624
Carotid Sinus Baroreceptor Reflex Control of Respiration
Martha J. Brunner, Marc S. Sussman, Andrew S. Greene, Clayton H. Kallman, and
Artin A. Shoukas
From the Department of Biomedical Engineering, The Johns Hopkins University School of Medicine, Baltimore, Maryland
Downloaded from http://circres.ahajournals.org/ by guest on June 16, 2017
SUMMARY. We have studied the effect of the carotid sinus baroreceptor reflex on respiration in 10
vagotomized, spontaneously breathing, pentobarbital anesthetized dogs. The carotid body chemoreceptor reflex response was eliminated by surgically excluding the carotid bodies from the carotid
sinus baroreceptor area. Steady state frequency, tidal volume, and minute ventilation were measured
after 25 mm Hg step changes in intrasinus pressure between 50 and 200 mm Hg. Over this range,
the step decreases in intrasinus pressure caused concomitant increases in mean arterial pressure
from 86 to 182 mm Hg. All of the respiratory response curves were sigmoidal in shape. Decreasing
intrasinus pressure from 200 to 50 mm Hg caused respiratory frequency to increase from 4.8 to 9.7/
min, and tidal volume to decrease from 704 to 515 ml. The calculated total ventilation, however,
increased from 3180 to 4530 ml/min. The time of inspiration decreased from 3.7 to 2.4 seconds, and
the time of expiration decreased from 9.8 to 4.1 seconds. These ventilatory responses are shown to
be baroreceptor reflex mediated, and not secondary to changes in arterial pressure. These findings
indicate that not only does the carotid sinus baroreceptor reflex control arterial pressure, but it also
simultaneously influences ventilation, through changes in both respiratory frequency and tidal
volume. (Ore Res 51: 624-636, 1982)
AS EARLY as 1932, Schmidt presented evidence that
the carotid sinus baroreceptor reflex can influence
respiratory function. However, there has been little
information published regarding the quantitative extent of carotid sinus baroreceptor reflex control of
respiration over the full working range of carotid
sinus baroreceptor pressures. In order to study the
baroreceptor reflex control, various input forcings or
stimuli such as carotid occlusion, aortic balloon inflation or hemorrhage, have been used, and have been
found to cause changes in ventilatory drive (Bishop,
1974; Grunstein, 1975; Delpierre, 1974; Miserocchi,
1980). However, in all these studies, it was difficult to
separate the specific effects of the baroreceptors from
those of the chemoreceptors or other reflex mechanisms. Heistad et al. (1975) have shown that baroreceptor and chemoreceptor reflexes interact closely in
the control of ventilation. The chemoreceptor reflex
ventilatory response to cyanide injections was increased at low intrasinus pressures and decreased at
high intrasinus pressures.
Therefore, to understand clearly the carotid sinus
reflex control of respiration, it was necessary to stimulate selectively only the carotid sinus baroreceptors,
and not chemoreceptors. To this aim, we used peripherally chemodenervated, vagotomized dogs. The purpose of this study was to examine, in a quantitative
manner, steady state baroreceptor reflex influence on
ventilatory function over the full working range of
carotid sinus pressures. In addition, we performed
experiments to test that these reflex changes were not
secondary to changes in arterial pressure.
Methods
Surgical Procedures
In the first series of experiments, 13 mongrel dogs (19.7
± 1.3 kg) of both sexes were anesthetized with sodium
pentobarbital (30 mg/kg, iv). A constant plane of anesthesia
was maintained by infusion of the anesthetic through a
catheter placed in the femoral vein. A catheter was placed
in the femoral artery and connected to a pressure transducer
(Statham P23AC) for the measurement of arterial pressure.
The trachea was intubated with a cuffed endotracheal
tube which was connected to a Fleisch pneumotachograph
(Gould model 7319). Ventilatory flow was continuously
monitored with a differential pressure transducer (Statham
PM 15 ETC). Inspiratory and expiratory flow were electronically integrated, using a Brush integrator (model 13-461570) to give tidal volumes. Tidal volumes were calibrated by
injection of known amounts of air through the pneumotachograph. The dogs were heparinized (7000 units) and
allowed to breathe spontaneously throughout the course of
the experiment. In addition, body temperature was maintained at 38 ± 0.5°C.
The vagi were cut in the mid-cervical region via a midline
incision in the neck. Vagotomy resulted in a pattern of
breathing characterized by a decreased frequency and increased tidal volume. Lim et al. (1958) showed that, although
vagotomy changed mechanical properties of the lung, total
ventilation did not change significantly. However, they also
showed that as a result of the altered pattern of breathing,
alveolar ventilation increased. Bilateral vagotomy was performed in order to eliminate the possible effects caused by
aortic arch baroreceptors and chemoreceptors, and the secondary effect from lung inflation and other cardiopulmonary receptors.
Figure lA shows the method for selective isolation of the
Brunner et a7./Baroreceptor Control of Respiration
625
Lingual a.
Carotid
sinus n.
Occipital a
Carotid
body
Downloaded from http://circres.ahajournals.org/ by guest on June 16, 2017
intrasinusal
pressure
REFERENCE
VOLTAGE
GENERATOR
B
Figure 1. Preparation of selective isolation
of carotid sinus baroreceptors. Panel A:
Surgical isolation of baroreceptors. The
carotid body is excluded from the area
exposed to changes in intrasinus pressure.
CCA. = common carotid artery. l.CA.
= internal carotid artery. E.C.A. = external
carotid artery. Panel B: Servo-controlled
pressure generating system for controlling
intrasinus pressure. R.CS. = right carotid
sinus. L.C.5. = left carotid sinus.
ELECTRIC
MANOMETER
SYRINGE PUMP
SERVO-AMPLIFIER
and
MOTOR DRIVING
SYSTEM
PRESSURE
TRANSMISSION
carotid sinus baroreceptors. The internal and external carotid arteries and any small branches originating from the
carotid bifurcation were completely ligated. A catheter was
placed in each lingual artery. The two lingual catheters were
joined and connected to a pressure transducer (Statham
P23AC) for recording intrasinus pressure. Sodium cyanide
(2 ml of 0.003 N) was injected into the carotid bifurcation
through the lingual catheter. The occipital artery then was
ligated at its origin from the external carotid, and the
cyanide injection was repeated in order to test the completeness of removal of the carotid body from the pressoreceptor
area. Control injections of 2 ml of saline were also performed before and after occipital ligations. By ligation of
the occipital artery at this point, the blood supply to the
carotid body is interrupted, and the carotid body is excluded
from the "blind sac" (Schmidt, 1932). Thus, the carotid
body is not exposed to changes in intrasinus pressure. Any
branches of the occipital artery distal to the carotid body
C
COMMON
CAROTID
AUTEHIES
were also ligated to prevent backflow into the carotid body.
