Respiratory frequency control during external elastic loading and chest compression

Respiration

Physiology

Publishing Company, Amsterdam

(1975) 23, 11-22; North-Holland

RESPIRATORY DURING EXTERNAL ELASTIC

FREQUENCY CONTROL LOADING AND CHEST COMPRESSION’

R. SHANNON’ The University

of New Mexico School of Medicine, Department of Physiology, Albuquerque, N. Mex. 87131, U.S.A.

Abstract. Experiments

were conducted

receptors

in the control

are involved

(EEL) and chest compression

(CCjin

to determine of respiratory

(i.e. chest wall) respiratory

if extravagal frequency

the anesthetized

during

(Dial), vagotomized

%o, curves obtained by breathing CO, mixtures were compared loading and chest compression. There was no significant difference cats or dogs indicating

there is no extravagal

respiratory

during

breathing

frequency

show that dogs

mechanoreceptors)

neural

mechanoreceptor

EEL. Comparison

respond

of curves

to chest compression

reflex increase

steady-state

mechano-

elastic loading

cat and dog. Frequency

with curves between

information obtained

external obtained

during

uersus elastic

the CO2 and EEL curves in contributing

to the control

of

with chest compression

and CO2

(i.e. involving

chest wall

with an extravagal

in frequency.

Chest compression

Respiratory

Chest wall

Vagotomy

Mechanoreceptors

Ventilatory

frequency response

to CO2

Evidence in the literature suggests that respiratory rate and depth are regulated to minimize the amount of work (Rohrer, 1925 ; Otis et al., 1950) or force (Mead, 1960) expended by the respiratory muscles in producing the body’s required alveolar ventilation. The chosen pattern of breathing appears to be dependent on the mechanics of the respiratory system (i.e. compliance and resistance). External mechanical loading of the respiratory system, which simulates a change in respiratory mechanics, has been used to study the mechanisms responsible for this regulation. The slowing offrequency and increase in tidal volume with external resistive loading and the increase in frequency and decrease in tidal volume with external elastic loading are well documented (for references, see Bland et al., 1967 ; Freedman et al., 1972 ; Freedman, 1974). These changes in the pattern of breathing during mechanical loading are Accepted for publication 20 August 1974.

’ This work was supported by NIH Grant HL/NS 14843 and institutional funds from GRS Grant 726225. : Department of Physiology, College of Medicine, University of South Florida, Tampa,

’ Current address Fla. 33620, U.SA.

11

12

K. SHANNON

consistent with the hypothesis. Accordingly, the switching to slow, deep breathing during resistive loading reduces resistive energy expended ; whereas, the switching to fast, shallow breathing during elastic loading reduces elastic energy expended. Conscious factors (Freedman, 1974; Margaria et al., 1973) and pulmonary receptors via vagal afferents (Mead, 1960) are known to be involved in these rate adjustments. Bland et al. (1967) presented evidence that extravagal afferent information also contributes to the regulation of frequency and depth of respiration during loading and suggested that ‘chest wall’ mechanoreceptors were the source of the information. It is known that afferent signals from intercostal muscle spindles and tendon organs are altered during mechanical loading and, respectively, facilitate and inhibit intercostal x-motoneurone activity via a spinal reflex (Sears, 1964; Corda et ul., 1965; Newsom Davis and Sears, 1970), but is is not known whether this information affects central respiratory activity during loading. It was recently shown that electrical stimulation of spindle afferents from external intercostal muscles inhibits central inspiratory activity, indicating the existence of pathways from these chest wall mechanoreceptors to the central respiratory centers (Remmers, 1973 ; Remmers et a/., 1973). This study was initially designed to determine if chest wall mechanoreceptors are indeed involved in the modulation of respiratory frequency during steady-state external elastic loading in the anesthetized cat. Preliminary experiments. however. indicated that the extravagal increase in frequency during elastic loading may also be due to increased chemical drive resulting from decreased alveolar ventilation. Experiments were then designed to elucidate the neural and chemical contributions to the reflex increase in frequency during elastic loading in the vagotomized cat ; the frequency responses to inhaled CO, and elastic loads were compared over the same range of increasing arterial CO,. The results of the first study show there is no extravagal mechanoreceptor information contributing to the control of respiratory frequency during external elastic loading in the cat, but that the changes observed are due to increased chemical drive. Since these results appeared to disagree with those of Bland et al., subsequent experiments were conducted to explain the difference. Were the differences due to species of animal and/or type of load used? Bland and co-workers used dogs with chest compression, whereas cats breathing from closed containers were used in this experiment. Subsequent experiments showed that (1) dogs, like cats, have no extravagal mechanoreceptor information contributing to the control of respiratory frequency during external elastic loading, and (2) dogs respond to chest compression with an extravagally mediated reflex increase in respiratory frequency. Methods GENERAL

