Circulation Research
JUNE
VOL. XXIV
1969
NO. 6
An Official Journal of the American Heart Association
Central Nervous Integration of the
Circulatory and Respiratory Responses
to Arterial Hypoxemia in the Rabbit
By Paul I. Korner, M.D., John B. Uther, M.B.,
and Saxon W. White, M.D.
Downloaded from http://circres.ahajournals.org/ by guest on June 18, 2017
ABSTRACT
Neural integration during arterial hypoxia was studied in sham-operated,
rhinencephalic, thalamic, high mesencephalic, and pontine rabbits, 3 hours
after operation under halothane anesthesia. All preparations except the pontine
recovered normal movement and posture 40 to 60 minutes after the operation,
and effects on the resting circulation specifically ascribable to transection were
small. Activation of diencephalic, and to a lesser extent of rhinencephalic,
centers was necessary to produce the large increase in autonomic peripheral
resistance effect and the autonomic slowing of heart rate characteristic of
normal rabbits. In animals with only pontine and high mesencephalic centers,
the autonomic peripheral resistance effect was smaller and there was an
autonomic rise in heart rate. The neocortex and rhinencephalon exerted
inhibitory influences related to the effects of hyperventilation. Suprabulbar
respiratory mechanisms were also activated during hypoxia, with diencephalic
mechanisms limiting the reflex response mediated by the pontine centers and
the cortex exerting disinhibitory effects on the diencephalic centers. The
cardiorespiratory response at different degrees of hypoxia probably depends on
differences in relative magnitude of inputs from the arterial chemoreceptors,
baroreceptors, and lung inflation receptors, producing different degrees of
excitation and inhibition of the various suprabulbar and bulbar centers.
ADDITIONAL KEY WORDS
rhinencephalic thalamic pontine preparations
neural ablation
suprabulbar control of respiration and circulation
• A number of regions in the cortex and
diencephalon when stimulated electrically can
From the Hallstrom Institute of Cardiology, Royal
Prince Alfred Hospital, and the Department of
Medicine, University of Sydney, Sydney, Australia.
This investigation was supported by grants from the
National Heart Foundation of Australia, the Life
Insurance Medical Research Fund of Australia and
New Zealand, and the National Health and Medical
Research Council. Dr. Uther is a Research Fellow of
the Life Insurance Medical Research Fund of
Australia and New Zealand. Dr. White is a Research
Fellow of the National Heart Foundation of
Australia.
Received October 16, 1968. Accepted for publication March 21, 1969.
Circulation Research, Vol. XXIV, June 1969
influence the activity of the autonomic effectors to the circulation (1-5), and some of these
effects can be modified in rum by stimulation
of the arterial baroreceptors and chemoreceptors (6-11). It remains uncertain whether the
suprabulbar regions contribute significantly to
cardiorespiratory integration of physiologically evoked reflex activity arising from these
peripheral receptors (12, 13). We have
examined this question in the present experiments in the rabbit by comparing the
respiratory and circulatory responses to arterial hypoxia and the autonomic effects in a
number of acute unanesthetized prepara757
758
Downloaded from http://circres.ahajournals.org/ by guest on June 18, 2017
tions, including sham-operated, rhinencephalic,
thalamic, high mesencephalic and pontine
animals (14-16). The circulatory effects of
different levels of hypoxia in the normal
unanesthetized rabbit, the role of the autonomic effectors, and the contribution to the
autonomic effects of peripheral inputs from
the arterial chemoreceptors, baroreceptors and
lung inflation receptors, have been the subject
of considerable analysis in this species (1723). The results of the present studies suggest
that suprabulbar regions of the nervous
system are normally important in the rabbit's
cardiovascular and respiratory responses to
arterial hypoxia.
Methods
Ninety-two offspring of New Zealand White
rabbits crossbred with the New Zealand Giant
strain, of average body weight 2.6 kg (range 2.3
to 3.4 kg) were used in this study.
Operations.—In all rabbits except the "deefferented" ones, a thermistor catheter was
inserted into the aorta at a preliminary operation
3 to 7 days before an experiment, as previously
described (24). To prepare autonomically "deefferented" animals, bilateral adrenalectomy was
performed 10 to 14 days before an experiment. A
thermistor was inserted into the aorta at the same
time, and the animals were maintained on fixed
adrenal steroid replacement therapy (25); from
the fourth postoperative day they were given
guanethidine sulfate (12.5 mg/kg/day iv), the
last dose being given on the morning of the
experiment; just before the test the animals were
fully atropinized (see below) (20,26).
On the day of the experiment, the central ear
artery and right atrium were cannulated and a
tracheotomy tube was inserted under local
lidocaine anesthesia (24).
The neurosurgical procedures were performed
under open-circuit halothane anesthesia. The
dura was opened through a bilateral frontoparietal craniotomy, and appropriate regions of the
brain were removed by a sucker fitted with a
blunt needle tip (16-21 gauge Luer). The tip
formed the active lead of a diathermy apparatus,
and the force of suction was controlled by partial
digital occlusion of a side-hole in the handle.
With this arrangement, brain stem distortion was
minimal, blood vessels were coagulated as they
were exposed during removal of brain tissue, and
the blood loss was small (3 to 8 ml measured as
hemoglobin from the sucker bottle and eluted
from swabs). The operations required about 1
hour. We prepared: (1) Sham-operated animals,
KORNER, UTHER, WHITE
in which bilateral craniotomy was performed and
8 ml of blood removed from the arterial catheter
during anesthesia, which was extended to 1 hour.
In most experiments the dura remained intact,
but in some it was accidentally torn, with
consequent escape of cerebrospinal fluid, but this
was without apparent effect on the postoperative
resting values of the different cardiorespiratory
variables or on the response to hypoxia. (2)
Rhinencephalic animals, in which the superior
colliculus was exposed by removing the occipital
lobe and the parietal cortex removed, leaving a
layer of white matter overlying the anterior
portion of the lateral ventricle. The cortical
removal extended ventrolaterally to the rhinic
fissure and medially to the lateral sagittal fissure,
leaving the area paracentralis (27) intact over
the middle of the hemisphere; in the anterior
quarter the area paracentralis was removed to the
midline (Fig. IB). (3) Thalamic animals in
which the cortex, olfactory tract, hippocampus,
fornix, basal ganglia, and septum pellucidum
were removed; thalamus, subthalamus, and hypothalamus including preoptic regions were preserved intact (Fig. 1C). (4) High mesencephalic
animals in which the transection extended from
the anterior border of the superior colliculus to
just behind the mammillary body (Fig. I D ) . (5)
Pontine animals, in which transection was from
the lower margin of the inferior colliculus to the
anterior pontine border near the point of
emergence of the oculomotor nerve (Fig. IE). In
three pontine animals, the carotid sinus and aortic
nerves were sectioned after decerebration (24).
After all bleeding had stopped, the exposed
brain was covered with moist absorbable gelatin
sponge (Sterispon, Allen & Hanbury, Ltd.), and
the skin was sutured. The animals recovered in a
sound-proof box, and were given 5.5% dextrose in
distilled water, 15 ml iv hourly. Body temperature
was maintained when necessary by an electric
blanket. At the end of the experiment, the brains
were fixed in 10% formol saline and examined
macroscopically. The lesions were reproducible,
and histological examination showed that there
was no underlying damage of parts of the brain
covered with pia mater (e.g., superior colliculus
in Fig. 1C); small, irregular patches of edema
and hemorrhage did not extend more than 200 to
500 ft from the cut surfaces.
Cardiorespiratory Measurements.—Cardiac output was measured by recording thermodilution
curves from the upper abdominal aorta, after
rapid injection into the right atrium of 0.46 ml of
5.5% dextrose in distilled water at room temperature (24). Mean ear artery and right atrial
pressures were recorded using Statham P 23 AC
strain gauges, and the heart rate was obtained
from the arterial pressure record. Total peripheral
Circulation Research, Vol. XXIV, June 1969
759
NEURAL INTEGRATION IN HYPOXIA
Downloaded from http://circres.ahajournals.org/ by guest on June 18, 2017
FIGURE 1
Photographs of normal rabbit brain (A), rhinencephalic preparation (B), thalamic preparation
(C), high mesencephalic preparation (D), pontine preparation (E).
resistance was calculated as (ear artery pressure —
right atrial pressure)/cardiac output. The animal
breathed room air through a low-resistance
respiratory flap-valve (17, 24) during control and
recovery periods, and a freshly prepared low O2N2 mixture from a light latex balloon during the
treatment period. Respiratory rate was measured
by counting the respiratory movements and the
minute volume (liters/min, ATPS) was determined by collecting expired gas. Arterial Po2,
Pco 2 and pH were measured with a blood gas
analyzer and pH meter. (Instrumentation Laboratory Inc., Model 113).