On two occasions, the carotid sinus baroreceptors were
inadvertently denervated as a result of these procedures.
The immediate reflex hypertensive response was seen and
there were no further arterial pressure changes in response
to carotid occlusion. These two animals were not used for
this study. One other animal whose occipital artery arose
from the internal carotid artery distal to the carotid bifurcation was also excluded.
When the carotid sinus baroreceptor isolation was complete, the left common carotid artery was tied, and a fourway connector was attached to the distal segment of each
common carotid artery, the proximal end of the right common carotid artery, and a servocontrolled, nonpulsatile
pressure-generating system (see Fig. IB). This system allowed normal blood pressure until the proximal end of the
right common carotid artery was clamped when intrasinus
pressure was to be controlled (Shoukas and Sagawa, 1973).
626
In the second series of experiments, five dogs (19.6 ± 1.3
kg) were prepared as described above. In addition, both
femoral arteries were cannulated and connected to an external reservoir, the height of which was adjusted to maintain arterial pressure constant.
For both series of experiments, pulsatile and mean arterial pressure, intrasinus pressure, ventilatory flow, and integrated tidal volume were recorded using ink recorders
(Brush models 2800 and 2400).
Experimental Protocol
Downloaded from http://circres.ahajournals.org/ by guest on June 16, 2017
In the first series of experiments, intrasinus pressure was
initially held constant at 125 mm Hg. At 5- to 10-minute
intervals, respiratory frequency, tidal volume, and total
ventilation were monitored. Over the course of the next 3060 minutes, and at intrasinus pressure equal to 125 mm Hg,
the pentobarbital infusion rate was adjusted until the monitored respiratory variables had reached steady values. The
rate of infusion which ranged from 25 to 150 mg/hr was
then maintained constant over the subsequent 1-hour experimental period during which intrasinus pressure was
step-changed. If, at the end of the 1-hour experimental
period, the respiratory variables returned to within ±15%
of the original control values, we considered that a constant
level of anesthesia had been sustained during the experimental run.
Arterial blood samples were taken every 15 minutes
during the control period. Arterial blood Pco2, P02, and pH
were measured with a blood gas analyzer (Instrumentation
Laboratory). Supplemental oxygen was given to maintain
the Pao2 at or above 100 mg Hg. Arterial pH was maintained
at 7.4 by infusion (50 ml/hr) of 1% sodium bicarbonate
when necessary, but never during an experimental run.
In series I dogs, intrasinus pressure was step changed
between 50 and 200 mm Hg. Between each step change in
intrasinus pressure, the dog was allowed to reach a steady
state. Steady state values of mean arterial pressure, frequency, tidal volume, and ventilation were monitored.
In some dogs, following the first experimental run, intrasinus pressure was again held constant for another control
period while the rate of anesthetic infusion was again
adjusted. At this new anesthetic infusion rate, a second run
was performed, starting at an intrasinus pressure of either
50 or 200 mm Hg. Intrasinus pressure was increased and
decreased over the same range as before. For each of the 10
dogs, either one experimental run or the average of two
runs is reported.
Arterial blood samples were taken at intrasinus pressures
of 50, 125, and 200 mm Hg for blood gas analysis.
In the second series of experiments, intrasinus pressure
was held constant at 125 mm Hg for a similar control period
until respiratory variables reached steady values. Intrasinus
pressure was step changed to only three different levels: 50,
125, and 200 mm Hg. Intrasinus pressure was changed
under two conditions: first, when arterial pressure was
allowed to vary, as in the previous experiments, and second,
when arterial pressure was held constant at 100 mm Hg.
For the first series of experiments, data analyses were
performed in two steps. In the first step, we examined the
shape of the relationship of each variable with intrasinus
pressure. Data recorded when intrasinus pressure was increasing and decreasing were analyzed separately. The repeated measures analysis of variance was applied using the
intrasinus pressure factors with seven levels (50 through
200 mm Hg). All variables demonstrated overall significant
differences (at least P < 0.05) within animals, i.e., between
intrasinus pressure levels. Linear, quadratic, and cubic orthogonal contrasts were used to test for linear, quadratic,
Circulation Research/Vo/. 51, No. 5, November 1982
and sigmoidal relationships. A sigmoidal relationship with
intrasinus pressure was found for all variables. The associated P values for these sigmoidal relationships are presented
here.
In a previous paper (Shoukas and Brunner, 1980), we had
described hysteresis in the arterial pressure response to
changes in carotid sinus pressure. We expected that we
might see similar behavior in the respiratory responses to
the same stimuli. To test for a hysteresis difference between
responses to increasing and decreasing intrasinus pressure,
only the 100,125, and 150 mm Hg intrasinus pressure levels
were examined. A repeated measures analysis of variance
was used with one direction of change factor (with two
levels) and one intrasinus pressure factor (with three levels).
The repeated measures analysis assumes that the intrasinus
pressure and direction of change factors are fixed and that
the dog (subject) factor is random. The model used is
saturated, i.e. includes all two- and three-way interactions.
Inspiratory and expiratory times were determined in only
five of the dogs. Values obtained at intrasinus pressures of
50 and 200 mm Hg were compared using a f-test on paired
data.
In the second series of experiments, arterial pressure,
respiratory frequency, tidal volume, and ventilation at intrasinus pressures of 50, 125, 200 with arterial pressure
uncontrolled and controlled were compared by two-way
analysis of variance.
The maximum changes in each variable over the full 50
to 200 mm Hg range for the uncontrolled and controlled
arterial pressure conditions were compared by paired r-test.
For all analyses, a P value less than 0.05 was considered to
be significant.
