented do lead to consideration of whether there are other explanations for the oscillations that occur in neural output to the respiratory muscles in non-REM sleep. The system controlling ventilation is not completely a closed-loop one. There are other inputs to this system that are not related to inputs from the peripheral and central chemoreceptors (see Fig. 1)

One particularly important input is that which is dependent on sleep state. This has been termed the "wakefulness stimulus" (Fink, 1961; Asmussen, 1977) although the neural origin of this is unknown. Perhaps the strongest evidence for the existence of this input is the observation that there is, with sleep, a very large increase in the apnea threshold, i.e. PC02 at which rhythmic ventilation ceases (Skatrud and Dempsey, 1981). Neurophysiological studies of the firing of respiratory-related neurons in the brainstem also reveal decrements in firing with sleep (Orem et al., 1985). Thus it is conceivable that oscillations in ventilation are secondary to oscillations in state that result in oscillations in the state-dependent input to the ventilatory control system. Certainly in the elderly, in whom sleep-disordered breathing is extremely common (Carskadon and Dement, 1981; Ancoli-Israel et al., 1985), there is evidence of instability in state control and sleep state is said to be poorly consolidated (Carskadon et al., 1982). The concept that oscillations in state can lead to oscillations in ventilation is not new since it was originally discussed by Bulow (1963) and more recently by Phillipson (1978). There is, however, at present no evidence to support this postulate.

If the ventilatory oscillations are secondary to oscillations in state one would predict that synchronous oscillations should occur in the frequency content of the electroencephalogram which exhibits characteristic changes with alterations in sleep state (for extensive discussion, see Hasan, 1983). To investigate this prediction, we have recorded simultaneous measurements of ventilation, using a respiratory inductive plethysmograph (Respitrace), and the electroencephalogram in five elderly subjects in Stage 1-2 sleep. All subjects exhibited periodic breathing and in two subjects more than 5 apneasjhour of sleep were observed. Measurements of minute ventilation were obtained for each breath. These data were then resampled using the technique described by Waggener et al. (1984) to obtain measurements equally spaced in time, in this case at 5.12 second intervals (5.12 seconds was chosen since this is close to the breathing frequency and it provides an appropriate number of points for a Fast Fourier Transform). In brief, this technique involves making the assumption that ventilation is constant over the duration of a breath, resampling the measured breath-by-breath data at a sampling frequency of 10 samples every 5.12 seconds, and then obtaining the average of each 10 sample points. This results in a time series with equally spaced samples at approximately the breathing frequency (one sample every 5.12 seconds). (The resampling at a higher frequency with subsequent averaging results in all breaths, no matter how brief, being represented in the final data string.)

The electroencephalographic data were also divided into 5.12 second intervals, with the begining and end of each interval corresponding to the times of the ventilatory measurements. For each such interval we computed a Fast Fourier Transform of the data. Each such frequency

219

STAGE 1-2 SLEEP Minute Venti lation

Me on Frequency (EEG l 12.0

t2.0 10.0 c:

E N :r

...... 8.0 ...1 >- 10.0 (.)

z L&J z

6.0 Q 1-ct ...1

4 .0 i= z L&J >

2.0

::::> 0 L&J 8.0 It: .... z < 6.0 L&J ~

0 2 4 6 8

TIME {Minutes)

Fig. 2. Time course of minute ventilation (- - -) and mean frequency (----) of the electroencephalogram in a single elderly subject in Stage 1-2 sleep. Relative ly synchronous oscillations in the two variables is evident. Similar results were obtained when the EEG parameter was alpha power .

spectrum was then reduced to two parameters - mean frequency and a lpha power. The latter was taken as the power in the frequency band (8-12 Hz) while the former was computed using t he formul a :

120 L P(fi)fi i=4

MF = [1] 120 L P(fi) i=4

where fi is the frequency in 0.25 Hz increments (the spectral resolution) and P(fi) is the spectral density .

These computational procedures resulted in there being three time series with samples at identical time points - minute ventilation, mean frequency of the EEG and alpha power. Examination of the data revealed t hat there were apparently relatively synchronous osci llations in the frequency content of t he EEG and in ventilation (s ee Fig . 2).

To analyze these data more definitive ly we cal culated the coherence between ventilation and mean frequency of the EEG as well as between ventilation and alpha power (see Table 1). High coherence values were obtained between the oscillations in ventilation and the parameters derived from the frequency spectra of the EEG (coherence is analogous to a corPelation coefficient). While these data indicate that there are oscillations in ventilation and the frequency content of the EEG at the same frequency, critical to investigat i on of t he postul ate i s determination of the time-lag between the two types of oscil lation . If oscillations in ventilation are the primary process then these should lead oscillations in the frequency content of the EEG. If, however, the primary proces is oscillatory changes in state , then the osci llations

220

Table 1

Subject No. Cycle Time (s)* Coherence Time-Lag (s)

1 - Ventilation/Mean Frequency 50 s 0.88 +5.12 s Ventilation/Alpha Power 71s 0.91 -15.36 s

2 - Ventilation/Mean Frequency 68 s 0.95 0.0 s Ventilation/Alpha Power 73 s 0.96 0.0 s


4 - Ventilation/Mean Frequency 81 s 0.93 -5.12 s Ventilation/Alpha Power 81 s 0.86 o.o s


Mean + SD

Ventilation/Mean Frequency 58.4 ± 13.8 0.89 ± 0.05 o.o

Ventilation/Alpha Power 63.6 ± 14.4 0.88 ± 0.05 -3.07

*Cycle Time is that which gave the maximum coherence between the two time series. Positive values for time-lag indicate that ventilation leads the EEG time series while negative values indicate the converse.

should be relatively synchronous. To determine this time-lag we calculated correlation coefficients between the different time series using a number of different values for lag (or lead). The value of timelag that produced the maximum correlation coefficient was taken as the appropriate estimate of this variable. This analysis revealed that the two oscillatory processes (ventilation and frequency content of the EEG) were virtually synchronous (see Table 1).

These data are compatible with the hypothesis that, at least in these elderly subjects, the primary abnormality is an oscillation in state. They cannot, however, prove this since measurement of the frequency content of the electroencephalogram is a relatively crude tool. It is difficult, however, at present, to investigate in a definitive way mechanisms producing periodicities in ventilation in non-REM sleep since no animal model of the process has been described.