Experimental Procedure.—After the initial cannulation procedures, each animal rested for 1
hour in a small rabbit box, kept in a larger soundproof box. Four measurements of the different
resting cardiorespiratory variables (cardiac output, arterial and right atrial pressures, heart rate,
respiratory rate and minute volume) were made
over a period of 10 minutes in each animal.
Neurosurgery or sham operation (craniotomy)
were then performed, and the main part of the
experiment began 3 hours later.
In the first series of experiments, all animals
were studied inside the sound-proof box 3 hours
after surgery while sitting in their normal posture,
except for the pontine animals, which required a
Circulation Research, Vol. XXIV, June 1969
special holder. The experiment consisted of a
resting period of 16 minutes breathing room air
(5 sets of cardiorespiratory measurements), a
treatment period of 44 minutes breathing 7.5%
to 12% O2 (17 sets of measurements), and a
recovery period of 13 minutes breathing room air
(4 sets of measurements). In each experiment,
measurements of arterial Po2, Pco2, and pH were
obtained while the animal was breathing room
air, and at 12 and 32 minutes after the start of
hypoxia. The changes in Pao2 minute by minute
were not studied, since previous studies showed
that the arterial O2 saturation falls close to its
new equilibrium level in less than 3 minutes (28,
29). The effects of only one level of hypoxia were
examined in each animal. In each preparation
three groups of rabbits were studied at Pao2
approximately 50, 40, or 30 mm Hg, using mostly
3 to 5 animals per group; two other groups of
normal animals were also studied at Pao2 35 mm
Hg and 26 mm Hg.
In the second series of experiments, the animals
again recovered from the effects of operation
while breathing spontaneously. Three hours after
operation they were connected to a respiratory
pump and ventilated at the average minute
volume (1 liter/min) and rate (60/min) of
normal rabbits. Muscular relaxation was induced
760
Downloaded from http://circres.ahajournals.org/ by guest on June 18, 2017
by decamethonium iodide, 1 mg/kg iv, with
supplements of 0.5 mg/kg iv every 20 to 30
minutes (26). No operative intervention was
undertaken while the animals were under the
influence of decamethonium, and they were
placed supine on a warm platform, wrapped in
cotton wool, and Dow Silicone Medical Fluid No.
360 instilled into their eyes to minimize somatic
stimuli. Each animal was subjected to three levels
of hypoxia in random order, each test consisting
of a resting period of 8 minutes (room air, four
measurements); a period of hypoxia of 10
minutes (five measurements); a recovery period
of 6 minutes (room air, three measurements).
During the treatment and recovery periods the
ventilation was raised to 2 liters/min for all levels
of arterial Po 2 to imitate the respiratory response
to severe hypoxia of a spontaneously breathing
normal animal (17, 19); this degree of hyperventilation during hypoxia resulted in approximately
the same changes in Paco 2 and pH as in the
spontaneously breathing animal subjected to Pao«
of about 30 mm Hg (Table 2), these changes in
blood composition being known to be tolerated by
the latter without apparent distress. Arterial blood
composition was determined before, and 8
minutes after, the start of hypoxia. A period of 30
minutes was allowed between tests.
A third series of experiments was performed in
the same way under controlled ventilation, but
the animals were given atropine sulfate, 3 mg iv
10 minutes before the test followed by a
continuous infusion of 0.2 mg/min iv.
Statistical Methods.—The effects of the neural
ablations and of sham operation on the resting
cardiorespiratory measurements were assessed in
each animal by comparing the average of four
preoperative normal resting measurements, with
the average of five resting measurements obtained
3 hours postoperatively. The data from all
preparations were examined by analysis of
variance, and the error mean square (EMS) of
each variable obtained after subtraction from the
total sums of squares of the effects due to times
(preoperative vs. postoperative); lesions; groups
within lesions (spontaneous vs. controlled respiration); between animals; times X lesions; and
times X groups within lesions. The standard error
(SE) of the mean of each group of n animals was
calculated as (EMS/II) ! , and the statistical
significance of the difference between pre- and
postoperative values and between the relative
effects of neural ablation and sham operation
were estimated using variance ratio, and t-test
(30).
In the first series of experiments, the 26
measurements of each variable were grouped into
the 10 larger time intervals, as shown in Figure 5.
In each animal, two or three measurements of
KORNER, UTHER, WHITE
each variable were averaged per larger time
interval, except for the first interval after the start
of hypoxia, for which only one set of measurements was obtained. The timing of measurements
was the same in each animal, and the mean
absolute value of each variable was obtained by
averaging the results of all animals in a particular
group at each time interval. To facilitate
comparisons between the different preparations,
the average changes at each time interval during
hypoxia and recovery have been expressed as a
percent of the mean resting value of the
particular group. The standard error of the mean
at a single time interval was calculated from
analysis of variance, after subtracting betweentime-intervals from the total sums of squares, as
(EMS/II)*, where n is the number of animals in
the group. In the second or third series of experiments, the standard error of the mean at a single
time interval was calculated similarly, but in these
series all time intervals were of only 2 minutes
duration, with one measurement per time interval
from each animal, and mostly three animals per
group.
Assessment of Autonomic Effects.—Changes in
autonomic effects were assessed indirectly from
the differences in the circulatory responses
between the various neurological preparations on
the one hand, and autonomically de-efferented
animals on the other, as discussed in detail
elsewhere (23, 26, 31). The cardiorespiratory
variables of the latter preparation at rest differ
only slightly from normal, and they have a normal
respiratory response to hypoxia. Since hypoxia
appears to have a direct action on the arteriolar
smooth muscle (32), the magnitude of which is
probably not altered by the de-efferentation
procedure (20) or by the difference in transmural
pressure between the preparations in the present
experiments (33), we have assumed that the
direct local effects of hypoxia on the peripheral
arterioles in normal animals are relatively independent of the degree of neural constrictor tone
and can be estimated approximately from the
responses of autonomically de-efferented animals
(23). The method estimates autonomic effects
and not autonomic neural activity, and is suitable
for studying only relatively gradual changes in
these effects.
Figure 2 illustrates the calculation in shamoperated normal animals of the autonomic effects
at Pao2 of 50, 40, and 30 mm Hg. To assess net
autonomic effects on total peripheral resistance,
the resting values in sham-operated and deefferented animals were superimposed, and the
changes in each preparation expressed as the
percent of its own resting value at the various
longer time intervals. The differences in total
peripheral resistance values at each time interval
Circulation Research, Vol. XXIV, June 1969
NEURAL INTEGRATION IN HYPOXIA
%
I 50 I
I 40 |
761
f 30 I
-40
Downloaded from http://circres.ahajournals.org/ by guest on June 18, 2017
AUTONOMIC
T.RR. EFFECT
20
60
20
60
60
TIME — minutes
FIGURE 2
Graph showing calculation of autonomic effects on
heart rate (H.R.) and total peripheral resistance
(T.P.R.). Low O z -?/ 2 mixtures breathed between
arrows, room air at other times. Results at Pao2
50 mm Hg from 3 normal (solid line) and 3 deefferented (broken line) rabbits; at Pao2 40 mm Hg
from 5 normal and 3 de-efferented rabbits; at Pao2
30 mm Hg from 4 normal and 3 de-efferented rabbits.