Results
Figure 2 shows the typical responses to saline and
sodium cyanide injection performed at the time of
selective isolation of carotid sinus baroreceptors from
the chemoreceptors. Before the occipital artery was
tied, intracarotid injection of saline, but particularly
sodium cyanide, caused carotid chemoreceptor reflex
increases in ventilatory How, respiratory frequency,
and mean arterial pressure. After ligation of the occipital artery, the responses to saline and cyanide
injection were abolished. We therefore considered
that the carotid body chemoreceptors had been excluded from the area exposed to the changes in intrasinus pressure.
Figure 3 shows a recording during 75 mm Hg step
changes in intrasinus pressure between 50 and 200
mm Hg. These large steps were used for the purpose
of illustration of clear-cut changes in respiration. Decreasing intrasinus pressure from 200 to 125 mm Hg
caused the characteristic reflex increase in mean arterial pressure from 60 to 107 mm Hg. The time
constant of the response was 8-12 seconds. Respiratory frequency increased from 5.5 to 7.3 breaths/min.
Tidal volume decreased from 670 to 530 ml. Ventilation increased from 3690 to 3870 ml/min. Respiratory
changes occurred with a similar onset of responses.
Decreasing intrasinus pressure from 125 to 50 mm Hg
caused a further increase in mean arterial pressure
(107 to 160 mm Hg), an increase in respiratory frequency from 7.3 to 10.4/min, and in an increase in
ventilation from 3870 to 4890 ml/min. Furthermore,
there was a decrease in tidal volume (530-470 ml).
Brunner ef ai./Baroreceptor Control of Respiration
Safine Inj.
627
Sodium
Cyanide Inj.
Saline Inj.
Saline Inj.
Occipital Ligation
Sodium
Cyanide Inj.
250Pulsatile
Arterial Pressure
(mm Hg)
Respiratory
Flow
(I/sec)
;
250Mean
Arterial Pressure
(mm Hg)
125
Downloaded from http://circres.ahajournals.org/ by guest on June 16, 2017
0—
t
2 min
Figure 2. The ventilatory responses to saline and sodium cyanide before and after occipital artery ligation. See text for explanation.
Increasing the intrasinus pressure resulted in opposite
changes in all the measured variables.
Arterial Blood Cases
Listed in Table 1 are the mean values for arterial
PaQ2, Paco2 and pH measured at intrasinus pressures
of 50, 125, and 200 mm Hg. Blood gas measurements
were obtained at the three intrasinus pressures for
most of the dogs. Although the vagotomy resulted in
a slight respiratory alkalosis (Lim, 1958), the dogs
maintained good acid-base balance through the
course of the experiment. Along with the resulting
Pulsatile
Arterial
Pressure
(mm Hg)
Intrasinus
Pressure
(mm Hg)
125-
Respiratory
Flow
250Mean Arterial
Pressure (mm Hg)
f
125-
1000-
INSPIRATIOH
500Tidal
Volume (ml)
II
lWfll
, ! i i l l i . .1
EXPIRATION
f
1 min
figure 3. Recording of arterial pressure and ventilatory changes in response to changes in intrasinus pressure. See text for details.
628
Circulation Research/Vo/. 51, No. 5, November 1982
TABLE1: 1
Arterial Blood Gas Values Measured at Three Intrasinus
Pressures
ISP == 50
(n = 9)
ISP = 125
(n = 9)
ISP = 200
(n = 7)
Pao2 (mm Hg)
128 ± 10
122 ± 28
102 ± 11
Paco2 (mm Hg)
26.5 ± 2.0
30.8 ± 2.8
27.0 ± 3.1
PH
7.46 ± 0.08
7.42 ± 0.06
7.37 ±0.09
changes in ventilation caused by changes in intrasinus
pressure, there were predictable changes in arterial
blood gas values.
Series I: Arterial Pressure Response to Changes in
Intrasinus Pressure
Downloaded from http://circres.ahajournals.org/ by guest on June 16, 2017
As intrasinus pressure was decreased from 200 to
50 mm Hg, mean arterial pressure increased significantly from a mean value of 85.9 ± 7.4 to 181.9 ±
18.5 mm Hg (n = 10) in a sigmoidal manner (P <
0.001). When intrasinus pressure was increased from
50 to 200 mm Hg, the reflex decrease in mean arterial
pressure from 183.0 ± 18.0 to 88.3 ± 7.8 mm Hg was
also sigmoidal in shape (P < 0.001), but followed a
curve that always fell below the curve for decreasing
intrasinus pressure (P < 0.01). The hysteresis in the
arterial pressure-intrasinus pressure relation curve
was seen consistently in all the animals.
Figure 4 shows mean arterial pressure loops for all
10 animals (panel A and B). Panel C shows mean
values for the 10 dogs. Listed in Table 2 are the mean
values of mean arterial pressure at each intrasinus
pressure.
Response of Respiratory Frequency to Changes in
Intrasinus Pressure
Stepwise decreases in intrasinus pressure from 200
to 50 mm Hg caused increases in respiratory frequency from 4.8 ± 0.53 to 9.7 ± 1.31 breaths/min.
Changes in intrasinus pressure resulted in consistent
changes in respiratory frequency. The dependency of
frequency upon intrasinus pressure was found to be
significant (P < 0.001). The shape of the frequencyintrasinus pressure relation curve was sigmoidal (P
< 0.01). Using the values of frequency at 100, 125,
and 150 mm Hg intrasinus pressure, we could not
demonstrate a statistical difference between frequency responses to increasing versus decreasing intrasinus pressure. Therefore, in panels A and B of
Figure 5, frequency responses for individual dogs for
increasing and decreasing pressure were pooled together. However, at intrasinus pressure of 100 mm
Hg alone, frequency responses when intrasinus pressure was decreasing were found to be significantly
larger than responses to increasing intrasinus pressure
(P < 0.02). Figure 5 shows respiratory frequency
responses from dogs 1 through 5 (panel A), 6 through
10 (panel B), and mean values for 10 dogs (panel C).