NEURAL MECHANISMS PRODUCING APNEA IN REM SLEEP

This lack of a suitable animal model does not apply, however, to apneas in REM sleep. Recently our laboratory has described that the English Bulldog exhibits apneas in REM sleep (Hendricks et al., 1987a in press). In this animal the apneas are commonly of the mixed type and they lead to marked oxygen desaturation. The animals are, moreover, hypersomnolent during the day (Hendricks et al., 1987a in press). In

221

this animal we have observed not only upper airway obstruction, but also alterations in the activity of the diaphragm in association with the phasic events of REM sleep (Kovalski et al., 1987). Currently, there is evidence for a number of different types of inhibition of the diaphragm's activity in REM sleep, although it is evident that the respiratory muscles must be relatively spared from the profound atonia that affects other skeletal muscles. The first type of inhibition is a tonic alteration that persists throughout REM sleep. Sieck et al. (1984) have described a drop out of activity of individual motor units in the diaphragm that persists throughout REM. There is, moreover, a brief intense inhibition of the diaphragm's activity that occurs in association with phasic events in REM. These inhibitions, termed fractionations by Orem (1980), last for 40-100 ms and their occurrence is cross-correlated with ponto-geniculo-occipital (PGO) waves, a marker of phasic events in REM sleep. Since phasic events tend to occur in clusters, fractionations of the diaphragm's activity also tend to occur in clusters. The third type of inhibition (or disfacilitation) that has been identified is what has been termed intermittent inhibition (Kline et al., 1986). Recognition of this was based on our observation that there was marked similarity in the time course of the diaphragmatic EMG during breaths in tonic REM, i.e. without phasic eye movements, and that the EMG in the majority of breaths in phasic REM had the same type of time profile. Intermittently, however, the rate of rise of the diaphragmatic EMG was reduced below the range found in tonic REM and intermittently it was increased. We termed this "intermittent inhibition" and "intermittent excitation", respectively (Kline et al., 1986). The frequency of these alterations in the rate of rise of the diaphragmatic EMG increases with phasic events. In cats, intermittent iru1ibition tended to occur for single breaths whereas in preliminary studies in humans we have observed that intermittent inhibition is found for six to eight consecutive breaths (Mahar et al., 1987). Alterations in neural activity associated with phasic events does not affect in a homogeneous way the respiratory muscles. Indeed, we have observed marked dissimilarity of firing patterns even within one muscle - the right and left costal diaphragms -such that one hemi-diaphragm was active while the other was silent (Hendricks, et al., 1987b). Such inhomogeneity in activation of the respiratory muscles must contribute to the periodic reductions in ventilation that are characteristic of this state.

Although these phenomena have been described, the neural mechanisms producing them are only beginning to be elucidated. One obvious neural structure that might be involved is the dorsal pontine tegmentum, lesions of which abolish the atonia of skeletal muscles in REM sleep, thereby creating an animal that exhibits behavior in this sleep state (see for example, Hendricks et al., 1982). To address the role of this structure in modifying the activity of the respiratory muscles in REM we have performed lesions of this area and have studied the resultant changes in neural activity. Following lesioning there is an increase in the rate of rise of the moving average of the diaphragm EMG in REM sleep. We cannot distinguish, however, whether this is due to loss of a tonic inhibition throughout REM or to loss of what we have termed intermittent inhibition. Following lesioning there were no long periods of tonic REM, i.e. without phasic eye movements; presence of this is essential for the analysis we employ to detect intermittent inhibition. We did note, however, that lesions of the dorsal pontine tegmentum abolished the occurrence of fractionations of the diaphragm's activity. Thus this area is involved in generation of tonic inhibition throughout REM and in brief intense phasic inhibition. Organization of the structure must be more complex than previously realized since skeletal muscle receives a profound tonic inhibition in REM but little phasic inhibition whereas the converse is the case for the respiratory muscles.

222

DIA EMG

SHAM STIMUlUS

DIA [ EMG

AUDITORY STIMUlUS

50 ms

Fig. 3. Averaged responses in REM of diaphragmatic activity to sham tests (electronic pulses but no auditory tones) in upper panel and to auditory tones (lower panel). Following the auditory tones there is a marked reduction in activity.

It seems unlikely, however, that phasic inhibition is primarily generated in the area of the brainstem. One possiblity is that such phasic inhibitions are manifestations of a startle response. This concept is based on several pieces of evidence. First, Bowker and Morrison (1976) have shown that startling stimuli, such as auditory tones, can induce PGO waves , a marker of phasic events in REM, even in non-REM sleep. Thus PGO waves in REM may be manifestations of what one might term "endogenous startles" in the highly activated brain of REM (for full discussion of this see Morrison and Reiner, 1985). Second, startling stimuli produce brief pauses in EMG activity in extensor muscles in non-REM sleep (flexor muscles show excitation). This is reviewed in Glenn (1985).

These arguments raise the question as t o whether the inhibitions of the diaphragm, which are coupled to phasic events in REM, are manifestations of a startle. To address this hypothesis we have induced startles in REM s leep using auditory tones (100 dB, 60 ms duration). The tones were delivered in the middle of the diaphragmatic burst, at a fixed magnitude of the moving average of EMG activity. A large number of trials were done (60 to 150) as were "sham" studies, i.e. an electronic pulse was generated but no tone ensued. The latter acted as a control. Signal averaging was employed to measure the responses. We have observed that these auditory tones induce a brief intense reduction in diaphragmatic activity, the duration of which is similar to that described for fractionations of diaphragmatic activity in REM sleep, i.e. 40-100 ms (Orem, 1980) (see Fig. 3). These results add weight to the concept that the brief but large r eductions in diaphragmatic activity that are found in REM are related to activation of the startle system.

CONCLUSION

An important component in our understanding of the mechanisms that produce apnea during sleep is elucidating the neural mechanisms that produce inhibition or disfacilitation of the activity of the upper airway and respiratory pump muscles . At the current time it seems most likely that this knowledge can be more easily obtained for mechanisms in REM

223

sleep as compared to non-REM. There is already considerable knowledge on the neural mechanisms that produce alteration in skeletal muscle behavior in this state. Certain of the concepts derived from such studies can be directly applied to studies of alterations in respiratory muscle performance in REM sleep. Moreover, the recent identification of an animal model that has sleep disordered breathing in REM sleep should facilitate study of the role of different neural mechanisms in the pathophysiology of apnea.

ACKNOWLEDGMENTS

The original studies discussed were supported by the following grants: AG-03934, HL-08805, HL-07163, HL-29596.

REFERENCES

Ancoli-Israel, S., Kripke, D.F., Mason, W., and Kaplan, O.J., 1985, Sleep apnea and periodic movements in an aging sample, J. Gerontology, 4: 419.

Asmussen, E., 1977, Regulation of respiration: "the black box", Acta Physiol. Scand., 99: 85.

Berssenbrugge, A., Dempsey, J., Iber, C., Skatrud, J., and Wilson, P., 1983, Mechanisms of hypoxia-induced periodic breathing during sleep in humans, ~ Physiol. (Land.), 343: 507.

Bowker, R.M., and Morrison, A.R., 1976, The startle reflex an PGO spikes, Brain Res., 102: 185.

Brusil, P.J., Waggener, T.B., Kronauer, R.E., and Gulesian, P., 1980, Methods for identyfying respiratory oscillations disclose altitude effects,~~ Physiol., 48: 545.

Bulow, C., 1963, Respiration and wakefulness in man, Acta Physiol. Scand., Suplement 209.

Carskadon, M.A., Brown, E.D., and Dement, W.C., 1982, Sleep fragmentation in the elderly: relationship to daytime sleep tendency, Neurobiol. Aging, 3: 321.

Carskadon, M.A., and Dement, W.C., 1981, Respiration during sleep in the aged human, ~Ger., 36: 420.

Cherniack, N.S., 1981, Respiratory dysrhythmias during sleep, ~Engl. ~ Med., 305: 325.

Cherniack, N.S., 1984, Sleep apnea and its causes,~ Clin. Invest., 73: 1501.

Elhefnawy, A., Saidel, G., Bruce, E., and Cherniack, N., 1987, Stability index for chemical control of breathing, Fed. Proc., 46: 654 (abstract)

Fink, B.R., 1961, Influence of cerebral activity in wakefulness on regulation of breathing,~~ Physiol., 16: 15.

Garay, S.M., Rapoport, D., Sorkin, B., Epstein, H., Feinberg, I., and Goldring, R., 1981, Regulation of ventilation in the obstructive sleep apnea syndrome, Am. Rev. Respir. Dis., 124: 451.