In the top row, mean heart rate changes have been
expressed as percent of resting, in each group, the
symbol on the left being the standard error of the
mean at a single time interval. In the second row, the
difference in percent response between normal and
de-efferented animals is the autonomic heart rate effect, and the time-effect curves obtained by joining
successive time intervals; the standard error of the
autonomic effect at a single time interval is on the
left. In the third row, the total peripheral resistance
percents are plotted; the net autonomic effect on
total peripheral resistance in the fourth row has been
obtained at each time interval from the difference in
percent response between the normal and de-efferented animals.
and the differences in heart rate response are used
to construct time-autonomic total peripheral
resistance and heart rate effect curves. As seen in
Figure 2, at Pao2 50 mm Hg, the heart rate
Circulation Research, Vol. XXIV, June 1969
increases during hypoxia in both sham-operated
and de-efferented groups, the rise being approximately twice as great in the former. The
autonomic component of the tachycardia is thus
less than the total rise in heart rate in the shamoperated animal. When, as at Pao2 50 mm Hg,
the rise in heart rate is relatively greater in shamoperated than in de-efferented animals, the
autonomic response is called an autonomic rise in
heart rate, and when the sham-operated animal
responds with more cardiac slowing than the deefferented animal (as at Pao3 30 mm Hg), it is
termed autonomic cardiac slowing. The standard error of the mean autonomic effect
at a single time interval, SEaut = (sE2normal +
SE2
de-ef()J. t r i e values in parentheses being the
standard error of a single time interval in the
total peripheral resistance or heart rate change of
sham-operated and de-efferented animals for (n t
+ n2 — 2) degrees of freedom, where r^ and n2
are the number of normal and de-efferented
animals, respectively. When calculating the significance of differences between two groups of
neurological preparations with n t and n2 animals
respectively, the degree of freedom is (nx + n2 —
2), and the de-efferented results contribute only to
the calculation of SE,lut of each group. The standard error of the mean autonomic effect over the
whole period of hypoxia in any group is SEaut/
(N)*, where N is the number of larger time
intervals during hypoxia.
The above method was used in the other
neurological preparations to construct time-autonomic effect curves at Pao2 approximately 50, 40,
and 30 mm Hg. Small corrections were applied to
the data from de-efferented rabbits using more
detailed response curves to allow for differences
in arterial Pao2 and pH between corresponding
neurological groups. The de-efferented animals
were not subjected to halothane anesthesia and
craniotomy because of a high mortality during
induction of anesthesia in these areflexic animals
(26). Since the same basic data from deefferented rabbits have been used with each
neurological preparation, any changes in local
circulatory responsiveness due to preceding
anesthesia and operation will systematically affect
the estimates of the autonomic effects in all the
preparations. However, it is unlikely that local
vessel sensitivity was much affected, since in one
de-efferented animal the circulatory response to
hypoxia was the same before as 3 hours after
anesthesia and craniotomy.
Assumptions and Limitations.—The use of
acute neural preparations in the analysis of
central nervous function requires that the ablation
procedures should produce few nonspecific effects
and that differences in resting state will not
preclude valid comparisons between the prepara-
762
KORNER, UTHER,
WHITE
TABLE 1
Comparisons of Mean Normal Resting Values with Values Three Hours after Surgery
Group*
Sham operation
(13)
Pre-op
Post-op
VE
1.03
1.37
f
67
112
±0.11
±12.2
VT
15.4
24.7
12.2
23.5
±1.1
Vo 2
Rhinencephalic
(8)
VE
1.32
1.97
f
81
101
VT
16.4
23.9
19.6
25.0
1.11
76
1.91
90
14.6
22.4
21.2
25.8
VE
1.11
f
81
1.21
77
VT
13.7
15.7
Vo,
VE
Downloaded from http://circres.ahajournals.org/ by guest on June 18, 2017
Thalamic
(12)
f
VT
Vo 2
High mesencephalic
(9)
SB
Variable
±0.15
±15.1
±1.4
±0.12
±12.3
P
0.01
<0.01
NS
<0.01
NS
NS
< 0.001
NS
±1.1
±0.14
±14.3
0.01
±0.12
±12.9
NS
±1.1
NS
NS
NS
Vo 2
Pontine
(11)
VE
1.15
f
72
1.31
75
VT
15.8
24.9
24.0
Vo2
NS
17.4
VE = respiratory minute volume (liters/min, ATPS); f = respiration rate (breaths/min);
Vo2 = oxygen consumption (ml/min, STPD); VT = tidal volume (ml), SE and P are standard
error and significance of difference between pre- and postoperative values obtained from the
same animal; NS = not significant.
*Number in parentheses is number of rabbits.
tions. Marked changes in the circulatory response
of a particular preparation (e.g., pontine) from
normal will then indicate that the centers
remaining after ablation by themselves account
inadequately for cardiorespiratory integration.
However, conclusions regarding the normal
function of these centers in animals with intact
brains must be more tentative, since the
hemodynamic changes may lead to a different
profile of stimulation of the peripheral receptors
normally involved in the response. Supplementary
information about the activation of various
centers from electrophysiological studies using the
evoked potential method would be valuable, but
has not been obtained in this study.
In the spontaneously breathing pontine animals, the mechanical effects of decerebrate
rigidity were minimized by using a special holder
which kept them in the same posture as normal
animals, and allowed free chest movements. In
these animals, resting respiratory minute volume
was not significantly different from that observed
after sham operation, but the depth of breathing
was somewhat greater. This could possibly be
related to change in discharge rates from the
stretch receptors of skeletal muscles, including
those of respiration, associated with the decerebrate rigidity. We have, however, assumed that
these differences do little to modify the pontine
animal's respiratory response to hypoxia, since it
is closely similar to that of normal (Fig. 4) and
high mesencephalic rabbits (Table 3), in which
decerebrate rigidity is absent. No data are
available on the work and efficiency of breathing
in the pontine rabbits but the overall effects of
decerebrate rigidity on muscle tone are not large
enough to affect total body oxygen consumption
(Table 1).
The muscle relaxant decamethonium when
used in large doses is known to affect chemoreceptor discharge during hypoxia and also autonomic ganglionic transmission (34, 35). In the
present dose, the effect on chemoreceptor discharge is probably minimal (35), and even with
Circulation Research, Vol. XXIV, June 1969
NEURAL INTEGRATION IN HYPOXIA
Downloaded from http://circres.ahajournals.org/ by guest on June 18, 2017
three times the present dose, the magnitude of the
heart rate changes produced by electrical stimulation of the vagus is virtually unaffected. Furthermore, in sham-operated animals the reflex
autonomic effects evoked by severe hypoxia are
closely similar during spontaneous respiration and
after decamethonium (with ventilation controlled
at approximately the same level). The dose of
atropine used in the present study prevents heart
rate changes during electrical stimulation of the
vagus in the anesthetized rabbit (31), but has no
apparent toxic effect in the unanesthetized animal
(26). We have assumed that its main action is to
block vagal efferents, that its effect on chemoreceptor discharge can be neglected in the doses
used (34), and that when used by itself in series
3, its recently demonstrated effects on sympathetic ganglionic transmission can probably be
neglected in the presence of the normal nicotinic
component (36).
The use of 'de-efferented' animals to assess the
local effects of hypoxia requires that their
autonomic effects have been completely inactivated. The adrenalectomy is complete, and the
animals do not survive when adrenal steroid
replacement therapy is discontinued (25); the
dose of guanethidine blocks sympathetic nerve
transmission to the hindlimb (37) and in normal
rats and rabbits depletes peripheral tissue
catecholamines except in the central nervous
system and adrenal medulla (38-40).
Results
RESTING BEHAVIOR
The sham-operated, rhinencephalic, thalamic, and high mesencephalic animals regained
normal posture and movement within 30 to 40
minutes after operation. The sham-operated
animals made normal sniffing and washing
movements typical of unoperated rabbits, and
appeared to be in no distress. The rhinencephalic and thalamic animals would sit in a
quiet, limp manner, with head drooping, but
after a painful stimulus or loud noise they
would arch their backs and with pupils
dilating, would run forward vigorously (sham
rage). We did not observe the placidity
described by Bard and Mountcastle (16) in
chronic rhinencephalic cats. The pontine
animals lay on their side, with the classic
posture of decerebrate rigidity (14).