Dogs # 1 and 4 exhibited nearly identical mean
arterial pressure responses to changes in intrasinus
pressure. However, their respiratory frequency responses were quite different. First, the baseline values
of frequency at intrasinus pressure of 125 mm Hg
were very different (4.9 for dog # 1 and 14.0 for dog
#4). Also, their sensitivities to changes in intrasinus
pressure were different. Over the full range of intrasinus pressures tested, dog #4 changed frequency
from 19.7/min at intrasinus pressure of 50 to 6.9/min
at intrasinus pressure of 200 mm Hg. Dog #1 changed
frequency from 5.8 to 4.2/min for the same change in
intrasinus pressure. We believe that this variability in
the baseline values and in the sensitivity of reflex
changes in ventilation is probably due to differing
levels of anesthesia maintained over the experimental
run. Although each dog was maintained at a constant
anesthetic level, dog #1 appeared to be more deeply
anesthetized than dog #4. The reflex changes in respiratory variables appeared to be much more sensitive
to the level of anesthesia than did the mean arterial
pressure responses.
Tidal Volume Response to Changes in Intrasinus
Pressure
Figure 6 shows tidal volume responses to a change
in intrasinus pressure. As intrasinus pressure was
stepwise decreased from 200 to 50 mm Hg, the tidal
volume decreased from 704 ± 47 to 515 ± 49 ml.
Tidal volume responses were found to be dependent
upon changes in intrasinus pressure (P < 0.001), and
sigmoidal in shape (P < 0.01). We could not demonstrate statistically significant hysteresis in the tidal
volume response to increasing vs. decreasing intrasinus pressures. Increasing and decreasing intrasinus
pressure responses at a given intrasinus pressure were
averaged and are shown in Figure 6, A and B.
As in the case of respiratory frequency, there are
differences in the responses of different animals. At
an intrasinus pressure of 125 mm Hg, dog #4 was
breathing at a higher frequency and lower tidal volume than dog #1. The tidal volume response of dog
#4 to changes in intrasinus pressure (i.e., the sensitivity of the reflex response) was greater than dog # 1 .
Again, although these differences in baseline values
and sensitivities are present, changes in intrasinus
pressure caused consistent changes in tidal volume.
Minute Ventilation Response to Changes in
Intrasinus Pressure
Responses of minute ventilation (frequency times
tidal volume) are plotted in Figure 7. Stepwise reductions in intrasinus pressure from 200 to 50 mm Hg
resulted in increases in minute ventilation from 3180
± 210 to 4530 ± 250 ml/min. For the ventilatory
response, it can also be seen that there exists a wide
variability in the baseline ventilations at the intrasinus
pressure of 125 mm Hg and that there are differences
in the sensitivity of the reflex response among individual animals. Minute ventilation was dependent
Brunner et ai./Baroreceptor Control of Respiration
629
340-
320220-,
220-
200-
200-
180-
180-
160-
160-
a. 140"
140-
£
UJ
a.
120-
Q.
120-
Downloaded from http://circres.ahajournals.org/ by guest on June 16, 2017
a.
a.
in
UJ
5 100-
100-
80
80-
60
60-
40-
A
50
4075
100
125
150
INTRASINUS PRESSURE
175
200
B
50
75
100
125
150
INTRASINUS PRESSURE
175
200
Figure 4. Mean arterial pressure responses to changes in intrasinus
pressure in dogs 1-5 (panel A), 6-10 (panel B), and mean values
for 10 dogs (panel C). Arrows indicate direction of intrasinus
pressure change.
200-I
180-
160-
upon intrasinus pressure (P < 0.001), and the relation
curve was sigmoidal in shape (P < 0.05). There is an
apparent minute ventilation response difference between increasing and decreasing intrasinus pressure,
but comparing values at intrasinus pressure of 100,
125, and 150 mm Hg, these differences were slightly
larger than the 5% significance level (P < 0.08). However, direct comparison of ventilation responses at
100 mm Hg intrasinus pressure shows a significantly
greater (P < 0.005) ventilation when intrasinus pressure was decreasing than when intrasinus pressure
was increasing.
140o>
I
E
120UJ
a.
3
UJ
ioo-
£ 8060-
Inspiratory and Expiratory Durations
40-
20-
50
75
100
125
150
INTRASINUS PRESSURE
175
200
The inverse of respiratory frequency, the respiratory period, varies from 13.7 ± 2.6 sec at an intrasinus
pressure of 200, to 6.7 ± 0.7 sec at an intrasinus
pressure of 50 mm Hg. In five dogs, the partition of
respiratory period into inspiratory and expiratory
phases was studied.
The duration of inspiration varied as a function of
intrasinus pressure, exhibiting a curve which was
630
Circulation Research/Voi. 51, No. 5, November 1982
TABLE 2
Responses to Changes in Intrasinus Pressure
Mean arterial pressure
(mm Hg)
n = 10
Intrasinus
pressure
200
175
150
125
100
Inc.
Dec.
85.9
±7.4
92.6
±9.0
101.3
±9.4
134.8
±9.6
176.5
88.3
±7.8
83.8
±7.7
87.2
±8.4
105.5
±10.0
145.5
±9.0
177.2
±18.2
183.0
±18.0
±7.5
75
Downloaded from http://circres.ahajournals.org/ by guest on June 16, 2017
50
191.4
±21.0
181.9
±18.5
Respiratory
frequency
Tidal
volume
Minute
ventilation
Duration of
inspiration
Duration of
expiration
(min ')
n = 10
(ml)
n = 10
(ml/min)
n = 10
(sec)
n =5
(sec)
n =5
4.8
704
±0.53
±47
3180
±210
3130
±208
3320
±284
3910
±378
4390
±332
4600
±280
4530
±250
4.9
691
±0.63
±53
5.2
668
±0.77
±69
7.1
615
±1.16
±56
9.1
9.8
537
±51
514
±1.23
±51
±1.04
9.7
515
±1.31
±49
similar in shape to that of the respiratory period (see
Fig. 8). At an intrasinus pressure of 200 mm Hg,
inspiratory duration was 3.7 ± 0.33 sec, and at an
intrasinus pressure of 50 mm Hg, inspiratory duration
was 2.4 ± 0.17 seconds. A r-test on paired data showed
a statistically significant (P < 0.01) difference between
these two values of intrasinus pressures in the five
dogs.