Experiments were performed on anesthetized (Dial@, 0.6 ml/kg for cats and 0.5 ml/kg for dogs), vagotomized, tracheotomized cats (2.G4.0 kg) and dogs (9.1-10.4 kg). Catheterization of the femoral vein was performed under ether anesthesia in cats and

RESPIRATORY

FREQUENCY

13

CONTROL

under thiopental anesthesia (25 mg/kg) in dogs. Atropine was administered (0.5 mg/kg intraperitoneally) to the cats one-half hour prior to ether induction to reduce respiratory tract secretions. Subsequent surgical and experimental procedures were performed following intravenous administration of the Dial. Rectal temperature was monitored and maintained at 38 “C. Arterial blood pressure was monitored, and arterial blood was periodically analyzed for PoZ, Pco, and pH. These parameters were used as an index of the animal’s general cardiovascular and respiratory status, and all experimental data presented are from animals with normal blood pressure and blood gases. LactatedRinger’s was given to replace the blood extracted for blood gas analysis. Tracheal pressure was monitored to determine respiratory frequency. Arterial and tracheal pressures were recorded on a direct-writing recorder. EXTERNAL

ELASTIC

LOADING

(EEL)

IN

CATS

The experimental set-up is illustrated in fig. 1. A two-way breathing valve was connected to the tracheal cannula, and a load producing container was connected to the inspiratory side of the breathing valve thru a 3-way stopcock. The stopcock allowed the animal to be exposed to the load, to a bag containing gas mixtures with CO, or 100% oxygen, or to room air (by removing the bag). The elastic loads were provided by a glass bottle containing water arranged so that the load could be changed by adjusting the water level in the bottle. Elastances between 0.30 and 0.90 cm H,O/ml were used, which is approximately two to six times the total elastance of the respiratory system of cats(0.15 cm H,O/ml; Croslill and Widdicombe, 1961). In order that the load would be constant for each inspiration, the system was vented through a solenoid valve to atmospheric pressure (via the O,-containing bag) during expiration.

Reservoir So;oc;id III

3-way Stopcock

Trachea Pressure

I valve

Elastic lood Fig. 1. Experimental

set-up.

The respiratory frequency response to inspired CO2 mixtures and elastic loads over the same range of increasing PacoZ was compared as shown in fig. 2. Frequency/ Pac.* (f/Pa,,,) curves were obtained during CO, breathing by exposing the animal to three different concentrations of CO2 (approx. 2,5 and 8%) in OZ. Frequency/Pa,,,

14

K. SHANNON

3or

k

IO

oh/

23

30 pa co2

Fig. 2. Relationship elastic

loading

of respiratory

frequency

to changes

I 35

I 40

I 45

1 50

(mmHg)

in Pa,,,

produced

(EEL). Each panel is data from a single cat and contains First trial : n =CO,,

I

0 = EEL. Second trial : A = CO,,

by CO, breathing the results

and external

of two trials (l&2).

l = EEL.

curves were obtained during elastic loading by exposing the animal to three different loads ; the level of loads varied slightly from animal to animal so that the increase in Paco2 would be in the range obtained during CO, breathing. The following protocol was used. The animal was first titrated with Dial until normal ventilation and blood gases were obtained. Since it has been shown that the dose of Dial used does not eliminate intercostal muscle motor activity (Shannon and Zechman, 1972), there was no concern that the anesthetic would eliminate muscle mechanoreceptor activity. Cervical vagus nerves were sectioned to eliminate the strong vagal effect on respiratory frequency and the animal allowed to recover from the trauma for one hour; this period of recovery resulted in a reasonably stable control respiratory frequency. In order to eliminate the complications of hypoxic drive, 100% oxygen was administered during the control and elastic loading periods, and the inhaled COz ‘was mixed with oxygen. The f/Pa,,, curves were then obtained in the following sequence :