RESTING MEASUREMENTS
The preoperative normal resting values in
each animal were compared with the results
obtained 3 hours postoperatively. Respiratory
Circulation Research, Vol. XXIV, June 1969
763
minute volume increased significantly in shamoperated (P < 0.01), rhinencephalic (P <
0.01), and thalamic (P<0.001) animals, but
not in the high mesencephalic and pontine
animals (Table 1). In the rhinencephalic and
thalamic animals, the rise in minute volume
was greater than after sham operation (P =
0.05, P<0.02, respectively), but the rise in
respiratory rate was somewhat less marked. In
both these groups, there was some hypocapnia, with a slight respiratory alkalosis in
thalamic animals (Table 2); alkalosis was
absent in the rhinencephalic animals.
In the first series of experiments, the
postoperative results were obtained during
spontaneous respiration (Fig. 3). In the shamoperated animals there was a moderate rise in
heart rate and fall in arterial pressure to 111%
and 87% of the respective preoperative values
(P<0.05), while the changes in cardiac
output and total peripheral resistance were not
statistically significant. In the rhinencephalic
animals the changes in heart rate, arterial
pressure, and cardiac output differed significantly from the effects in the sham-operated
animals, but the differences were small. In the
thalamic animals, the rise in heart rate was
again greater than after sham operation, and
the changes after operation in blood pressure
and cardiac output also differed slightly
(P<0.05). In the pontine animals, the heart
rate and blood pressure remained unchanged
from the preoperative value, but the reduction
in cardiac output to 79% of preoperative and
the rise in total peripheral resistance to 127%,
were significantly different (P<0.05) from
the effects of sham operation.
In the second series of experiments, the
normal preoperative values (spontaneous respiration) were compared with the postoperative values obtained during controlled ventilation (Fig. 3). The circulatory variables after
operation were now more closely similar in
sham-operated, rhinencephalic, and thalamic
animals, suggesting that the differences in the
spontaneously breathing animals reflected
differences in respiratory or muscular activity
or both, rather than different degrees of
trauma. However, in the pontine group the
ownloaded from http://circres.ahajournals.org/ by guest on June 18, 2017
X
X
TABLE 2
3
3
3
De-efFerented
0.4
1.6
1.9
1.8
0.8
52 ± 2.0
42 ± 1.0
32 ± 1.3
53 ± 1.5
39 ± 2.7
32 ± 1.4
39 ± 2.2
31 ± 1.3
48 ± 0.9
39 ± 2.3
32 ±1.5
53 ±1.6
41 ± 3.0
32 ± 1.1
50 ±
41 ±
36 ±
32 ±
27 ±
51 ±0.8
41 ± 0.8
31 ±1.1
33
34
33
25
19
15
26
22
16
30
29
32
52 ± 1.4
40 ± 2.0
32 ± 1.0
14
16
17
15
14
15
17
34
34
38 ± 1.5
30 ± 0.9
16
16
17
16
15
12
18
16
16
17
15
19
18
15
13
15
18
17
18
34
33
22
23
25
21
18
23
19
28
32
23
19
25
48 ± 0.9
38 ± 1.2
31 ± 1.5
52 ± 1.0
42 ± 1.0
33 ± 0.9
51 ± 0.4
40 ±1.1
35 ± 1.0
30 ± 1.0
26 ± 0.7
1.3
1.3
1.3
1.3
1.1
1.1
1.7
0.9
1.1
1.3
1.0
1.3
1.1
1.1
1.1
0.9
1.3
0.9
Arterial carbon dioxide pressure
(mm Hg)
Hypoxia
R
12 min
32 min
±SEf
7.45
7.50
7.46
7.45
7.44
7.45
7.49
7.46
7.51
7.60
7.54
7.46
7.44
7.49
7.47
7.48
7.50
7.47
R
pH
7.59
7.67
7.68
7.54
7.52
7.52
7.65
7.63
7.63
7.63
7.51
7.43
7.53
7.49
7.60
7.64
7.58
7.62
7.55
7.63
7.59
7.49
7.45
7.38
7.62
7.51
7.61
7.64
7.52
7.54
7.39
7.39
7.59
7.60
7.62
7.64
Hypoxia
12 min
32 min
Respiration was spontaneous in all animals. *Mean ± SEM of each individual group to indicate range of variation. fObtained by analysis of variance
within-animal comparisons. R = at rest.
90 ± 1.7
93 ± 2.4
92 ± 3.0
103 ± 1.2
100 ± 3.0
107 ± 3.8
3
4
4
Pontine
High mesencephalic
96 ± 2.0
96 ± 1.3
107 ± 4.0
112 ±4.2
110 ±4.0
4
Thalamic
3
5
2
6
108 ± 3.8
104 ± 2.8
98 ± 2.0
3
4
4
Rhinencephalic
R
97 ± 1.0
96 ±3.1
95 ± 1.8
99 ± 3.6
96 ± 2.0
rabbits
Normal
Group
No.
Arterial oxygen pressure*
(mm Hg)
Hypoxia
12 min
32 min
Mean Values of Arterial Oxygen Pressure, Carbon Dioxide Pressure, and pH during Rest and during Hypoxia
0.013
0.013
0.013
0.013
0.011
0.011
0.016
0.009
0.011
0.013
0.010
0.013
0.011
0.011
0.011
0.009
0.013
0.009
±SEf
Z
jo
O
NEURAL INTEGRATION IN HYPOXIA
SPONTANEOUS
CONTROLLED
200
m
s
3 ft ID DID]
400
3OO
ZOO
r
rh
Downloaded from http://circres.ahajournals.org/ by guest on June 18, 2017
H Hill Oil
loo 1 -
l—LJ
l_LJ
Sh
Rh
l_Ld
Th I Po
Sh
Rh
Th
Po
FIGURE 3
Left: Comparisons of the mean normal resting preoperative value, with the value obtained 3 hours
after surgery in 13 sham-operated (Sh), 8 rhinencephalic (Rh), 12 thalamic (Th), and 11 pontine (Po)
rabbits, breathing spontaneously. The variables were
heart rate/min (H.R.), arterial pressure, mm Hg (B.P.),
cardiac output (CO.) in ml/min and total peripheral
resistance (T.P.R.) in arbitrary units. A significant
(P < 0.05) change from preoperative value in each
group is shown as stippled postoperative value. Solid
circle above a preparation indicates that its response
differs (P < 0.05) from effect of sham operation. The
height of a rectangle equals mean value, and the
height of the line near the upper edge of the preoperative rectangle is twice SE of the mean pre- or
postoperative value, obtained by Analysis of Variance. Right: Similar comparisons in 3 normal, 4
rhinencephalic, 3 thalamic and 3 pontine animals. The
preoperative normal value was obtained during spontaneous respiration; the postoperative value was obtained during artificial ventilation after administration
of decamethonium.
total peripheral resistance again rose more
markedly than after sham operation (P =
0.05), thus showing the same effect as in
spontaneously breathing animals.
BEHAVIOR DURING HYPOXIA
The normal (sham-operated) animals became restless for about half a minute after the
start of hypoxia, especially with the lower
inspired oxygen mixtures, often sniffing near
the front of their box as though looking for a
means of escape. However, they quickly
settled down and remained quiet at all levels
Circulation Research, Vol. XXIV, June 1969
20
20
TIME-
60
20
60
minutes
FIGURE 4
Mean changes in respiratory minute volume (VE)
and frequency (f), during hypoxia at Pao2 about 50
mm Hg in 3 normal, 3 rhinencephalic, 4 thalamic,
and 3 pontine animals; 40 mm Hg in 5 normal, 4
rhinencephalic, 3 thalamic, and 4 pontine; and 30
mm Hg in 4 normal, 4 rhinencephalic, 5 thalamic, and
4 pontine, expressed as each group's percent resting
value just before hypoxia. Symbol of each group on
left is ± 1 SE of mean at a single time interval.
of Pac>2 studied, without apparent loss of
consciousness. In rhinencephalic and thalamic
animals, the initial movements were less
purposeful, but of about the same duration as
in normal animals. There was little increased
activity in the high mesencephalic and pontine
animals. No ill effects were observed after
hypoxia in any of the preparations.