Expiratory duration was also affected by changes
in intrasinus pressure. The duration of expiration at
an intrasinus pressure of 200 mm Hg was 9.8 ± 1.6
sec, and at an intrasinus pressure of 50 mm Hg,
expiratory duration was 4.1 ± 0.3 sec. For expiratory
durations, a r-test using paired data showed a significant difference (P < 0.05) between the two values of
intrasinus pressure.
Series II: Influence of Arterial Pressure
Figure 9 shows a chart recording of a typical experiment. In panel A, intrasinus pressure was increased
when arterial pressure was allowed to decrease. Respiratory frequency and ventilation decreased, and
tidal volume increased. Panel B shows the response
in the same dog when arterial pressure was controlled
at 100 mm Hg. The response is very similar to that
when arterial pressure varied.
Results of series II experiments (n = 5 dogs) are
listed in Table 3. Regardless of whether arterial pressure was controlled, changes in intrasinus pressure
always resulted in statistically significant changes in
respiratory frequency, tidal volume, and ventilation.
Changes in each variable, over the full range of
values (50 to 200 mm Hg intrasinus pressure) were
compared for controlled and uncontrolled arterial
pressure conditions using a paired f-test. Changes in
tidal volume and ventilation were not significantly
different when arterial pressure was controlled. However, changes in respiratory frequency were signifi-
3.7
±0.33
3.6
±0.33
3.4
±0.33
3.1
±0.30
2.6
±0.23
2.5
±0.20
2.4
±0.17
9.8
±1.6
10.1
±1.6
9.8
±1.7
7.5
±1.6
4.9
±0.6
4.1
±0.5
4.1
±0.3
cantly greater (P < 0.05) when arterial pressure was
controlled than when it was allowed to vary.
Discussion
The results presented demonstrated that the carotid
sinus baroreceptor reflex controls ventilation through
changes in both frequency and tidal volume. In addition, both inspiratory and expiratory durations were
shown to be affected by the carotid sinus baroreceptor
reflex.
This study employed a vagotomized preparation.
The advantages of this approach are that we could
eliminate vagal afferents which mediate aortic baroreceptor, chemoreceptor, and lung inflation reflexes.
However, we could not eliminate inputs from nonvagally mediated afferents, and it is possible that
reflex effects mediated by these afferents could be
activated as a result of changes in pulmonary pressures and volumes, and could contribute to steady
state ventilatory changes. However, the ventilatory
responses to changes in intrasinus pressure occur
within the first breath and therefore could not be due
solely to changes in the pulmonary circulation.
The stimuli used in this study were confined specifically to the carotid sinus baroreceptors. Although
the blood supply to the carotid bodies was eliminated,
it is possible that they continued to function. At the
end of two experiments, we injected sodium cyanide
directly into the carotid body through a distal branch
of the occipital artery. There was no ventilatory or
circulatory response to the injection at this point,
some 4 hours after the isolation procedure. If any
peripheral chemoreceptor activity had remained during the course of the experiment, it would tend to
oppose the action of the carotid sinus baroreceptor
responses measured. For example, as intrasinus pressure was decreased from 200 to 50 mm Hg, systemic
Brunner et al./Baroreceptor Control of Respiration
631
20-.
20-
19-
19-
18-
18-
17-
17-
16-
16-
15-
15-
14-
14-
f
13-
13-
|
12-
12-
>
u
a
11-
IV
10-
10a
9-
cr
o
98-
8O
7-
5
Downloaded from http://circres.ahajournals.org/ by guest on June 16, 2017
6-
a.
ES
5-
a.
4-
7654-
3-
3-
2-
2-
1-
1-
0
0
50
75
100
125
150
175
200
INTRASINUS PRESSURE ( mm Hg )
B
50
75
100
125
150
175
200
INTRASINUS PRESSURE ( mm Hg )
15-
Figure 5. Respiratory frequency responses to changes in intrasinus
pressure dogs 1-5 (panel A), 6-10 (panel B), and mean values for
10 dogs (panel C). Arrows indicate direction of intrasinus pressure
change in panel C.
1413
12IV
10-
a
UJ
EC
CC
o
987-
6-
r
5<
43210
50
75
100
125
150
175
200
INTRASINUS PRESSURE ( mm Hg )
arterial pressure and ventilation rose. This increase in
arterial pressure might have caused an increase in
backflow to the carotid bodies, through residual anastomoses. Since a steady state increase in perfusion
pressure to the carotid body has been shown not to
affect or to decrease ventilation (Biscoe et al., 1970),
the baroreflex increase in arterial pressure could only
buffer and not augment the ventilatory response to
baroreceptor stimulation.
The same argument holds for the participation of
the central chemoreceptors during carotid sinus baroreceptor pressure changes. Increasing intrasinus
pressure from 50 to 200 mm Hg would decrease
arterial pressure. The accompanying decreases in cerebral perfusion pressure and pH, and increases in
Paco2 would only cause ventilation to increase. However, we always saw decreases in ventilation under
this condition. These measured decreases in ventilation may have, in fact, been buffered by any response
to cerebral ischemia. In order to avoid cerebral hypoxia, we maintained the Pao2 above 100 mm Hg
throughout the experiment. Cerebral ischemia was
unlikely since, in most of the dogs, arterial pressure
did not fall below 80 mm Hg, even at high intrasinus
pressures. An adequate cerebral blood supply through
the vertebral arteries must have existed throughout
this experiment, since baroreceptor reflex control of
arterial pressure was always present.
Nevertheless, decreases in cerebral perfusion pressure could possibly have influenced the respiratory
changes which we have attributed to carotid sinus
baroreceptors. Nonvagally mediated afferents could
also respond to drastic changes in arterial pressure. If
Circulation Research/Vo/. 51, No. 5, November 1982
632
1000 -,
1000
• 10
900900
800-
1.
..
800
700700-
600600-
£
1
500-
500-
UJ
UJ
s
E
§ 400-
400-
i
K
Downloaded from http://circres.ahajournals.org/ by guest on June 16, 2017
300H
300-
200-
200-
100-
100-
B
50
75
100
125
150
175
200
50
75
100
125
150
175
200
INTRASINUS PRESSURE ( mm Hg )
INTRASINUS PRESSURE ( mm Hg )
Figure 6. Tidal volume responses to changes in intrasinus pressure
in dogs 2 -5 (panel A), 6-10 (panel B), and mean values for 10 dogs
panel C. Arrows indicate direction of intrasinus pressure change in
panel C.