Glenn, L.L., 1985, Brainstem and spinal control of lower limb motoneurons with special reference to phasic events and startle reflexes, in: "Brain Mechanisms of Sleep", D.J. McGinty, R. DruckerColin, A. Morrison and P.L. Parmeggiani, eds., Raven Press, New York,

Hasan, J., 1983, Differentiation of normal and disturbed sleep by automatic analysis, Acta Physiol. Scand., Supplement 526.

Hendricks, J.C., Kline, L.R., Kovalski, R.J., O'Brien, J.A., Morrison, A.R., and Pack, A.I., 1987a, The English Bulldog: A natural model of sleep disordered breathing, ~ ~ Physiol. (in press).

Hendricks, J.C., Kline. L.R., Tuttle, S.V., Davies, R.O., and Pack, A.I.,

224

1987b, Heterogenous activation of the costal diaphragm during rapid eye movement sleep (REMS), Fed. Proc., 46: 653 (abstract).

Hendricks, J.C., Morrison, A.R., and Mann, G.L., 1982, Different behaviors during paradoxical sleep without atonia depend on positive lesion site, Brain Res., 239: 81.

Khoo, M.C.K., Kronauer, R.E., Strohl, K.P., and Slutsky, A.S., 1982, Factors inducing periodic breathing in humans: a general model, J. ~ Physiol., 53: 644.

Kline, L.R., Hendricks, J.C., Davies, R.O., and Pack, A.I., 1986, Control of the activity of the diaphragm in rapid eye-movement sleep, ~ ~ Physiol., 61: 1293.

Kovalski, R.J., Hendricks, J.C., Kline, L.R., and Pack, A.I., 1987, Activity of the respiratory muscles during rapid eye movement sleep in the English Bulldog, Am. Rev. Respir. Dis., 135: A47 (abstract).

Lahiri, S., Maret, K., and Sherpa, M.G., 1983, Evidence of high altitude sleep apnea on ventilatory sensitivity to hypoxia, Respir. Physiol., 52: 281.

Littner, M., Young, E., McGinty, D., Beahm, E., Riege, W., and Sowers, J., 1984, Awake abnormalities of control of breathing and of the upper airway, Chest, 86: 573.

Longobardo, G.S., Gothe, B., Goldman, M.D., and Cherniack, N.S., 1982, Sleep apnea considered as a control system instability, Respir. Physiol., 50: 311.

Mackey, M.C., and Glass, L., 1977, Oscillations and chaos in physiological control systems, Science, 197: 287.

Mahar, P.J., Kline, L.R., Grippi, M.A., and Pack, A.I., 1987, Changes in respiratory muscle activity during REM sleep in man, Am. Rev. Respir. Dis., 135: A47 (abstract).

Millman, R.P., Bevilacqua, J., Peterson, D.D., and Pack, A.I., 1983, Central sleep apnea in hypothyroidism, Am. Rev. Respir. Dis., 127: 504.

Morrison, A.R., and Reiner, P.B., 1985, A dissection of paradoxical sleep, in: "Brain Mechanisms of Sleep", D.J. McGinty, R. DruckerColin, A. Morrison and P.L. Parmeggiani, eds., Raven Press, New York,

Onal, E., and Lopata, M., 1982, Periodic breathing and the pathogenesis of occlusive sleep apneas, Am. Rev. Respir. Dis., 126: 676.

Orem, J., 1980, Medullary respiratory neuron activity: relationship to tonic and phasic REM sleep,~~ Physiol., 48: 54.

Orem, J., Osorio, I., Brooks, E., and Dick, T., 1985, Activity of respiratory neurons during NREM sleep,~ Neurophysiol., 54: 1144.

Pack, A.I., Silage, D.A., Millman, R.P., Knight, H., Shore, E.T., and Chung, D.C.C., 1988, Spectral analysis of ventilation in elderly subjects awake and asleep, ~ ~ Physiol., in press.

Phillipson, E.A., 1978, Control of breathing during sleep, Am. Rev. Respir. Dis., 119: 909.

Sieck, G.C., Trelose, R.B., and Harper, R.M., 1984, Sleep influences on diaphragmatic motor unit discharge, Exp. Neural., 85: 316.

Shore, E.T., Millman, R.P., Silage, D.A., Chung, D.C., and Pack, A.I., 1985, Ventilatory and arousal patterns during sleep in normal young and elderly subjects,~~ Physiol., 59: 1607.

Skatrud, J.B., and Dempsey, J.A., 1981, Interaction of sleep state and chemical stimuli in sustaining ventilation,~~ Physiol., 55: 813.

Waggener, T.B., Brusil, P.J., Kronauer, R.E,. Gabel, R.A., and Inbar, G.F., 1984, Strength and cycle time of high-altitude ventilatory patterns in unacclimatized humans,~~ Physiol., 56: 576.

Waggener, T.B., Stark, A.R., Cohlan, B.A., and Frantz, I.D., 1984, Apnea duration is related to ventilatory oscillation characteristics in newborn infants.~~ Physiol., 57: 536.

Webb, P., 1974, Periodic breathing during sleep, ~ ~ Physiol., 37: 899.

225

White, D.P., Gleeson, K., Pickett, C.K., Rannels, A.M., Cymerman, A., and Weil, J.V., 1987, Altitude acclimatization: influence on periodic breathing and chemoresponsiveness during sleep, ~ ~ Physiol., 63: 401.

226

COMPARISON OF THE IMPORTANCE OF MECHANISMS TAKING PART IN THE

SELF-REGULATION OF BREATHING

V.D. Glebovski

Pediatric Medical Institute, Leningrad, USSR

INTRODUCTION

The central breathing pattern generator (CPG) activity is controlled by several self-regulation mechanisms. Among those are feedbacks that include medullary and arterial chemoreceptors, pneumotaxic mechanism, stretch receptors of the respiratory muscles, "flow" receptors of the upper airways. The specificity and relative importance of separate selfregulation mechanisms of breathing have not yet been fully determined.

We have investigated the self-regulation mechanisms of automatic breathing. We did not take into consideration the behavioral changes of breathing and the self-regulation mechanisms involving the suprapontine structures. We tried to determine the importance of the mechanisms mentioned above in the self-regulation of quiet breathing and hypercapnic hyperpnoea.

METHODS

Studies in our laboratory were carried out mainly with spontaneously breathing, tracheostomized, midcollicularly decerebrated cats without anesthesia. The "integrated" (time constant 50 ms) phrenic nerve activity (Cs root), intercostal and abdominal muscles electromyograms were recorded. Tidal volume (VT) was obtained by spirograph of Krogh type (volume 500 - 700 ml) filled with 02. For determination of VT vs. TI relationship rebreathing from the spirograph was performed. Monitoring began after the cessation of hyperventilation apnoea and ended when maximal hyperpnoea had developed (FC02 in spiro~raph about 101). On the basis of obtained records the lung ventilation (V), central inspiratory excitation (CIE) intensity, inspiratory off-switch threshold (10-ST), inspiratory (TI) and expiratory (TE) durations and for respiratory muscles - myotatic reflexes intensities were determined. The details of the methods are described in the original papers from our laboratory (see references).