RESPIRATORY RESPONSE
In each preparation we studied the responses to at least three levels of hypoxia at
approximately Pao2 50, 40, and 30 mm Hg
(Table 2, Fig. 4). In the normal animal at
Pao2 50 mm Hg, the respiration increased
KORNER, UTHER, WHITE
766
TABLE 3
Percent Change from Resting Values in Respiratory Minute Volume and Rate
No.
rabbits
Pao2
VE
Normal
3
6
4
4
6»
50
41
36
31
26
Rhinencephalic
3
4
4
52
41
32
146 it 4.9
179 ± 3.2
202 ± 2.9
194 ± 5.8
176 ± 7.3
129 ± 3.8
4
48
39
31
Group
Thalamic
3t
5t
Downloaded from http://circres.ahajournals.org/ by guest on June 18, 2017
High mesencephalic
Pontine
De-efferented
39
30
2*
6+
3
4*
4*
52
40
32
3
3
3
51
41
31
174 it 6.3
180 ± 2.4
135 ± 2.6
123 ± 3.4
138 ± 3.7
249 ± 4.6
222 it 5.0
139 ± 3.2
207 ±6.1
218 ± 6.6
147 ± 2.7
187 ± 8.1
170 ± 7.3
102 ± 2.8
128 ± 2.8
142 ± 3.2
118 ±3.2
121 ±4.5
108 ±1.6
140 ± 5.3
127 ± 2.9
115 ±2.1
114 ± 2.0
107 ± 1.9
150 ± 3.3
148 ±1.8
116 ±1.8
131 ± 2.4
136 ± 2.2
130 ± 3.2
133 ± 5.9
127 ± 6.6
Values are means ± SE of the mean difference based on comparisons within animals. Pao,,
values are means from samples 12 and 32 minutes after start of hypoxia. Mean respiratory
effects were calculated from the integrated time-response curves (Fig. 4).
*VE and f significantly lower (P = 0.01) than in normal animal at Pao,, 36 mm Hg.
f VE significantly less (P < 0.001) than in other preparations at corresponding Pao2.
|VE significantly greater (P<0.05) than in normal, rhinencephalic, and thalamic animals
at corresponding Pao2.
soon after the onset of hypoxia, and rose
within about 5 minutes to an approximate
plateau level. At Pao2 40 and 30 mm Hg,
ventilation increased to a maximum within 1
minute from the onset of hypoxia, settling to a
lower plateau level 5 to 10 minutes later. The
rise in respiratory rate was less marked than
the rise in minute volume. Two other groups
of normal animals were studied at PaO2 35 mm
Hg and 26 mm Hg. The results in Table 3
indicate that the greatest mean respiratory
response was attained at approximately Pao2
35 mm Hg (202 ±2.9 (SE)% of resting); at
Pao2 30 mm Hg, the response was almost
unaltered, but at 26 mm Hg, it had fallen off
significantly to a mean of 176 ±7.3$ of
resting.
At each level of Pao2, respiratory response
in rhinencephalic animals was closely similar
to normal (Table 3, Fig. 4). In the thalamic
animals the response was considerably less at
Pao2 40 and 30 mm Hg than in the other
preparations. On the other hand, the pontine
and high mesencephalic animals had an
essentially normal or somewhat increased
response at all levels of PaOo. The mean pH
change during hypoxia in normal and deefferented animals was in agreement with
previous findings (17, 20, 31) and showed
some respiratory alkalosis (Table 2). This was
somewhat less marked in thalamic and high
mesencephalic animals. However, the pontine
and rhinencephalic animals developed a moderate acidosis toward the end of hypoxia,
possibly related to greater epinephrine release
(see Discussion).
Circulation Research, Vol. XXIV, June
1969
NEURAL INTEGRATION IN HYPOXIA
NORMAL
•
30
767
THALAM1C
RHINENCEPHALIC
•
t
30 1
*
30
MESENCEPHALIC
1
t
30 t
PONTINE
I
30l
r
Downloaded from http://circres.ahajournals.org/ by guest on June 18, 2017
60
20
60
20
60
20
60
20
60
TIME - minutes
FIGURE 5
Mean changes during severe hypoxia (Paos 30 mm Hg) in heart rate (H.R.), arterial pressure
(B.P.), cardiac output (CO.), and total peripheral resistance (T.P.R.), all expressed as percent
of resting, in 4 normal, 4 rhinencephalic, 5 ihalamic, 6 high mesencephalic and 4 pontine
rabbits, breathing spontaneously. Symbol on left of each variable is ± 1 SE of mean at a single
time interval.
CIRCULATORY CHANGES AT Poo- 30 mm Hg
The results in the sham-operated group
were similar to findings in rabbits subjected to
only the minor cannulation procedures and
previous implantation of the thermistor (17,
19). During severe hypoxia, there was bradycardia, transient early reduction in cardiac
output followed by a subsequent increase
above the resting value, and transient early
elevation in arterial pressure and in total
peripheral resistance (Fig. 5). The circulatory
findings in the thalamic animals were closely
similar. In the rhinencephalic animals, marked
bradycardia was also present, but the cardiac
output increased more rapidly than normal,
and the early rise in total peripheral resistance
was less pronounced. In the pontine animals,
the early bradycardia was either minute or
absent, and in all animals, heart rate and
cardiac output increased rapidly soon after the
onset of hypoxia, while the total peripheral
resistance fell to 80% of resting. The circulatory response of the pontine group thus
differed more markedly from normal than that
of the other groups. In the high mesencephalic
animals in which posture and movement were
Circulation Research, Vol. XXIV. June 1969
essentially normal, a response closely similar
to that of pontine animals was observed.
In the normal rabbit, the reflex changes in
respiration and circulation during hypoxia are
primarily evoked from the receptors of the
carotid sinus and aortic arch region (17-20).
After section of the carotid sinus and aortic
nerves in animals with an intact central
nervous system, the resting total peripheral
resistance was about 50% higher than before
nerve section, and during hypoxia the respiration no longer rose, the bradycardia was
greatly reduced, and the arterial pressure and
total peripheral resistance fell markedly instead of rising (Fig. 6). In pontine animals,
the resting total peripheral resistance was also
increased by about 50% after section of the
carotid sinus and aortic nerves, and during
hypoxia the rise in respiration and tachycardia
were virtually abolished. The same large
reduction in arterial pressure and total peripheral resistance now occurred as in animals
with intact central nervous system (Fig. 6),
probably as a result of the local dilator effects
of hypoxia (17, 20). The reason for the
greater magnitude of vasodilatation compared
768
KORNER, UTHER, WHITE
NORMAL
%
1
t
t
,-.
250 200
30
PONTINE
30
"
' "'
f
No
/
' * • ' . . ,
•
^
1
" "
V
- - % . . ,
UJ
|
"1
> 150
•il
;
-
r
100
125
. 100
r
75
50
'••i..
Downloaded from http://circres.ahajournals.org/ by guest on June 18, 2017
200
150
QL
i 100
50
20
60
TIME -
20
60
minutes
FIGURE 6
Mean changes in respiratory minute volume (VE)
heart rate (H.R.), arterial pressure (B.P.) and total
peripheral resistance (T.P.R.), all expressed as percent
of resting, during hypoxia (Pao2 30 mm Hg) in spontaneously breathing animals with intact central nervous system (normal), and in pontine animals. The
former group comprises 4 animals with carotid sinus
and aortic nerves (C & A) intact (broken lines), and
3 animals with section of these nerves (solid lines).
A similar notation is used to show responses of 4
pontine animals with these nerves intact and 3 pontine
animals with section of these nerves. Mean absolute
values of the variables in order, with nerves intact,
then after section: Normal rabbits: VE = 1.37 liters/
min, 0.92 liters/min; H.R. = 273/min,
306/min;
B.P. = 84 mm Hg, 126 mm Hg; T.P.R. 175 units,
261 units. Pontine rabbits: VE = 1.31 liters/min,
0.91 liters/min; H.R. = 245/min, 261/min; B.P. = 92
mm Hg, 110 mm Hg; T.P.R. = 218 units, 340 units.
with de-efferented rabbits probably depends
on differences in Paco2, since de-efferented
animals hyperventilate normally (31). The
findings suggest that the carotid and aortic
reflex zones are the primary sources of
respiratory and autonomic excitation during
hypoxia in pontine as well as in normal
animals.
20
60
T I M E - minutes
FIGURE 7
Time-autonomic total peripheral resistance effect
curves at approximately Pao2 50, 40, and 30 mm Hg
in the various preparations of animals breathing
spontaneously. The number of animals is the same
as in Figure 4.