1000-
900-
800-
700-
600E
-
500-
O
>
400-
300-
200-
100-
50
75
100
125
150
175
200
INTRASINUS PRESSURE ( mm Hg )
changes in ventilation were due simply to decreases
in cerebral perfusion pressure, then one would not
expect to see respiratory changes when arterial pres-
sure was controlled. However, we always saw changes
in ventilation with baroreceptor stimulation when
arterial pressure was controlled. For the five dogs
studied, the changes in respiratory frequency when
mean arterial pressure was controlled (as in Fig. 9B)
were significantly greater (P < 0.05) than when mean
arterial pressure was allowed to vary.
To see whether changes in mean arterial pressure
per se could influence ventilation, intrasinus pressure
was held constant, and mean arterial pressure was
step changed in some series II animals. Small changes
in ventilation were seen when arterial pressure was
increased above 80 mm Hg. However, when mean
arterial pressure was decreased below this point, consistent increases in ventilation were seen, most probably due to a cerebral ischemic response. This was
opposite to the response seen when intrasinus pressure was increased to 200 mm Hg; arterial pressure
fell and ventilation decreased. The ventilatory responses to increasing intrasinus pressures which we
are reporting could only have been blunted by any
cerebral ischemic response. Therefore, our results
may actually underestimate the contribution of the
reflex system in this range of arterial pressure below
80 mm Hg. The amount of increase in ventilation seen
when arterial pressure was lowered was dependent
upon the level at which intrasinus pressure was held
Brunner ef a/./ Baroreceptor Control of Respiration
633
6000-
6000-
5000-
5000-
E
I
4000-
4000-
3000-
3000-
2000-
2000-
z
UJ
Downloaded from http://circres.ahajournals.org/ by guest on June 16, 2017
50
75
100
125
150
175
200
INTRASINUS PRESSURE ( mm Hg )
75
100
125
150
175
200
INTRASINUS PRESSURE ( mm Hg )
Figure 7. Minute ventilation responses to changes in intrasinus
pressure in dogs 1-5 (panel A), 6-10 (panel B), and mean values
for 10 dogs (panel C). Arrows indicate direction of intrasinus
pressure change in panel C.
6000-
5000-
.£
4000-
Z
2
50
3000-
s
2000-
50
75
100
125
150
175
200
INTRASINUS PRESSURE ( mm Hg )
constant. For the same decrease in mean arterial pressure (from 150 to 60 mm Hg), ventilation increased
by 100% when intrasinus pressure was 50 mm Hg,
but increased only 30% when intrasinus pressure was
200 mm Hg. These intriguing results may imply an
interaction between baroreceptor inputs and the cerebral ischemic response and warrant further, more
rigorous study.
Of the 10 animals whose arterial pressure responses
are shown in Figure 4, only three (dogs #2, 3, and 8)
ever fell below the level of 80 mm Hg. Thus, in the
remaining seven animals, the observed responses to
changes in mean arterial pressure could not have
contributed to the changes in ventilation seen. These
changes can therefore be considered to be due to
changes in baroreceptor input alone, and not secondary to changes in mean arterial pressure. This gives
further proof that the arterial baroreceptor reflex does
control ventilation.
The respiratory responses to carotid sinus baroreceptor pressure changes were similar to the mean
arterial pressure responses. After each step change in
intrasinus pressure, the onset of the respiratory responses was identical to that of the pressor response.
The responses occurred within the first breath, with
a time constant of 8-12 seconds. The steady state
changes in all respiratory variables were also similar
to the pressor response. All respiratory response
curves followed the characteristic sigmoidal shape of
the pressor response.
634
Circulation Research/Vo/. 51, No. 5, November 1982
16-
' Respiratory Period
' Expiratory Duration
• Inspiratory Duration
15141312111098765432Downloaded from http://circres.ahajournals.org/ by guest on June 16, 2017
150
75
100
125
150
175
200
INTRASINUS PRESSURE ( mm Hg )
Figure 8. Respiratory period, expiratory and inspiratory duration
responses to changes in intrasinus pressures. Mean values for dogs
6-10 are shown. Arrows indicate direction of intrasinus pressure
changes.
Hysteresis was always clearly seen in the mean
arterial pressure response to changes in intrasinus
pressure. These findings confirm results of a previous
study (Shoukas and Brunner, 1980) in which intrasi-
nus pressure was changed in steps of 25 mm Hg. The
mean arterial pressure response to decreasing intrasinus pressure always followed a curve which was
greater than that of increasing intrasinus pressure.
Recently, Coleridge et al. (1981) have shown that this
hysteresis, for the aortic baroreceptors, occurs at the
receptor level—that is, that baroreceptors in the aortic
nerve fired more when the aortic pressure was increasing than when it was decreasing. This behavior
could explain the hysteresis seen in our mean arterial
pressure and ventilatory response curves. If hysteresis
does occur at the level of the receptor, one would
expect to see such behavior in both the pressor and
the respiratory responses to changes in intrasinus
pressure. If hysteresis was found in the respiratory
response as well as the pressor response, then this
would clearly indicate that the changes in ventilation
seen are indeed due to the carotid sinus baroreceptor
reflex. We could not demonstrate statistically that
hysteresis in the frequency, tidal volume, and ventilation curves was present at three levels of intrasinus
pressure as we found in the pressor response. However, we could show that at an intrasinus pressure of
100 mm Hg, frequency and ventilation were greater
when intrasinus pressure was decreasing than when
intrasinus pressure was increasing. Hysteresis was
never seen in the tidal volume responses. In any single
animal, our ability to see hysteresis clearly in the
ventilatory responses depended upon the constancy
of anesthesia maintained during the experimental run,
as discussed below. The most probable explanation
for the difference between observed behavior of hysteresis in the pressor and ventilatory responses con-
250
250-
Arterial
pressure
Arterial
pressure Mjjnrmrrr.(mmHg)
(mmHg)
:
0250-
250
Intrasinus
pressure
Intrasinus
pressure
(mmHg)
(mmHg)
Ventilatory
flow
Ventilatory
flow
d a l
500
volume
volume •-
(ml)
(ml)
-fmiri—I
l-«-1 mirt—•»!