227

150

50

0 2 T

Fig. 1. Vr vs. Tr relationships in decerebrated cats (averaged data). Series with pronounced Vr to Tr relationships. Ordinate- Vr, ml, abscissatime, s. 1, 3 - Tr, 2, 4 - TE, 1, 2 - vagi nerves intact, 3, 4 - after vagotomy. Hori zontal bars -±SE. Arrows - appearance of abdominal oblique muscle activity. (Glebovski and Gizatullina, 1977a).

RESULTS AND DISCUSSION

The Chemoreceptors

They are the basis of the most potent self-regulation breathing mechanism. Since CPG cannot act without chemoreceptor s ignals, they can be regarded as a part of the CPG .

A good illustration of the chemoreceptor s igni f i cance for the function of CPG is given by the examination of Vr vs . Tr curves obtained in vagotomized cats (Fig. 1., Glebovski and Gizatullina, 1976, 1977a ,b). In most cats (in 8 out of 11) the Vr vs. TI reversed relationship was retained in agreement with the results of Widdicombe and Winning (1974) and Romaniuk et al. (1976). In these cases the first breath after the apnoea had a relatively long duration (3 - 23 s). We believe that t he lack of such relationship in the work of Cl ark and Euler (1972) can be explained by the fact that these authors recorded TI only for Vr which surpassed the level of quiet breathing. In fact in this diapason of Vr changes (70 ml and more) Vr vs.Tr relationship is weak or absent. More seldom (in 3 cats out of 11) the first inspiration after cessation of apnoea had a comparatively short duration (0.68- 2.1 s). With accumulation of C02 in the spirograph TI did not change much.

As hyperpnoea progressed, TE always increased considerably. This is in variance with the data of Widdicombe and Winning (1974) and Miserocchi

228

50

0 2 T J

Fig. 2. Vr vs. Tr relationship in decerebrated cats (averaged data). Series with weak Vr to Tr relationship. (Glebovski and Gizatullina, 1977a). Marking as in Fig. 1.

and Milic-Emili (1975) obtained on anaesthetized cats but in agreement with the results of Romaniuk et al. (1976) in anaesthetized rabbits.

Examination of the results showed that chemoreceptor si~1als are capable of increasing several times the V, the CIE intensity, the inspiratory off-switch thresho ld, the TE, and of reducing Tr to a certain limit. Hypercapnia in such conditions usually brings forth opposite changes of Tr and TE. TE prolongation is probably due to a facilitation of the inspiration-inhibiting mechanism (Euler, 1977,1986). Chemoreceptors regulate the activity of all the main blocks of CPG.

The ~ungs Mechanoreceptors

The pulmonary mechanoreceptors (or those of lower airways more precisely) s lowl y and rapidly adapting regarded in complex , do not influence V considerably. After vagotomy changes of lung ventilation are weak and inconsistent (Lim et al., 1958; Widdicombe, 1964; Glebovski and Zhdanov , 1969; Glebovski and Sukhova, 1983). Both, inhibitory (Bradl ey et al., 1975; Euler et al ., 1976) and facilitatory (DiMarco et al., 1981) influences of lung mechanoreceptors on t he CIE generator are absent after vagotomy.

The curves of VT vs. Tr in cats with pronounced re l ationship got before and after vagotomy, as compared in Fig. 1. show that the lung mechanoreceptor impulses reduce Tr considerably , moving the curves to t he left and down. Secondly, they strongly reduce TE and in consequence both, Vr vs. Tr and Vr vs. TE curves become similar (Glebovski and Gizatullina, 1977). As after vagotomy, the Vr vs . Tr relationship in some cats (5 out of 13) was weak (Fig. 2). But also in these experiments changes of Tr and TEare parallel only when vagal input was intact (Glebovski and Gizatullina, 1977a). Similar changes of Tr and TE with development of hypercapnia were described in awake cats (Gautier , 1976). Variances in the expression of Vr to Tr relationships were pointed out a l so by St.John (1979).

229

So the lung mechanoreceptor activities reduce considerably the inspiratory off-switch threshold, Tr, and especially TE·

The Pneumotaxic Mechanism

The breathing changes after local lesions of NPBM-KF complex depend on the lesion size. They can be relatively small (Euler et al., 1976). But with vast lesions breathing can be disturbed more than after vagotomy (Glebovski and Obukhova, 1978). The extensive lesions of NPBM-KF

Fig. 3. Levels of pontine transections by caudal border of NPBM. Dotted line - rhomboidal fossae plan. BC - brachium conjunctivwn, CP - posterior colliculus, LVN - lateral vestibular nucleus, NPBM- medial parabrachial nucleus, PG - pontine grey, ST - solitary tract, V - motor trigeminal nucleus VII - facial nerve nucleus. (Glebovski, 1983).

complex are accompanied by V decrease (eg. St.John, 1972; Glebovski and Obukhova, 1978; Webber and Peiss, 1979). A complete electrolytic NPBM-KF complex destruction is difficult because of its large dimension (Caille et al., 1981).

Therefore, in order to attain complete abolition of pneumotaxic mechanism we made a transverse brain stem sections in the region of NPBM leading to strong disturbances of breathing that are similar to those evoked by large NPBM-KF complex lesions accompanied by V decrease. But transections in the region of caudal NPBM border (Fi~. 3) lead to less pronounced changes of breathing without significant V changes. These results suggest that extensive injuries of NPBM-KF complex can damage CPG function more strongly than complete pneumotaxic mechanism elimination (Glebovski, 1983). So we evaluated the pneumotaxic mechanism importance by changes in breathing after the brain stem transection near the caudal border of NPBM.

230

roo

0

%

Fig. 4. Comparison of the breathing parameters changes (in percent of quiet breathing values) after vagotomy (white, n = 10), and pneumotaxic mechanism elimination (hatched, n = 8). (Glebovski, 1983). x- significant changes (p < 0.05). All differences between vagotomy and pneumotaxic mechanism elimination are not significant (p > 0.1).

The abolition of the pneumotaxic mechanism brings forth the breathing changes close in their direction and magnitude to those after vagotomy: the increase of VT, Tr and TE as well as of the inspiratory off-switch threshold (Fig. 4) with small changes of V.

Although the di f ferences between the effects of vagotomy and pneumotaxic mechanism elimination were insignificant, the pneumotaxic mechanism abolition was followed by the diminished CIE slope (Breslaw and Glebovski, 1981; Cohen et al., 1986). This was not seen after vagotomy.

Pneumotaxic mechanism seems to control both the inspirat ory offswitch and inspir at ion-inhibiting mechanisms, as well as the CIE slope.

The Nose "Flow" Receptors

The on-switch of the nasal breathing with the background of breathing t hrough the tracheostomy reduces V by about 10~ mainly due t o reduction of VT (Fig. 5). Changes of TJ and TE in vagotomized cats wer e not signif i cant . The responses were abolished by trigeminal nerve blockade (Glebovski, 1981; Glebovski and Sukhova, 1983). The inhibitory inf luences of t he nasal breathing weakened with time .

The nose "flow" receptors (at least those wit h afferent f ibres i n ethmoidal nerves) are cold receptors (Glebovski and Bajev, 1984, 1986). Their i mpulses reduce the sensitivity of CPG to the chemoreceptor signals. The i ntensity of this effect is relatively small.

231

8

Fig. 5. Breathing changes after transition from tracheostomic (before mark) to nasal breathing. From top to bottom: tracheal pressure (em H20), abdominal oblique muscles EMG,"integrated" phrenic activity. Time, s. Before (A) and after (B) trigeminal nerves blockade. (Glebovski and Sukhova, 1983).