AUTONOMIC EFFECTS DURING
SPONTANEOUS RESPIRATION
Autonomic Total Peripheral Resistance Effect.—In the normal animals, the mean autonomic total peripheral resistance effect over
the whole period of hypoxia at Pao2 50, 40,
and 35 mm Hg increased by a small but statistically significant amount (P<0.02) to an
approximately constant value (516 ± 115, 603
± 130, 490 ± 140 arbitrary units, respectively)
and then increased sharply at Pao2 30 mm Hg
to 1530 ± 214 units, following attainment of
the maximum respiratory response (Table 3).
A similar rise in mean autonomic total peripheral resistance effect was observed at Pao2 26
mm Hg (1730 ± 214 units), when there was
some falling off in respiratory response.
The time-autonomic total peripheral resistance effect curves (Fig. 7) in the three
Circulation Research, Vol. XXIV, June 1969
NEURAL INTEGRATION IN HYPOXIA
I
Downloaded from http://circres.ahajournals.org/ by guest on June 18, 2017
20
so l
60
I
40
20
I
769
•
30 •
„
60
TIME- minutes
FIGURE 8
Time-net autonomic heart rate effect curves at Pao2
about 50, 40, and 30 mm Hg in the same animals
as in Figures 4 and 7.
suprabulbar preparations typically showed a
transient increase in autonomic total peripheral resistance effect, and a smaller late effect,
both components increasing in magnitude
with increasing severity of hypoxia. The early
transient response was completely suppressed
in normal and rhinencephalic animals at Pao2
50 mm Hg and was most pronounced in the
thalamic animals. In the rhinencephalic animals at Pao2 40 mm Hg, the early rise in
autonomic effect was maintained during the
latter part of hypoxia at a higher level than in
normal (P<0.05) or thalamic (P = 0.01)
animals at this Pao2. At Pao2 30 mm Hg, each
of the three suprabulbar preparations had
prominent early and late autonomic total
peripheral resistance effects. In the pontine
animals at Pao2 50 and 40 mm Hg, the
autonomic effects did not change significantly
from the resting level, while at Pao2 30 mm
Hg, the mean response during hypoxia averCirculalion Research, Vol. XXIV, June 19C9
aged 40% of the value observed in the
suprabulbar preparations, but only the early
effect was statistically significant.
Autonomic Heart Rate Effect.—In the three
suprabulbar preparations, the autonomic effect typically consisted of cardiac slowing
during the early part of hypoxia, becoming
less marked with time (Fig. 8). The early
autonomic slowing became more pronounced
with increasing severity of hypoxia in all
suprabulbar preparations, and was most prominent in the thalamic animals at each Paoo. In
normal animals at Pao2 50 mm Hg, the early
autonomic slowing was completely suppressed
and there was a gradual autonomic rise in
heart rate. Late autonomic cardiac slowing
was more pronounced in rhinencephalic animals at Pao2 40 mm Hg than in the other
preparations (P < 0.01).
In the pontine animals, there was a gradual
autonomic increase in heart rate at Pao2 50
and 40 mm Hg, rising to a plateau during the
first 10 to 20 minutes of hypoxia, with a time
course similar to that of the normal animal at
Pao2 50 mm Hg. At Pao2 30 mm Hg, the
response of the pontine animal was biphasic,
with a very brief period of autonomic cardiac
slowing followed by the more gradual autonomic rise in heart rate. The transient cardiac
slowing in the pontine animal corresponds to
the time of initial rise in arterial pressure (Fig.
5). In one experiment in a pontine animal,
raising the carotid sinus pressure in the
isolated sinus by 100 mm Hg for a few
seconds resulted in reflex bradycardia due to
increased vagal and reduced cardiac sympathetic activity. However, hypoxia never resulted in prolonged autonomic cardiac slowing in
the pontine animal.
AUTONOMIC EFFECTS DURING CONTROLLED
VENTILATION (2 liters/min)
The differences in autonomic total peripheral resistance effects observed during the early
part of hypoxia between spontaneously
breathing normal, rhinencephalic, and thalamic animals were now greatly reduced (Fig.
9), as were corresponding differences in
autonomic heart rate effects (Table 4). In the
pontine animals, the autonomic total periph-
KORNER, UTHER, WHITE
770
TABLE 4
Mean Change in Autonomic Heart Rate Effect over Ten-Minute Period of Hypoxia
Group
Normal
No drug
Atropine
Rhinencephalic
No drug
Atropine
Thalamic
No drug
Atropine
Pontine
No drug
Atropine
PaOa 50 nun Hg
PaO2 40 mm Hg
PaOs 30 mm Hg
-3.0 ± 2.4
-2.0 ±1.7
-19.0 ±2.9*
- 3.0±3.6f
- 3 5 . 0 ± 2.5*
-12.0 ± 1.6*t
+3.6 ± 1.4
-22.0 ± 3.7*
-26.0 ± 2.2*
+4.4 ± 1.3
+1.2 ±2.3
-17.0 ± 3.3*
-15.0 ±2.7*
-28.0 ± 5.2*
-10.0 ± 1.8*t
+2.0 ± 1.6
+5.2 ±1.6
- 0.2 ±2.4
+ 10.6±3.0f
+ 3.8 ± 1.9
+ 6.4 ± 2 . 6
Values are means ± SE of means for three rabbits with ventilation controlled at 2 liters/min.
*P < 0.05. fP < 0.05 for difference from response with no drug.
Downloaded from http://circres.ahajournals.org/ by guest on June 18, 2017
eral resistance effect was still significantly
smaller (P < 0.01), its greatest increase at
Pao2 30 mm Hg being again about 40% of the
value in the suprabulbar groups. In the
pontine animals there was either no change or
an increase in autonomic effect on heart rate,
while in the suprabulbar animals there was
progressively more marked autonomic cardiac
slowing with more severe hypoxia.
In normal animals at Pao2 50 mm Hg, the
small autonomic change in heart rate was little
affected by atropine (Table 4), but at 40 mm
Hg the autonomic cardiac slowing was almost
completely abolished, suggesting that it was
largely vagal in origin at this Pao2; at 30 mm
Hg, it was reduced by about two-thirds after
atropine, suggesting that there was now an
additional component due to reduction in
cardiac sympathetic activity. In thalamic
animals the autonomic cardiac slowing at Pao2
40 mm Hg was unaffected by atropine,
suggesting that it was predominantly due to
reduced cardiac sympathetic activity, but at
30 mm Hg, it appeared to depend on
increased vagal activity and reduced sympathetic activity in the same way as in normal
animals. In the pontine group the autonomic
rise in heart rate was slightly greater after
atropine (0.1 > P>0.05), suggesting slight
masking of the early autonomic increase in
heart rate by increased vagal activity.
• 50 •
I 40 1
• 30 •
160,-
NORMAL
80
RHINENCEPHALIC
80
o;
o
160
o
o
80
THALAMIC
0
801
PONTINE
5
10 15 20
5 10 15 20
TIME minutes
5
10 15 20
FIGURE 9
Time-autonomic total peripheral resistance effect
curves at Paos of approximately 50, 40, and 30
mm Hg, obtained at each level in 3 normal, 4 rhinencephalic, 3 thalamic and 3 pontine animals during
10 minutes of hypoxia with ventilation controlled at
2 liters/min in all experiments.
Circulation Research, Vol. XXIV, June 1969
771
NEURAL INTEGRATION IN HYPOXIA
Discussion
NEUROLOGICAL PREPARATIONS
Downloaded from http://circres.ahajournals.org/ by guest on June 18, 2017
The results suggest that nonspecific factors
resulting from the surgical interventions exert
relatively few effects 3 hours postoperatively
since (1) the effects of sham operation on the
cardiorespiratory variables are slight; (2)
controlling ventilation almost abolishes the
differences in resting circulatory variables
between the three suprabulbar preparations;
(3) all groups can withstand prolonged
hypoxia without apparent ill effect. The
somewhat greater muscular activity in response to slight stimuli observed in rhinencephalic and thalamic animals is probably a
specific release phenomenon after ablation of
the neocortex (14-16), and the rise in
respiratory minute volume may be associated
with this. The elevation in resting total
peripheral resistance in the pontine animals is
probably a specific effect of the lesion; it does
not appear to result from baroreceptor involvement due to greater blood loss, since
after section of the carotid sinus and aortic
nerves there is a further rise in total peripheral
resistance similar in magnitude to that found
in normal animals (19, 24).