RESPIRATORY

FREQUENCY

CONTROL

15

(1) Control period with 10 minutes of 100% OZ. (2) Seven minutes of breathing each COZ mixture (low concentration to high). (3) Ten minutes reeovery breathing 100% 0, with the last minute used as the control for the elastic loading series. (4) Seven minutes of exposure to each elastic load (small load to large). (5) After twenty minutes recovery breathing air the sequence was repeated. During the CO, breathing and loading periods, frequency stabilized in approximately five minutes, which indicates that the seven minute periods were more than sufficient for reaching a steady state. Arterial blood samples were collected over the last 30 seconds of each seven minute run for determination of Pco2 and pH. The average respiratory frequency was also determined over the last 30 seconds of each run. Series (2) and (4) a b ove were occasionally switched for technical reasons. curves were plotted for sequential series of CO2 breathing and Frequency/Pa,,, elastic loading. Data for plotting the curves were chosen with the following criteria ; (1) the control points on the graph (1 and 3 above) had to be within two breaths/min and one mm Hg of each other, and (2) the frequency and PacoI had to return to this range during recovery after (4) above. It is assumed that if the control points fall within this range, there are negligible changes in the sensitivity of the chemical regulating system and level of anesthesia during the entire sequence. The test points (controls and recovery) were often out of the specified range during the 70 minutes required for a complete sequence; as a result, approximately 50% of the series are reported. The data are interpreted in the following manner. Displacement of the elastic loading f/Pa,o, curve above the curve for COZ breathing would indicate extravagal mechanoreceptor involvement in respiratory frequency control during elastic loading. On the other hand, if the curves were superimposable, no mechanoreceptor involvement could be concluded. EXTERNAL

ELASTIC

LOADING

IN DOGS

The techniques were the same as for cats, with the exceptions that a larger two-way breathing valve and a larger load-producing bottle (EEL) were used. Elastances used were between 0.09-0.19 cm H,O/ml, which is approximately 1.5-3.5 times the total elastance of the respiratory system of dogs (0.056 cm H,O/ml, Crosfill and Widdicdmbe, 1961). The experimental protocol was the same as that for cats. CHEST

COMPRESSION

(cc)

IN DOGS AND CATS

The animal’s chest was lightly compressed by inflating a sphygmomanometer cuff, which was strapped around the chest. Cuff inflation pressures (inside cuff) for dogs were 30,40 and 50 mm Hg ; these pressures are in the range used by Bland et al., thus allowing comparison of our results. Because of the different frequency response in cats and dogs, it was necessary in cats to use inflation pressures between 15 and 40 mm Hg to obtain curves over the range of Paco2 seen during the CO2 breathing. Frequency/ Paco2 curves were obtained for CO, breathing and chest compression using the same

16

K. SHANNON

protocol as for the previously described studies, except that chest compression was substituted in the sequence for EEL. Results EXTERNAL

ELASTIC LOADING

IN CATS

Frequency/P+,,, curves obtained with CO, breathing and elastic loading were compared in six cats (fig. 2). There was variability inthe frequency response to inspired COz between animals, three animals showing an increase, two a decrease, and one no change in frequency with increasing arterial CO,. The different CO, curves obtained in each animal probably reflects the changing level of anesthetic or general deterioration of the animal with time. The generally low control Paco2’s probably reflect the slight hyperventilation that accompanies the light level of anesthesia. Comparison of the f/Pa,-,, curves for CO, breathing and elastic loading shows that the curves are essentially superimposable, regardless of the direction of response or shape of the curve. The data were not subjected to statistical analysis because of the variability between animals and the variability within an animal, as well as the close fit of the EEL and CO, curves. CHEST COMPRESSION

IN CATS

curves from CO2 breathing and chest compression were compared Frequency/Pa,,, in three cats (fig. 3). Chest compression elicited a small increase in frequency above that due to CO, drive in cat-l, a large increase in cat-2 and no difference in cat-3. The 30

____-y----a _

n

20 E

20

F

04)’

25

’ 30

1 35 %02

j

40 (mmtig)

45

50

’ 55

OLI

--

-----

1 25

4

^..