The Diaphragm Stretch Receptors

There are very few stretch receptors in the diaphragm (Glebovski, 1962a,b; Ya~argil, 1962; Corda et al., 1965). The attempts to discover reflex responses to the stimulation of the diaphragm stretch receptors (Glebovski and Pavlova, 1962; Sant'Ambrogio et al., 1962; Sant'Ambrogio and Widdicombe, 1965) or of the corresponding afferent fibres (Hoffman and Keller, 1934; Gill and Kuno, 1963; Glebovski, 1970) were negative. Only very weak and inconsistent inhibitory influences were sometimes noticed. We conclude that the function of the diaphgram proprioceptors is performed mostly by the lung mechanoreceptors (Gl ebovski, 1970).

The Intercostal Stretch Receptors

An excitation of the intercostal muscle spindles results in myotatic reflexes at segmental level (Ramos and Mendoza, 1959; Glebovski, 1963, 1965; Sears, 1964; Eklund et al., 1964; Budzinska and Romaniuk, 1986). These reflexes increase the force of muscles contractions with the increase of the load on them and through the influence on the neighbouring segmental motoneurones (Sears, 1964; Glebovski, 1965) assist to interrelation of the neighbour intercostal interspace muscles activity.

The afferent discharges evoked by twitches of the diaphragm (Decima and Euler, 1969; Euler, 1973), mechanical stimulation of the rib cage (Decima et al., 1969; Remmers, 1970, Shimaraeva and Glebovski, 1975), synchroneous electrical stimulation of the intercostal afferent fibres (Remmers, 1973; Shannon, 1980) can evoke intercostal-to-phrenic reflexes, activate the inspiratory off-switch mechanism and change the activity of the brain stem respiratory neurones. Yet the attempts to discover changes of the phrenic motoneurones activity with reduction of breathing movements and changes of the thorax volume with the speeds close to those

232

seen in quiet breathing and hyperpnoea in vagotomized animals, showed that such changes are either absent or very weak (Glebovski and Pavlova, 1962; Glebovski, 1963).

Fig. 6. Changes of intercostal and phrenic activity under tracheal occlusion and lungs inflat ions and deflations in vagotomized decerebrated cats. From top to bottom: External intercostal muscles (3rd interspace) and phrenic (C5) "integrated" activities. Short mark - tracheostomy tube occlusion, prolonged mark - lung inflation (+) or deflat i on (-) i n ml relative t o FRC. b, c, d - lung volume changes at the end of expiration, and e, f - during inspirat ion. Time in s. Note very weak phrenic and pronounced intercostal activity changes. (Shimaraeva, 1978).

We applied occlusion of the airways, lung inflations and deflations (Shimaraeva, 1976, 1978), pressing the abdomen with a pneumatic cuff for the abdomina l wall compliance limitation (Glebovski and Pashkevitch, 1986). All phrenic activity changes were weak or absent, but changes of intercostal musc l es activity were well pronounced at the expense of their myotatic r efl exes (Fig. 6). The most marked changes were evoked by airway occlusion at the end of expiration (on average the Tr decrease by 2~, P < 0.01, the phrenicogram- by 5~, P < 0.01). Before vagotomy such i nfluences produced a 62~ Tr and 126~ ampl i tude increase . The lung inflation on 40 ml after vagotomy did not evoke any significant changes except those of the inter costal muscles excitation intensity. Our data do not support the hypothesis of great import ance of the intercostal muscles receptors in CPG and phrenic motoneurones activity regulation in

233

Table 1. Changes of Breathing Parameters in Responses to Stimulation or Elimination of Separate Feedbacks in Decerebrate Cats (Percents of Values in Quiet Breathing).

The feedback sources

Chemoreceptors (effects of stimulation)

Lung mechanoreceptors (changes after elimination)

Pneumotaxic mechanism (changes after elimination)

Nose "flow" receptors (effects of stimulation)

Diaphragm stretch receptors (effects of stimulation)

Intercostal stretch receptors (effects of stimulation)

The controlled parameters

CIE IO-ST

+560 +160 +106 -9 +40 ?

-10

0

0

-28* +89 +83 +113 ?

-13 +127 +109 +62 ?

-8

0

0

0 0

0 or -2 very weak

?

0 0

+4 +55

IO-ST - inspiratory off-switch threshold, SSR - segmental stretch reflexes. Strong changes are underlined. - - inconsistent changes, 0- absence of effects. *-after data of DiMarco et al., 1981.

normal automatic breathing.

Table 1. shows the summary of the data described above. Lung ventilation in decerebrated cats is determined mostly by the chemoreceptors stimulation; it can be reduced a little by the nasal "flow" receptors. CIE is influenced much by the chemoreceptors and less by the pneumotaxic mechanism and the lung mechanoreceptors. The inspiratory off-switch threshold is lowered by the lung mechanoreceptors stimulation. In conseq\ience Tr is strongly controlled by chemo-receptors, lung mechanoreceptors and by pneumotaxic mechanism. The relatively small effect of chemoreceptor stimulation on TI (-9%) is explained by the fact that the measurement was taken from the level of quiet breathing; in the range from apnoea to hyperpnoea TI can decrease by about 74%. The basic feedback of the diaphragm is provided by the lung mechanoreceptors. The essential reflex response of the intercostal muscles stretch receptors in normal breathing consists in the myotatic reflexes.

REFERENCES

Bradley, G.W., Euler, C.von. Marttila, I., and Roos, B., 1975, A model of the control and reflex inhibition of inspiration in the cat, Biol. Cybernet., 19: 105.

Breslaw, I.S., and Glebovski, V.D., 1981, "Regulation of Breathing", Nauka, Leningrad. (in Russian).

234

Budzinska, K., and Romaniuk, J.R., 1986, Phrenic reflexes in the decerebrated and spinal rabbit, Bull. Eur. Physiopathol. Respir., 22: 65.

Caille, D., Vibert, J.F., and Hugelin, A., 1981, Apneusis and apnea after parabrachial or Kolliker-Fuse N. lesion: influence of wakefulness, Respir. Physiol., 45: 79.

Clark, F.J., and Euler, C. von, 1972, On the regulation of depth and rate of breathing, ~ Physiol., (London), 222: 267.

Cohen, M.I., See, W.R., and Sica, A.L., 1986, Factors influencing the inspiratory-facilitatory response to lung inflation in decerebrate cat, ~ Physiol., (London), 371: 237.

Corda, M., Euler, C. von, and Lennerstrand, G., 1965, Proprioceptive innervation of the diaphragm,~ Physiol., (London), 178: 161.

Decima, E.E., and Euler, C. von, 1969, Excitability of phrenic motoneurones to efferent input from lower intercostal nerves in the spinal cat, Acta Physiol. Scand., 75: 580.

Decima, E.E., Euler, C. von, and Thoden, U., 1969, Intercostal-to-phrenic reflexes in the spinal cat, Acta Physiol. Scand., 75: 568.

DiMarco, A.F., Euler, C. von, Romaniuk, J.R., and Yamamoto, Y., 1981, Positive feedback facilitation of external intercostal and phrenic inspiratory activity by pulmonary stretch receptors, Acta Physiol. Scand., 113: 375.

Eklund, G., Euler, C. von, and Rutkowski, S., 1964, Spontaneous and reflex activity of intercostal gamma motoneurones, ~ Physiol., (London), 171: 139.