CARDIORESPIRATORY CONTROL DURING HYPOXIA
The peripheral receptors which contribute
to the normal rabbit's cardiorespiratory response to hypoxia include the arterial chemoreceptors and baroreceptors, which produce
autonomic excitation (17-19), and the effects
of lung inflation mediated through vagal
afferents, which inhibit the primary circulatory effects of strong chemoreceptor stimulation (22, 41, 42). The maximum initial
autonomic effects suggest involvement of ratesensitive elements in the control system at a
time when the Pao2 is changing rapidly, but
equilibrium between the local and autonomic
effects at the periphery has not yet been
reached, while the attenuated late response
suggests proportional elements responding
primarily to the gradual changes in state of
the peripheral beds with prolonged exposure
to the severe hypoxic stress.
RESPIRATORY CONTROL
The respiratory response to hypoxia of the
Circulation Research, Vol. XXIV, June 1969
NEOCORTEX
RHINENCEPHALON
DIENCEPHALON
LUNG INFLATION
RECEPTORS
(Thalamic)
CHEMORECEPTORS
BARORECEPTORS
FIGURE 10
Schema of suggested inputs to, and outputs from,
cardiovascular and respiratory centers in pons and
medulla (pontine animals), diencephalon (thalamic
animals), rhinencephalon and neocortex concerned
tvith cardiorespiratory integration in hypoxia. Centers
A and B in rhinencephalon influence the early and
late components, respectively, of the autonomic effector response, and have no special anatomical significance. High mesencephalic centers are not considered
to contribute to reflex response to hypoxia in view of
identity of effects in pontine and high mesencephalic
animals. Inputs arising from arterial chemoreceptors,
baroreceptors, and lung inflation receptors relay in
the bulb and along ascending relays suggested by
upward flowing arrows (solid lines, known projections;
broken lines, suggested projections); projections including vagal afferents to pons and medulla concerned
with Hering-Breuer reflex, possible baroreceptor afferents to pontine respiratory center and possible
afferents from muscle stretch receptors, are not shown
in this schema. Downward flowing large black arrows
indicate excitation of underlying centers or effectors;
white arrows indicate inhibition of -underlying centers;
half white and half black arrows from diencephalon to
bulbar heart rate centers indicate that autonomic cardiac slowing results from increased vagal and reduced
cardiac sympathetic activity. Effectors: V = vagus;
C.S. = cardiac sympathetic; S.-A. = sympathoadrenal
effects on systemic vessels; Resp. = Respiration. For
further description see text.
pontine animals is slightly greater than normal
and arises from arterial chemoreceptor stimulation. In the thalamic preparation the rise in
respiration is markedly smaller, and this is
772
Downloaded from http://circres.ahajournals.org/ by guest on June 18, 2017
probably not just a mechanical limitation of
respiratory performance due to high resting
ventilation, since the rhinencephalic animal
has an almost normal response to hypoxia.
The net chemoreceptor drive during hypoxia probably differs little in the various
preparations. In normal and thalamic animals,
the arterial blood composition is similar, and
the potentially stronger stimulus in rhinencephalic and pontine rabbits due to lower pH
appears to be offset by greater reduction in
Paco2 in these animals (Table 2), and their
respiratory responses are approximately the
same as in normal rabbits.
The present results suggest that (1) the
diencephalon limits the increase in respiration
during hypoxia by exerting inhibitory influences on the pontine centers and (2) that this
inhibitory influence is in turn inhibited by
activation of rhinencephalic centers (with
probably a small additional contribution from
the neocortex), thus allowing the normal
animal to have almost as great a respiratory
response as the pontine animal down to a Pao2
of 35 to 30 mm Hg, and a greater response
than the thalamic animal (Fig. 10). The
afferent sources for activating the suprabulbar
respiratory mechanisms have not been studied
in these experiments. In the cat and monkey,
suprabulbar respiratory centers in the diencephalon and cortex have been identified in
electrical stimulation studies (1, 4, 13), and
these may correspond to the suprabulbar
respiratory centers in the rabbit and also
participate in the reflex control of respiration.
CIRCULATORY CONTROL
The autonomic effector response of the
suprabulbar preparations differs from that of
pontine animals in having a greater rise in
autonomic total peripheral resistance effect at
each Pao2 and in having directionally different
heart rate effects. The transition from suprabulbar to pontine autonomic effector response
is sharp, between the thalamic and high
mesencephalic preparations. The differences
between the three suprabulbar preparations
are at least partly related to differences in
respiration, muscular activity, or both, since
they are greatly reduced during controlled
KORNER, UTHER, WHITE
ventilation plus decamethonium, but the
differences between suprabulbar and pontine
responses are still present after controlling
ventilation. The difference in heart rate effects
does not depend on a different profile of
baroreceptor stimulation, since the arterial
pressure response during hypoxia is the same
in normal and pontine animals (Fig. 5). The
elevated resting total peripheral resistance of
the pontine animal could have contributed to
the difference in estimated autonomic total
peripheral resistance effect. However, an even
larger difference in resting total peripheral
resistance is observed between pontine and
normal animals subjected to section of the
carotid sinus and aortic nerves, and in these
the fall in resistance due to identical local
effects of hypoxia is the same when expressed
as a percent of each group's resting value.
The smaller autonomic total peripheral
resistance effect during hypoxia in the pontine
animal could be due to a smaller rise in
sympathetic neural constrictor activity, a
greater amount of beta-receptor dilator activity owing to greater secretion of epinephrine
—virtually the rabbit's sole adrenal medullary
hormone (43)—or both. The moderate 'metabolic' acidosis which occurs in pontine (and
rhinencephalic) rabbits (Table 2) probably
reflects a greater epinephrine secretion (44);
it is not observed in autonomically de-efferented animals, suggesting that it results
from a reflex change in autonomic effector
activity rather than from prolonged hypoxia in
a preparation with inadequate circulatory response.
CARDIORESPIRATORY INTEGRATION
Figure 10 presents a tentative hypothesis for
the integration of cardiorespiratory effects
during hypoxia in the rabbit, relating the
present findings to the inputs from the various
peripheral receptors (17, 19, 22). Afferents
from lung inflation receptors (Hering-Breuer
afferents) and possible baroreceptor afferents
to the pontine respiratory centers may have
relevance in cardiorespiratory integration in
hypoxia, but are not shown in Figure 10 since
they were not investigated in this study.
In pontine animals, the primary stimulus
Circulation Research, Vol. XXIV, June 1969
NEURAL INTEGRATION IN HYPOXIA
Downloaded from http://circres.ahajournals.org/ by guest on June 18, 2017
arises, as in normal animals, from the carotid
and aortic regions, and the circulatory findings
suggest autonomic excitation through the
arterial baroreceptor system (17, 19). Normally an autonomic rise in heart rate would occur
in response to a fall in arterial blood pressure,
but the present findings in hypoxia can be
explained if we postulate that chemoreceptor
and baroreceptor afferents terminate on common neurones in the bulbar circulatory
centers. If we suppose that the increased
chemoreceptor activity renders these neurones
less than normally responsive to incoming
baroreceptor traffic, through inhibitory mechanisms (45), this will alter the "set" reference
blood pressure level for the baroreceptor reflex
to a value somewhat above resting. Under
these conditions: (1) Autonomic cardiac
slowing will occur only during the largest
blood pressure rise in the first minute of
hypoxia (Figs. 5 and 8). (2) An autonomic
rise in heart rate will occur even though the
blood pressure is slightly elevated above
resting, and the heart rate effect will increase
further as the blood pressure tends to fall
during the latter part of hypoxia. (3)
Sympathoadrenal activity to the peripheral
vessels will also increase while the blood
pressure is somewhat raised.