--===w

--

! x)

’ 35 %02

’

’ 40

’

1 45

1 50

’ 55

ImmHgl

Fig. 3. Respiratory frequency/Pa,,, relationship during CO, breathing and chest compression (CC) in vagotomized cats. Cat-2 was also exposed to EEL. The results of two trials are shown for each animal.

RESPIRATORY

FREQUENCY

CONTROL

CAT

17

CONTROL

CHEST COMF? 40 mmHg

RECOVERY

0

I

f = 24/min

Arterul PWZSU~ ImmHp)

100

-

0

CAT 2

f = 17

I

CAT 3

0t

f=21

Fig. 4. Relationship

of respiratory

frequency

from top to bottom

f=20

and blood pressure

are examples

during

chest compression

in cats. Panels

from the cats in fig. 3, respectively.

mechanism responsible for the reflex increase in frequency in cat-l and cat-2 is obviously not active in cat-3. Figure 4 shows arterial blood pressure and tracheal pressure (respiratory frequency) during control, chest compression and recovery periods. The panels from top to bottom are examples from the cats in fig. 3, respectively. Chest compression caused a decrease in blood pressure, probably resulting from a reduced cardiac output. Comparison of figs. 3 and 4 suggests that the reflex increase in respiratory frequency with chest compression in these cats may be related to the decreased arterial blood pressure. The cat showing the largest frequency change (fig. 3, cat-2) with chest compression was also tested with EEL. There was no significant difference between the f/Pa,,, curves during CO2 breathing and EEL. There is also no decrease in blood pressure during EEL. CHEST COMPRESSION

AND ELASTIC LOADING

IN DOGS

The frequency response to chest compression and elastic loading was tested in two dogs and the response to only elastic loading in one dog. These f/Pa,,, curves were

18

R. SHANNON

compared with CO2 response curves. Figure 5 shows no significant difference between f/Pa,,, curves obtained with CO2 breathing and elastic loading. Chest compression elicited an increase in respiratory frequency much greater than obtained with CO2 breathing (fig. 6). The slope of the f/Pa,,, curve with chest

Ok/’

1 30

I

’ 35

ko* Fig. 5. Respiratory

relationship

frequency/Pa,,,

I

’ 40

’ 45

(mmHg)

during

EEL and COZ breathing

in vagotomized

dogs.

9

‘, 20

B d

IO

30

alto

’

25

’ %02

Fig. 6. Respiratory

I

30

’

35 CmmHg)

’

’

40

’

J

45

i

i

r

OLf

I

I

I

25

30

35 (mmHg

I

1 40

I 45

1

relationship during chest compression and CO2 breathing frequency/Pa,,, tomized dogs. The results of two trials are shown for each animal.

in vago-

RESPIRATORY

FREQUENCY CONTROL

19

compression was generally negative and the curves for CO; breathing and elastic loading were positive. Contrary to CO, breathing and elastic loading, the Paco2 decreased during chest compression (an increase in alveolar ventilation). The level of chest compression used did not produce a decrease in blood pressure in dogs. Larger inflation pressures surely would have decreased cardiac output and lowered blood pressure, but they were not necessary because the reflex response was elicited with low inflation pressures. In dogs, as in cats, the shape of the f/Pa,,, curve during CO, breathing varied between animals. The trend, however, is an increased respiratory frequency in vagotomized dogs during CO2 breathing. Discussion