Euler, C. von, 1973, The role of proprioceptive afferents in the control of respiratory muscles, Acta Neurobiol. Exp., 33: 329.

Euler, C. von, 1977, The functional organization of the respiratory phase-switching mechanisms, Fed. Proc., 36: 2375.

Euler, C. von, 1986, Brain stem mechanisms for generation and control of breathing pattern, in: "Handbook of Physiology", N.Cherniack and J.G. Widdicombe, ed., Am. Physiol. Soc., Washington.

Euler, C. von, Marttila, I., Remmers, J.E., and Trippenbach, T., 1976, Effects of lesions in the parabrachial nucleus on the mechanisms for central and reflex termination of inspiration in the cat, Acta Physiol. Scand., 96: 324. --

Gautier, H., 1976, Pattern of breathing during hypoxia or hypocapnia of the awake or anesthetized cat, Respir. Physiol., 27: 193.

Gill, P.K., and Kuno, M., 1963, Excitatory and inhibitory action on phrenic motoneurones, ~ Physiol. (London), 168, 258.

Glebovski, V.D., 1962a, On the physiological properties of the sensitive fibres of the diaphragmatic and intercostal nerves, Bull. Exp. Biol. Med., 53: 17.

Glebovski, V.D., 1962b, Stretch receptors of the diaphragm, Sechenov Physiol. ~, 48: 545 (in Russian); 1963, Fed. Proc., Transl. Suppl., 22: T405.

Glebovski, V.D., 1963, Intercostal muscle reflexes to adequate stimulation of pulmonary and thoracic receptors, Sechenov Physiol. ~, 49: 965. (in Russian).

Glebovski, V.D., 1965, Stretch reflexes of intercostal muscles, Sechenov Physiol. ~.51: 1420. (in Russian); 1966, Fed. Proc., Trans!. Suppl., 25: T937.

Glebovski, V.D., 1970, On the diaphragmal proprioceptive reflexes, Sechenov Physiol. ~, 61: 1405. (in Russian).

Glebovski, V.D., 1981, Changes of respiration after the blockade of trigeminal rerves in the decerebrated cats, Sechenov Physiol. ~. 67: 865. (in Russian).

Glebovski, V.D., 1983, Changes of respiration after pontine transections in decerebrated cats, Sechenov Physiol. ~, 69: 445. (in Russian).

235

Glebovski, V.D., and Bajev, A.V., 1984, Stimulation of nasal cavity mucose trigeminal receptors with respiratory airflows, Sechenov Pysiol. ~. 70: 1534. (in Russian).

Glebovski, V.D., and Bajev, A.V., 1986, Depressing effect of carbon dioxide on excitation of nasal cavity cold receptors in cats, Sechenov Physiol. ~. 72: 595. (in Russian).

Glebovski, V.D., and Gizatullina, N.S., 1976, On the relationship between tidal volume and inspiratory duration in decerebrated cats, Sechenov Physiol. ~. 62: 1636. (in Russian).

Glebovski, V.D., and Gizatullina, N.S., 1977a, Relationship between depth of breathing and expiratory duration in decerebrated cats, Sechenov Physiol. ~. 63: 524. (in Russian).

Glebovski, V.D., and Gizatullina, N.S., 1977b, Importance of vagus nerves in breathing frequency changes in decerebrated cats, Sechenov Physiol. ~. 63: 1167. (in Russian).

Glebovski, V.D., and Obukhova, E.A., 1978, The significance of pneumotaxic centers in regulation of inspiratory duration in decerebrated cats, Sechenov Physiol. ~. 64: 818. (in Russian).

Glebovski, V.D., and Pashkevitch, B.P., 1986, The changes of the activity of phrenic motoneurons during lowering of the abdominal wall compliance (on the importance of the excitatory intercostal-to-phrenic reflex), Sechenov Physiol. ~. 72: 1533. (in Russian).

Glebovski, V.D., and Pavlova, N.A., 1962, Diaphragm reflexes in response to adequate stimulation of the receptors of lungs and respiratory muscles, Sechenov Physiol. ~. 48: 1444. (in Russian); 1963, Fed. Proc., Transl. Suppl., 22: T651.

Glebovski, V.D., and Sukhova, G.K., 1983, The effect of the trigeminal nerves blockade on breathing in vagotomized cats, Sechenov Physiol. J,, 69: 1207. (in Russian).

Glebovski, V.D., and Zhdanov, V.A., 1969, Vagotomy effect on lung ventilation in quiet and forced respiration, Sechenov Physiol. ~. 55: 1118. (in Russian).

Hoffmann, P., and Keller, C.J., 1934, Untersuchungen Uber Atemreflexe mit Hilfe Actionsstrome, Ber. Ges. Physiol., 50: 296.

Lim, T.P.K., Luft, U.C., and Grodins, F.S., 1958, Effects of cervical vagotomy on pulmonary ventilation and mechanics, ~ ~ Physiol, 13: 317.

Miserocchi, G., and Milic-Emili, J., 1975, Contribution of hypercapnic stimuli and of vagal afferents to the timing of breathing in anesthetized cats, Respir. Physiol., 25: 71.

Ramos, G.J., and Mendoza, L.E., 1959, On the integration of respiratory movements, II. The integration at spinal level, Acta Physiol., Lat.Amer., 9: 257.

Remmers, J.E., 1970, Inhibition of inspiratory activity by intercostal muscle afferents, Respir. Physiol., 10: 358.

Remmers, J.E., 1973, Extra-segmental reflexes derived from intercostal afferents: phrenic and laryngeal responses, ~ Physiol. (London), 233: 45.

Romaniuk, J.R., Ryba, M., and Grotek, A., 1976, The effects of C02 on the components of breathing pattern, Acta Physiol. Pol., 27: 215.

Sant'Ambrogio, G., and Widdicombe, J.G., 1965, Respiratory reflexes acting on the diaphragm and inspiratory intercostal muscles at the rabbit, ~ Physiol. (London), 180: 766.

Sant'Ambrogio, G., Wilson, M.F., and Frazier, D.T., 1962, Somatic afferent activity in reflex regulation of diaphragmatic function of the cat, ~ ~ Physiol. , 17: 829.

Sears, T.A., 1964, Some properties and reflex connections of respiratory motoneurones of the eat's thoracic spinal cord, ~ Physiol. (London), 175: 386.

236

Shannon, R., 1980, Intercostal and abdominal muscle afferent influence on medullary dorsal respiratory group neurones, Respir. Physiol., 39: 73.

Shimaraeva, T.N., 1976, The role of extravagal reflexes in regulation of diaphragmatic activity after tracheal occlusion or volume changes of the thorax in cats, Sechenov Physiol. ~. 62: 1652. (in Russian).

Shimaraeva, T.N., 1978, The influence of phrenicotomy on extravagal diaphragm reflexes, Sechenov Physiol. ~. 64: 802. (in Russian).

Shimaraeva, T.N., and Glebovski, V.D., 1975, Reactions of phrenic nuclei to chest deformations in the cat, Sechenov Physiol. ~. 61: 1779. (in Russian).

St.John, W.M., 1972, Respiratory tidal volume responses of cats with chronic pneumotaxic center lesions, Respir. Physiol., 16: 92.

St.John, W.M., 1979, An analysis of respiratory frequency alterations in vagotomized decerebrate cats, Respir. Physiol., 36: 167.