Figure 9. Effect of arterial pressure on ventilatory changes due to increasing intrasinus pressure from 125 to 200 mm Hg. Panel A: Response
when arterial pressure is allowed to decrease reflexly. Panel B: Response when reflex decrease in arterial pressure is prevented; arterial
pressure controlled at 100 mm Hg.
Brunner et ai./Baroreceptor Control of Respiration
635
TABLE 3
Intrasinus
pressure
(mm Hg)
Mean arterial
pressure
(mm Hg)
Respiratory
frequency
(min"')
Tidal volume
(ml)
Ventilation
(ml/min)
Uncontrolled arterial pressure (n = 5)
50
125
200
157.8 ± 7.3
117.6 ± 6.8
77.4 ± 8.7
15.0 ± 3.3
12.2 ± 2.1
9.1 ± 2.7
528 ± 192
602 ± 209
705 ±218
5945 ± 824
5979 ± 1186
4791 ± 849
•P < 0.005
*P < 0.05
•P < 0.025
*P < 0.025
Controlled arterial pressure (n =5)
50
125
200
100.8 ± 2.1
100.6 ± 2.1
99.4 ± 2.5
17.6 ± 2.9
12.4 ± 2.2
8.0 ± 1.7
417 ± 127
483 ±189
531 ±179
6260 ± 904
5869 ± 1030
4212 ± 850
N.S.
*P < 0.005
*P < 0.025
*P < 0.025
Downloaded from http://circres.ahajournals.org/ by guest on June 16, 2017
cerns the difference in sensitivity of the two responses
to different levels of anesthesia.
The major differences between respiratory and circulatory responses to changes in intrasinus pressure
seemed to be related to the level of anesthesia in a
particular animal. Pressor responses seemed to be
relatively insensitive to an individual dog's anesthetic
level. Threshold and saturation levels of pressure for
the pressor reflex as well as the reflex sensitivity
(gain) remained fairly constant over differing levels of
anesthesia. However, for the respiratory response
curves, there seemed to be a consistent dependence
of thresholds, saturations, and sensitivities upon the
level of anesthesia. In two of the dogs, repeated runs
at different levels of anesthesia produced different
respiratory response curves without affecting the
pressor response curve. Although we have no objective measure of the depth of anesthesia attained, it
should be fairly apparent that the anesthetic level was
well correlated with the different sensitivities of the
respiratory responses. Regardless of anesthetic level,
changes in intrasinus pressure always caused directionally similar, if not equivalent, changes in respiratory variables. The five dogs used in series II experiments were most probably on a lighter anesthetic
plane, and exhibited higher respiratory frequencies
and ventilations than those of the series I experiments.
We conclude that pentobarbital anesthesia affects the
respiratory response limb of the carotid sinus baroreceptor reflex much more than it affects the vasopressor response limb. This may account for the
statistical differences in hysteresis behavior of the
pressor and respiratory responses.
The increase in frequency and ventilation seen
when intrasinus pressure is decreased is in agreement
with earlier studies. Schmidt (1932) and Bishop (1974)
both reported increases in frequency with decreases
in vascularly isolated carotid sinus pressure. Levy
(1966) reported a decrease in respiratory frequency
when isolated aortic arch baroreceptor pressure was
increased. These experiments were performed in vagally intact animals.
However, it has also been reported that, in response
to aortic balloon inflation (Grunstein, 1975) and hemorrhage (Miserocchi, 1980), respiratory frequency did
not change in vagotomized cats. In our vagotomized
preparation, we always found increases in steady state
respiratory frequency when intrasinus pressure was
decreased from 200 to 50 mm Hg. The differences in
results may be due to the difference in species used,
or in specificity of the baroreceptor stimulus. Both the
removal of vagal afferents and the elimination of
carotid chemoreceptors in our preparation may be an
important factor in accounting for differences in results.
We have shown that carotid sinus baroreceptors
can influence both the inspiratory and expiratory
duration. Bishop (1974) showed that decreasing intrasinus pressure increased both diaphragm and abdominal muscle activity. Gabriel and Seller (1969) showed
an increased expiratory neuron activity when carotid
sinus pressure was increased, thereby demonstrating
that baroreceptors have medullary connections with
expiratory neurons. Our data support the postulate
that carotid sinus baroreceptors project to both inspiratory and expiratory neurons.
The tidal volume responses reported here are in
direct disagreement with data reported by Heistad et
al. (1975), Grunstein et al. (1975), and Miserocchi and
Quinn (1980). Grunstein reported a decrease in tidal
volume (for the same Paco2) with aortic balloon inflation in neurally intact cats. Miserocchi reported an
increase in tidal volume following hemorrhage in
neurally intact cats. In both of these studies, sodium
pentobarbital was the anesthetic used. Again, the
differences in results could very well be due to the
intact vagus nerve and differences in stimuli used.
Heistad et al. showed that tidal volume and frequency decreased with increasing perfusing pressure
on the isolated left carotid sinus. They showed a
linear inverse relationship between carotid perfusion
pressure and minute ventilation. Their anesthetic was
chloralose-urethane. With vagi and peripheral chemoreceptors intact, an increase in carotid perfusion
pressure from 100 to 200 mm Hg caused ventilation
to decrease from 10.5 to 10.0 1/min. In our vagoto-
Circulation Research/Vo/. 51, No. 5, November 1982
636
Downloaded from http://circres.ahajournals.org/ by guest on June 16, 2017
mized, peripherally chemodenervated dogs, the total
ventilation at any intrasinus pressure was lower, but
changed to a larger degree for the same intrasinus
pressure change, from 4.4 to 3.2 1/min. We also saw
a sigmoidal relationship between ventilation and intrasinus pressure. They reported results from only
two levels of intrasinus pressure; the exact shape of
the ventilation-intrasinus pressure relationship was
not a main focus of their study. However, they did
show an important interaction between carotid baroreceptor and chemoreceptor stimulation. When both
receptors are simultaneously activated, the combined
response is greater than the sum of separate individual
stimulations. This emphasized the importance in this
experiment of separating individual effects in order
to characterize the sole contribution of carotid sinus
baroreceptors in the control of ventilation. It may be
that during a complex stimulus like hemorrhage,
where many reflexes act in concert, carotid baroreceptor reflex control of ventilation is, in fact, augmented, and plays a significant role in the maintenance of homeostasis.