The respiratory frequency versus Pacoz curves during CO2 breathing and steady-state external elastic loading are essentially superimposed in the vagotomized cat and dog. Furthermore, no animal showed an increase in frequency during EEL unless there was an increase with CO2 breathing as well. These results demonstrate that the increase in respiratory frequency observed during external elastic loading in vagotomized animals is due only to the increased chemical drive accompanying the hypoventilation. If there were a neural reflex component, the elastic loading curve would be shifted to the left of the CO, curve. It is thus concluded that there is no extravagal mechanoreceptor information (i.e. from chest wall muscle spindles, tendon organs or joint receptors) modulating respiratory frequency during external elastic loading in the cat or dog. This conclusion agrees with the findings of Mead (1960) in guinea pigs. Freedman (1974) reported that respiratory frequency does not change in response to externally applied elastic (or resistive) loads in anesthetized humans, indicating that extravagal (or vagal) proprioceptive mechanisms do not contribute to frequency control in humans either. Thus, it appears that extravagal proprioceptive mechanisms are not involved in the respiratory pattern changes during steady-state (or long-term) elastic loading in dogs, cats, guinea pigs or humans. There is increasing evidence in the literature which suggests that frequency changes during mechanical loading are due primarily to vagal reflexes in anesthetized animals and cortical influence in conscious humans (Margaria et al., 1973 ; Freedman, 1974). The conclusion that chest wall mechanoreceptors are not involved in frequency control during EEL disagrees with the conclusion of Bland et al. (1967). The results in this study on chest compression in vagotomized dogs agree somewhat with their results, but they did not distinguish between EEL and chest compression in their conclusions. They apparently assumed that chest compression and EEL load the respiratory system in the same manner; that is, the loads similarly affect chest wall mechanoreceptors. Results from this study show this assumption is not valid. As discussed above, respiratory frequency changes during EEL in the vagotomized dog are due to increased chemical drive. During chest compression, the frequency increase above that attributed to CO, drive indicates that mechanoreceptors affecting respiratory frequency are also stimulated. Chest compression thus appears to stimulate

20

R. SHANNON

mechanoreceptors which are unaffected by EEL or alters their output differently than EEL ; one possibility is the muscle spindles of the intercostal respiratory muscles. Remmers (1970) demonstrated that intercostal muscle spindles (T3-T9) can inhibit peripheral inspiratory discharge during chest compression in dogs, but did not indicate whether the stimulus was from inspiratory or expiratory muscles. It has also been suggested that vertebral joint receptors may be involved, but this was ruled out in Remmers’ study. Indirect effects through changes in cardiac output and subsequent baroreceptor stimulation are also ruled out since blood pressure did not change significantly during chest compression in the dogs. There appear to be differences between the cat and dog with regard to chest compression. The application of light pressure to the chest of cats did not elicit the same large increase in frequency and simultaneous decrease in Paco2 noted in dogs. Furthermore, the reflex increase in frequency in the cats occurred primarily at large cuff pressures. No attempts were made in this series of experiments to equate the force and geometric pattern of chest compression in the cat and dog. The extent and character of chest deformation caused by a given cuff pressure is probably different in the two species due to differences in thoracic geometry, chest wall compliance and cuff size relative to chest size. These differences could result in diverse patterns of mechanoreceptor stimulation in the cat than dog, thus negating comparison of results at a given cuff pressure. However, it is assumed that at very light but different cuff pressures the extent and character of chest deformation, and thus mechanoreceptor stimulation, would be similar in the cat and dog. The fact that light pressures did not elicit the same response in cats as dogs strongly suggests that the mechanism responsible for the frequency increase in dogs is not active or is weaker in cats, implying a species difference. The neural reflex increase in frequency during chest compression in cat-l and cat-2 (fig. 3) may be an artifact induced by the drop in arterial pressure, rather than due to stimulation of chest wall mechanoreceptors. There were large decreases in arterial pressure at the higher cuff pressures, and there appears to be some correlation between the drop in arterial pressure and the increase in frequency. It is well known that stimulation of baroreceptors by a decrease in blood pressure can result in an increased respiratory frequency (Widdicombe, 1964). These chest compression studies pose some interesting questions which cannot be answered with this study. Results from this study also allow some conclusions about respiratory frequency control during increased Paco2. Many investigators have suggested that the increase in respiratory frequency in response to carbon dioxide is primarily, if not entirely, dependent on vagal mechanisms (for references, see Clark and von Euler, 1972; Rosenstein et al., 1973). However, it was shown in these experiments that there was an increase in frequency during CO, breathing in some vagotomized cats and all the vagotomized dogs, indicating involvement of an extravagal mechanism. This mechanism is apparently stronger in some animals than others and is definitely of minor importance. The response is both qualitatively and quantitatively different than in the intact animal. In most of the vagotomized animals in which the frequency

RESPIRATORY

FREQUENCY

21

CONTROL

increased with COZ, the f/Pa,,, curve increased over the lower Paco2 range and then plateaued ; whereas, it continues to increase over the observed Paco2 range in intact animals. Since Marsh et al. (1973) reported that peripheral chemoreceptors do not contribute appreciably to CO, drive when desensitized by exposure to 1OO’A02, it may be concluded that a mechanism of central origin (i.e. central chemoreceptors or direct effect on central respiratory centers) is responsible for the increase in frequency during CO2 breathing in vagotomized, hyperoxic animals. Acknowledgements

The author is very gratefu! for the excellent technical assistance of Sharon Jennings and Louie Gallegos.