Webber, C.L., and Peiss, C.N., 1979, Pentobarbital-induced apneusis in intact, vagotomized and pneumotaxic-lesioned cats, Respir. Physiol., 38: 37.

Widdicombe, J.G., 1964, Respiratory reflexes, in: "Handbook of Physiology", Sect. 3: Respiration, V. I, Am. Physiol. Soc., Washington.

Widdicombe, J.G., and Winning, A., 1974, Effects of hypoxia, hypercapnia and changes in body temperature on the pattern of breathing in cats, Respir. Physiol., 21: 203.

Ya~argil, G.M., 1962, Proprioceptive afferenzen im N. phrenicus der Katze, Helv. Physiol. Acta, 20: 39.

237

CLOSING REMARKS

Being now quite sure that the Symposium has not been a disaster, I would like to express my gratitude to everybody who contributed to its success, and that means all members of the scientific and organ1z1ng committees, speakers, discussants, technicians and - last but not least -my colleagues from the Medical Research Centre.

The list of persons to whom I am particularly indebted is very, very long, but I must mention at least a few names: I have in mind - our old friends, veterans of the 1971 and 1975 Warsaw Symposia Curt von

Euler and Abe Guz; Professor Victor Glebovski, who could not come to our first Symposium but has now finally arrived; leading personalities of this Meeting - Professors Geoffrey Dawes and John Severinghaus; our Italian co-organizers -Professors V. Bellia, G. Bonsignore and E. Lugaresi; Doctors P. Gutkowski, M. Migdal and L. Orlowski from the Child Memorial Health Centre.

Finnaly, I must say that it is difficult to find proper words to thank all - members of "The Warsaw Gang" (to quote its Honorary Member, the man who

gave us this proud name - Curt von Euler) - Urszula Jernajczyk, Joanna Kulesza, Ryszard Romaniuk, Leszek Kubin - who have been the real organizers of this Meeting; Pawel Grieb, Andrzej Kukwa and other girls and boys from our Department;

They have all been working very, very hard - and their work is not yet over, I am afraid.

Thanks are also due to Somebody responsible for the weather: we have been quite lucky, particularly if we take into account the disastrous summer in this part of Europe.

I would like to wish you all the best. I hope we will meet again in not too distant future.


239

PARTICIPANTS

Altose, M.D. (USA) Anderssen, S.H. (N) Becker, H. (FRG) Bellia, V. (I) Berry Borowiecki, B. de, (USA) Bonsignore, G. (I) Brzecka, A. (PL) Burza, J. (PL) Cherniack, N.S. (USA) Czerwosz, L. (PL) Darowski, M. (PL) Dawes, G.S. (GB) Doboszynska, A. (PL) Dutton, R.E. (CDN) Euler, C. von, (S) frank-Piskorska, A. (PL) Gardner, W.N. (GB) Gaultier, Cl. (F) Gautier, H. (F) Gianotti, A. (I) Glebovsky, V.D. (USSR) GJogowska, M. (PL) Gfowicki, K. (PL) Goztikirmizi, E. (TR) Grieb, P. (PL) Gromysz, H. (PL) Gutkowski, P.S. (PL) Guz, A. (GB) HaJuszka, J. (PL) Redner, J. (S) Hurle, M.A. (F) Janczewski, W. (PL) Jernajczyk, U. (PL) Jernajczyk, W. (PL) Jurkiewicz, B. (PL) Kaminski, B. (PL) Karczewski, W.A. (PL) K~pa, L. (PL) Komorowska, A. (PL) Koziorowski, A. (PL) Kubin, L. (PL) Kucharzewski, M. (PL) Kukwa, A. (PL) Kulesza, J. (PL) Lugaresi, E. (I) Marlot, D. (F)

Maszczyk, Z. (PL) Mazurek, H. (PL) Mier, A. (GB) Migdal, M. (PL) Morin-Surun, M.P. (F) Mossakowski, M.J. (PL) MUller, J. (PL) Nar~bski, J. (PL) Nink, M. (FRG) Orlowski, L. (PL) Pack, A.I. (USA) Pantaleo, T. (I) Polanski, L. (PL) Radwan, L. (PL) Ramet, J. (B) Romaniuk, J.R. (PL) Rondio, z. (PL) Rudowski, R. (PL) Ryba, M. (PL) Sanci, S. (I) Severinghaus, J.W. (USA) Shea, S.A. (GB) Siafakas, N. (GR) Sidorowicz, S. (PL) Siegelova, J. (CS) Siesjo, B.K. (S) Skar~ynski, H. (PL) Sknta, L. (PL) Smejkal, V. (CS) Szereda-Przestaszewska, M. (PL) Tendera, M. (PL) Walczak, J. (PL) Ward, D.S. (NL) Wilim, G. (PL) Zielinski, J. (PL)

241

INDEX

ACTtl, see \eurohormones and respiratory control

Adenotomy, see Sleep apnea therapy

Alcohol, 13, 31 Anesthesia

and brai.n metabolism, see Brain metabolism

and metabolic rate, 167-171 and respiration, 107-110,

163-166, 167-171 side effects, 107-110

Anesthetics, 107-109 chloralose, 109, 129, 170, 193 enflurane, 108

and brain metabolism, 105 and respiratory control,

167-171 halothane, 108, 163, 171, 183,

187 ketamine, 105, 108, 167, 190 isoflurane, 108, 171 methohexital, 108 pentobarbital, 102, 149, 177,

199 thiopental, 108, 163

Apnea, s~e Sleep apnea Apnea index, 11, 14, 40, 51,56 Area postrema, 128, 162 Aspartate metabolism, 149-153 Autocorrelation, :203-206

BHtzinger complex, 177-180, 186 Bradycardia, 55, 143, 144, 208 Brain metabolism

of aminoacid neurotransmitters, 149-152

and anesthesia, 105-106 Breathing control

chemical, 33, 127, 163-166, 167-172, 2:28-229,

in the fetus, 133-136 postnatal, 136-137, 210-213 reflex, 173-175

Breathing pattern diurnal variation in the

fetus, 137 Breathing pattern

and larynx in animals, 193-195 in humans, 73

oscillations in, 34, 203-206, 217-224

during sleep, 89-93 spectral analysis during

exercise, 203-206 Breuer-Hering reflex

expiratory-facilitatory, 212-213

inspiratory-inhibitory, 77-81, 126, 189, 211-213

in preterms, 77-81

Cardiac arrhytrunia, 21, 23, 55 Cardiomegaly, 5, 15, 68 Central Pattern Generator (CPG),

119, 123-128, 227-233 Cephalometry, 10, 13, 61-64,

100-103 Cheyne-Stokes breathing, see

Periodic breathing Chemosensitive medullary areas,

se~. Ventral medullary surface

Chloralose, §_e~ Anesthetics Chronic Alveolar Hypoventilation

(CAH), 5, 68 Chronic Obstructive Pulmonary

Disease (COPD), 19-21, 24-25, 47, 49-52, 108

Chronic pulmonary denervation, 89 Cholecystokinin, ~ee

Neuropeptides Conchoplasty, ?e~ Sleep apnea

therapy Continuous Positive Airway

Pressure (CPAP), see Sleep apnea therapy

243

Corticotropin Releasing Factor (CRF), §_~~ Neurohormones and respiratory control

Cotransmission, 121, 125 Cyanosis, 69-70

Daytime somnolence, 9-11, 14, 97 Diaphragm stretch receptors, 232 Diaphragm weakness, 16, 65 Diaphragmatic fatigue, 83-86 Dorsal Respiratory Group (DRG),