We wish to express our thanks to Dr. Kiichi Sagawa for his
advice and concern during the course of these experiments, to
Linda Watermeier, Joan Midthun, and Chobad Djawdan for technical assistance, and to Kathy Hartzell for typing the manuscript.
Supported by U.S. Public Health Service Grant HL-00219 and
HL-19039
Dr. Brunner is a Department of Biomedical Engineering Andrew
W. Mellon Foundation Postdoctoral Fellow.
Mr. Sussman was supported by Johns Hopkins Biomedical
Research Support Grant 2S07RR05378-20.
Address for reprints: Dr. Artin A. Shoukas, Department of
Biomedical Engineering, The Johns Hopkins University School of
Medicine, 720 Rutland Avenue, Baltimore, Maryland 21205.
Received November 13, 1981; accepted for publication August
5, 1982.
References
Aviado D, Schmidt C (1959) Cardiovascular and respiratory reflexes from the left side of the heart. Am J Physiol 164: 726-730
Biscoe T, Bradley G, Purves M (1970) The relation between carotid
body chemoreceptor discharge, carotid sinus pressure and carotid body venous flow. J Physiol (Lond) 208: 99-120
Bishop B (1974) Carotid baroreceptor modulation of diaphragm
and abdominal muscle activity in the cat. J Appl Physiol 36: 1219
Chungchareon D, Daly M Schweitzer A (1952) The blood supply
of the carotid body in cats, dogs and rabbits. J Physiol (Lond)
117: 347-358
Coleridge H, Coleridge J, Kaufman M, Dangel A (1981) Operational
sensitivity and acute resetting of aortic baroreceptors in dogs.
Circ Res 48: 676-684
Davis A, McCloskey D, Potter E (1977) Respiratory modulation of
baroreceptor and chemoreceptor reflexes affecting heart rate
through the sympathetic nervous system. J Physiol (Lond) 272:
691-703
Delpierre S, Jammes Y, Vanuxem D, Vanuxem P, Grimaud C
(1974) Modifications ventilatoires par stimulation des afferences
barosensibles du sinus carotidien isole de la circulation cerebrale.
C R Acad Sci [B] (Paris) 169: 989-993
Gabriel M, Seller H (1969) Excitation of expiratory neurones adjacent to the nucleus ambiguus by carotid sinus baroreceptor and
trigeminal afferents. Pfluegers Arch 313: 1-10
Grunstein M, Derenne J, Milic-Emili J (1975) Control of depth and
frequency of breathing during baroreceptor stimulation in cats.
J Appl Physiol 39: 395-404
Haymet B, McCloskey D (1975) Baroreceptor and chemoreceptor
influences on heart rate during the respiratory cycle in the dog.
J Physiol (Lond) 245: 699-712
Heistad D, Abboud F, Mark A, Schmid P (1974) Interaction of
baroreceptor and chemoreceptor reflexes: modulation of the
chemoreceptor reflex by changes in baroreceptor activity. J Clin
Invest 53: 1226-1236
Heistad D, Abboud F, Mark A, Schmid P (1975) Effects of baroreceptor activity on ventilatory response to chemoreceptor stimulation. J Appl Physiol 39: 411-416
Kostreva D, Hopp F, Zuperku E, Igler F, Coon R, Kampine J (1978)
Respiratory inhibition with sympathetic afferent stimulation in
the canine and primate. J Appl Physiol 44: 718-724
Levy M, Ng M, Zieske H (1966) Cardiac and respiratory effects of
aortic arch baroreceptor stimulation. Circ Res 19: 930-939
Lim T, Luft U, Grodins F (1958) Effects of cervical vagotomy on
pulmonary ventilation and mechanics. J Appl Physiol 13: 317324
Miserocchi G, Quinn B (1980) Control of breathing during acute
hemorrhage in anesthetized cats. Respir Physiol 41: 289-305
Richter D, Seller H (1975) Baroreceptor effects on medullary respiratory neurones of the cat. Brain Res 86: 168-171
Schmidt C (1932) Carotid sinus reflexes to the respiratory center.
I. Identification. Am J Physiol 102: 94-119
Shoukas A, Brunner M (1980) Epinephrine and the carotid sinus
baroreceptor reflex: Influence on capacitive and resistive properties of the total systemic vascular bed of the dog. Circ Res 47:
249-257
Shoukas A, Sagawa K (1973) Carotid sinus baroreceptor reflex
control of total systemic vascular capacity. Circ Res 33: 22-33
Von Euler C, Herrero F, Wexler I (1970) Control mechanisms
determining rate and depth of respiratory movements. Resp
Physiol 10: 93-108
Winder C (1938) Isolation of the carotid sinus pressoreceptive
respiratory reflex. Am J Physiol 122: 306-318
INDEX TERMS: Respiratory frequency • Tidal volume • Ventilation • Respiratory control
Carotid sinus baroreceptor reflex control of respiration.
M J Brunner, M S Sussman, A S Greene, C H Kallman and A A Shoukas
Downloaded from http://circres.ahajournals.org/ by guest on June 16, 2017
Circ Res. 1982;51:624-636
doi: 10.1161/01.RES.51.5.624
Circulation Research is published by the American Heart Association, 7272 Greenville Avenue, Dallas, TX 75231
Copyright © 1982 American Heart Association, Inc. All rights reserved.
Print ISSN: 0009-7330. Online ISSN: 1524-4571
The online version of this article, along with updated information and services, is located on the
World Wide Web at:
http://circres.ahajournals.org/content/51/5/624
Permissions: Requests for permissions to reproduce figures, tables, or portions of articles originally published in
Circulation Research can be obtained via RightsLink, a service of the Copyright Clearance Center, not the
Editorial Office. Once the online version of the published article for which permission is being requested is
located, click Request Permissions in the middle column of the Web page under Services. Further information
about this process is available in the Permissions and Rights Question and Answer document.
Reprints: Information about reprints can be found online at:
http://www.lww.com/reprints
Subscriptions: Information about subscribing to Circulation Research is online at:
http://circres.ahajournals.org//subscriptions/