References Bland, S., L. Lazerou,

G. Dyck and R. M. Cherniack

(1967). The influence

rate and depth. Respir. Physiol. 3 : 47-54. Clark, F. J. and C. von Euler (1972). On the regulation

of the ‘chest wall’ on respiratory J. Physiol. (London)

of depth and rate of breathing.

222: 267-295. Corda,

M., G. Eklund

changes

and C. von Euler (1965). External

in respiratory

Crosfill, J. L. and J. G. Widdicombe breathing Freedman,

in different

S., K. J. Dalton,

respiratory Freedman,

response

(1961). Physical

characteristics

and phrenic

a-motor

responses

to

D. Holland

and J. M. S. Patton

to CO2 in man. Respir. Physiol.

by L. D. Pengelly,*A.

of the chest and lungs and the work of

species. J. Physiol. (London

mammalian

S. (1974). The effects of added loads in man

edited

intercostal

load. Acta Physiol. Sand. 63 : 391-400.

S. Rebuck

) 158

: l-14.

(1972). The effects of added elastic loads on the 14: 237-250.

: Conscious

and anesthetized.

and E. J. M. Campbell.

Ontario,

In: Loaded Longman

Breathing,

Canada

Ltd..

pp. 22-25. Margaria,

C. E., S. Iscoe, L. D. Pengelly, J. Couture,

response

to elastic loads and positive pressure

Marsh, R. H. K., K. R. Lyen, G. A. D. McPherson, by-breath

effects of imposed

alternate-breath

H. Don and J. Milic-Emili

(1973). Immediate

ventilatory

in man. Respir. Physiol. 18 : 347-369. S. B. Pearson oscillations

and D. J. C. Cunningham

of alveolar

CO,.

(1973). Breath-

Respir. Physiol. 18: S&91.

Mead. J. (1960). Control of respiratory frequency. J. Appf. Physiol. 15: 325-336. Newsom Davis, J. and T. A. Sears (1970). The proprioreceptive reflex control of the intercostal

muscles

during their voluntary activation. J. Phrsiol. (London) 209: 711-738. Otis, A. B., W. 0. Fenn and H. Rahn (1950). Mechanics of breathing in man. J. Appl. Physiol. 2 : 592-607. Remmers,

J. E. (197Q). Inhibition

of inspiratory

activity

by intercostal

muscle afferents.

Respir. Physiol. 10:

358-383. Remmers,

J. E. (1973). Extra-segmental

responses. Remmers,

J. E., I. Marittila

of anesthetized Rohrer,

reflexes derived

from intercostal

afferents:

Phrenic

F. (1925).

and C. von Euler (1973). Effect of afferent stimulation

on the respiratory

oxygen

rhythm

cats. Physiologist 16: 432. Physiologie

der Atembewegung.

In: Handbuch

der normalen

und pathologischen

Physiologie, edited by A. T. J. Bethe et al. Berlin, Springer, 2: 7&127. Rosenstein, R., L. E. McCarthy and H. L. Borison (1973). Rate versus depth of breathing alveolar

and laryngeal

J. Physiol. (London) 233 : 45-62.

in decerebrate

independent

of

cats. Respir. Physiof. 19: 8&87.

Sears, T. A. (1964). Some properties and reflex connections spinal cord. J. Physiol. (London) 175 : 386403.

of respiratory

motoneurons

of the cat’s thoracic

‘22

K. SHANNON

Shannon, R. and F. W. Zechman (1972). The reflex and mechanical increased airflow resistance. Respir. Physiol. 16: 51-69. Widdicombe,

J. G. (1964). Respiratory

reflexes. In : Handbook

edited by W. 0. Fenn and H. Rahn. Washington,

response

of Physiology.

D.C., American

of the inspiratory

muscles to an

Section 3. Respiration.

Physiological

Vol. 1,

Society, pp. 585-630.