§_~~- Respiratory neurons

Endorphin, ~~~ Neuropeptides Enflurane, s_~ Anesthetics Enkephal in, s_e~ Neuropeptides Expiratory time, 187, 194-196,

227

Facial nerve, 183-184, 187 Fentanyl, se~ Opioids exogenous Functional Residual Capacity

(FRC), 20, 210-212

GABA respiratory effects of, 31-32 as neurotransmitter, 111, 121,

123 metabolism, 149-153

Glutamate iontophoresis, 111-114 metabolism, 149-153

Glutamine, 151-152 Glycine, 121, 150-152

Halothane, se~ Anesthetics Heavy Snorers Disease, s_e_e


Hypercapnia and aminoacid metabolism,

149-153 and respiration, 163-166

Hypertension pulmonary arterial, 5, 7, 12,

20, 47 systemic arterial, 5, 7, 13,

58, 208 Hypoglossal nerve, 32, 102,

187-191 Hypothyroidism, 10, 218 Hypoxemia during sleep, 7, 20-22,

s_e_e <t_l_sg S 1 eep apnec1s Hypoventilatioh

chronic, 7, 11 chronic idiopathic, 15-16

Indomethacin, see Prostaglandins Inspiration volume, 173-175

244

Inspiratory-inhibitory reflex, see Breuer-Hering reflex

Inspiratory flow, 173-175 Inspiratory time, 77, 194-196,

227 Intercostal muscles

inhibition during REM sleep, 20 stretch receptors, 232

Intrinsic properties of neurons, 127

Iontophoresis, 111-114, 124-125, 129, 155

Irritant receptors, 32, 214 Isoflurane, se~ Anesthetics

J-receptors, 32, 210-212

KHlliker-Fuse nucleus, 113, 157, 235

Ketamine, see Anesthetics

Laryngectomy, 73 Locus coeruleus, 58

Microiontophoresis, §ee Iontophoresis

~1ethohexital, ~~ Anesthetics Morphine, _§_~e_ Opioids exogenous Myelination of vagal fibers,

209-210 Mylohyoid nerve, 183-186

Naloxone, 31, 108, 126, 130-131, 137' 155-157

Nenrohormones and respiratory control, 159-162

ACTH in the fetus, 136 Neuromodulation, 122-123 Neuropeptides

cholecystokinin (CCK), 111, 114-115, 155-158

endogenous opioids endorphins, 31, 108, 125, 13C enkephalins, 112-113, 116,

125-126, 155-158 neurotensin, 111-117, peptide YY, 125, 128 respiratory effects of, 111,

115-116 substance P, 114-115, 125

Nenrotransmission, 119-123 Neurotransmitters

classical, 121, 125, 128, 149-152

i onotropic, 121 metabotropic, 121

Non-REM sleep, 11-13, 20-22, 24-26, 43, 74, 83-85, 142, 206, 217-224

Nose "flow·· receptors, 231 Nucleus parabrachialis medialis

(NPml), see Kolllker-Fuse nucleus

Nucleus tractus solitarii (NTS), 111-112, 115, 125-129, 155, 162

Obesity, 5, 14, 27, 43 Obstructive Sleep Apnea Syndrome

(OSAS), 5-15, 19-21, 23-25, 47, 49, 67-70, 83, 95, 97, 99

clinical stages, 9-11 epidemiology, 13-14 etiology, 12-13 therapy, ~~-~ Sleep apneas

Oculocardiac reflex, 77, 80 and SIDS, 144

Opioids endogenous, see Neuropeptides

Opioids exogenous

fentanyl, 108-109 morphine, 108-110, 117, 126,

129

Periodic brea-ching, 3'+, 218 (se_e a]~9 Breathing pattern oscillations in)

Pentobarbital, §_~~Anesthetics Phrenic nerve, 112, 177-180,

183-186, 187-191, 227-234 Pickwickian syndrome, 5 Plethysmography, 40, 43, 73, 90,

112, 219 Pneumotaxic centre, 113, 186 (§.~~

also Kolliker-Fuse nucleus)

Pneumotaxic mechanism, 230-231 Polycythemia, 5 PoJysomnography, 7, 15, 40, 43,

49-52, 69 Post-apneic arousal, 7, 20 Prostaglandins, 135

synthetase inhibiotors, 135-136 indomethacin, 135

Pulmonary denervation, 89-93 Pulmonary mechanoreceptors,

187-191, 210-211, 229-230

Rebreathing, 37, 168, 277 Recurrent laryngeal nerve,

193-196 Retrofacial nucleus, 186 REM sleep, 7, 11-12, 14-15,

20-22, 43, 47, 74, 83, 217-224

Respiratory neurons, 111-116, 124-125, 149, 219

Dorsal Respiratory Group (DRG), 112-114, 124-125

spinal, 197-200 Ventral Respiratory Group

(VRG), 112-114, 124-125, 177-180

Respiratory oscillations, 33-34, 217-224

Respiratory Pattern Generator (RPG),~~~ Central Pattern

Generator spinal, 197

Respiratory sensitivity to hypoxia, 31-32, 34

blunted, 20 to hypercapnia, 163-166,

167-170 blunted, 16

Sleep stages, ~~~ Non-REM sleep, REM sleep

Sleep apneas (§.~ also Obstructive Sleep Apnea

Syndrome) central, 7, 16, 33, distribution during sleep

stages, 7, 44-46, 52 mixed, 7

Sleep apneas neural mechanism, 217-224 obstructive, 7, 51, 83 role of trigeminal nerve in,

§_ee Trigeminal nerve therapy, 5, 14-15

by nasal Continuous Positive Airway Pressure (CPAP), 7, l4-15, 24, 39-42

by oxygen administration, 21, 23

pharmacological, 14, 24 surgical, 7, 14, 69, 95-96,

97-98, 99-102, 207-208 by weight reduction, 14, 23

Snoring, 5, 7-9, 11, 13 Sudden Infant Death Syndrome

(SIDS), 141-145 Superior laryngeal nerve,

32, 193-196 Swallowing reflex, 33, 128 Sympathetic activity during

apnea, 12, 55-58

Tachycardia, 7, 55-58 Taurine metabolism, 149-153 Thiopental, s_~_e_ Anesthetics Tidal volume control, 173-175,

187, 227-231

245

Tonsillectomy, s_~~ Upper airway obstruction surgical therapy

Trigeminal nerve, 183-186 TRH, ~~~ Neurohormones and

respiratory control Tube breathing, 163-166

Upper airway muscles, 29-36 Upper airway obstruction

mechanism of, 11-12, 35-36, 61, 68, 99

surgical therapy, 207-208 (~~~ a)_§_~ Sleep apneas therapy surgical)

246

adenotomy, 97-98 tonsillectomy, 68-70, 97-98 conchoplasty, 95-96

Lipper airway patency, 61-64 evaluation, see Cephaiometry

Upper airway skeleton, 14, 99-101

Vagotomy, 209 midcervical, 193-196, 213

Vagus nerve, 77-81, 134, 155, 187, 193, 209-213

Ventral meduliary surface, 30, 113' 155' 158

Ventral Respiratory Group (VRG), see Respiratory neurons

Wakefulness stimulus, 219

Control of Breathing During Sleep and Anesthesia

Documents

Transcript of Control of Breathing During Sleep and Anesthesia