Since the thalamic animal, unlike the
pontine and high mesencephalic animals, has
an autonomic effector response similar to
hypoxia in a normal animal, the diencephalic
vasomotor centers would appear to play a
major part in the normal response. The
autonomic effects depend to an important
extent on arterial chemoreceptor stimulation
(17-19), and occur during perfusion of the
carotid chemoreceptors with hypoxic blood
(46), suggesting that the diencephalic vasomotor center is one important suprabulbar
projection site for chemoreceptor afferents (8,
9). Chemoreceptor afferents, baroreceptor
afferents, or both (8-10, 26) may also project
to the rhinencephalon, accounting for the late
sustained effects in the rhinencephalic animals
at Paoo 40 mm Hg.
The autonomic effects of strong chemoreceptor stimulation are inhibited by the effects
Circulation Research, Vol. XXIV, June 1969
773
of hyperventilation mainly through vagal
afferents (22, 41, 42). The vagus has known
cortical projections (47), and if we postulate
that the main input from lung inflation
receptors terminates in the rhinencephalon
and neocortex this could explain (1) the
attenuation of the early autonomic total
peripheral resistance effect in rhinencephalic
animals compared with the thalamic at Pao2
50 and 30 mm Hg; (2) the complete absence
of autonomic cardiac slowing in normal
animals at Pao2 50 mm Hg; and (3) the
attenuation in the normal animal at Pao2 40
mm Hg of the late rhinencephalic response.
The similar time course of the autonomic rise
in heart rate in normal and pontine animals at
Pao2 50 mm Hg suggests that it probably
normally arises from the bulbar heart rate
centers as a result of neocortical inhibition of
rhinencephalic and thalamic mechanisms.
Vagal afferents may also project to the
diencephalon, since the circulatory findings
are similar in artificially ventilated thalamic
animals to those in the other suprabulbar
preparations.
The various regions of the central nervous
system probably compare information arising
from the various peripheral receptors during
hypoxia. In mild and moderate hypoxia, the
relative magnitudes of input from lung
inflation and from the chemoreceptors will
result in inhibition of most of the early
autonomic effects arising from the diencephalic centers in the normal animal. This inhibition would allow each peripheral bed to adjust
its blood flow according to its needs, with a
modest degree of control of the blood
pressure. With more severe hypoxia the input
from the chemoreceptors will increase more
than the input from the lung inflation
receptors, and the autonomic cardiac slowing
and increased total peripheral resistance effects become more prominent, resulting in
major redistribution of peripheral blood flow
(20). The falling off in respiratory response in
the normal animal (Table 3) can be explained
if strong stimulation of the diencephalic
vasomotor centers is associated with stimulation of the diencephalic (inhibitory) respira-
774
KORNER, UTHER, WHITE
Downloaded from http://circres.ahajournals.org/ by guest on June 18, 2017
tory centers rendering the cortical respiratory
disinhibition less effective, and would reduce
the work of breathing at a time when the
alveolar ventilation could no longer maintain
the animal's oxygen supply (29).
The rabbit is a 'small-lung' burrowing animal, with a cardiorespiratory response to hypoxia intermediate between that of diving
animals to submersion asphyxia (with maximum bradycardia and peripheral vasoconstriction and complete respiratory inhibition
[48-50]), and that of man or the dog to
hypoxia (with a large hyperventilation response [51] and only slight circulatory adjustments [17, 52]). However, the effects of
hypoxic stimulation of the isolated chemoreceptors, the reflex effects through baroreceptors, and the effects of lung inflation are
qualitatively similar in the rabbit and dog
(17, 19, 22, 41, 42, 46, 53) and the species
differences appear thus to be mainly quantitative. The large lung and greater respiratory
response to hypoxia in man or the dog will
maintain a relatively high Pao2 for a given
inspired oxygen pressure, resulting in a preponderance of reflex effects due to lung inflation with inhibition of primary circulatory
chemoreceptor effects. In the rabbit, however,
the limited respiratory response will be associated with more marked circulatory chemoreceptor effects at higher inspired oxygen
pressure and its capacity for circulatory adjustments through activation of the various
suprabulbar mechanisms becomes unmasked
more easily than in the dog (17, 41, 53).
References
1. RANSONT, S. W., AND MAGOUN, H. W.:
The
hypothalamus. Ergeb. Physiol. 41: 56, 1939.
2. HOFF, E. C , KELL, J. F., JR., AND CARROLL, M.
N., JR.: Effects of cortical stimulation and
lesions on cardiovascular function. Physiol.
Rev. 43: 68, 1963.
3. WALL, P. D., AND DAVIS, G. D.: Three cortical
systems affecting autonomic function. J. Neurophysiol. 14: 507, 1951.
4. LIVINGSTON, R. B., FULTON, J. F., DELGADO, J.
N. R., SACHS, E., JR., BRENDLER, S. J., AND
DAVIS, G. D.: Stimulation and regional
ablation of orbital surface of frontal lobe. Res.
Publ. Assoc. Nervous Mental Dis. 27: 405, 1948.
1948.
5. KAADA, B. R., FRIBRAM, K. H., AND EPSTEIN, J.
A.: Respiratory and vascular responses in
monkeys from temporal pole, insula, orbital surface and cingulate gyrus. J. Neurophysiol. 12:
347, 1949.
6. PITTS, R. F., LARRABEE, M. G., AND BRONK, D.
W.: Analysis of hypothalamic cardiovascular
control. Am. J. Physiol. 134: 359, 1941.
7. MANNING, J. W.: Cardiovascular reflexes following lesions in the medullary reticular formation.
Am. J. Physiol. 208: 283, 1965.
8. BIZZI,
E.,
LIBRETTI,
A.,
MALLIANI, A., AND
ZANCHETTI, A.: Reflex chemoceptive excitation
of diencephalic sham rage behavior. Am. J.
Physiol. 200: 923, 1961.
9. BACCELLI, G., GUAZZI, M., LIBRETTI, A., AND
ZANCHETTI, A.: Pressoceptive and chemocep-
tive reflexes in decorticate and in decerebrate
cats. Am. J. Physiol. 208: 708, 1965.
10. BARTORELLI, C ,
BIZZI, E., LIBRETTI, A., AND
ZANCHETTI, A.: Inhibitory control of sinocarotid pressoceptive afferents of hypothalamic
autonomic activity and sham rage behavior.
Arch. Ital. Biol. 98: 308, 1960.
11. PEISS,
C.
N.:
Concepts
of
cardiovascular
regulation; past, present and future. In
Nervous Control of Heart Rate, edited by W.
C. Randall. Baltimore, Williams & Wilkins,
1965, p. 154.
12. FOLKOW,
B.,
HEYMANS,
C,
AND NEIL,
E.:
Integrated aspect of cardiovascular regulation.
In Handbook of Physiology, sec. 2, vol. 3,
Circulation, edited by W. F. Hamilton and P.
Dow. Washington, D. C , American Physiological Society, 1965, p. 1787.
13. WANC,
S.
C,
AND NGAI,
S.
H.:
General
organization of central respiratory mechanisms.
In Handbook of Physiology, sec. 3, vol. 1,
Respiration, edited by W. O. Fenn and H.
Rahn. Washington, D. C , American Physiological Society, 1964, p. 487.
14. BARD, P., AND MACHT, B.: Behavior of chronical-
ly decerebrate cats: CIBA Foundation Symposium on the Neurological Basis of Behavior,
edited by G. E. W. Wolstenholme. London,
Churchill, 1958, p. 55.
15. BARD, P., AND RIOCH, D. MCK.: Study of four
striatal cats deprived of neocortex and additional parts of the forebrain. Bull. Johns
Hopkins Hosp. 40: 73, 1937.
16. BARD,
P.,
AND MOUNTCASTLE, V.
B.:
Some
forebrain mechanisms involved in the expression of rage with special reference to the
suppression of angry behavior. Res. Publ.
Assoc. Nervous Mental Dis. 27: 362, 1948.
17. KORNER, P. I.: Role of the arterial chemoreceptors and baroreceptors in the circulatory
response to hypoxia of the rabbit. J. Physiol.
(London) 180: 279, 1965.
Circulation Research, Vol. XXIV, June 1969
775
NEURAL INTEGRATION IN HYPOXIA
18.
AND MOK, H. I. Y.: Role of the sympathetic
nervous system in the circulatory response of
the rabbit to arterial hypoxia. J. Physiol.
(London) 181: 175, 1965.
19.
CHALMERS, J. P., KORNER, P. I., AND WHITE, S.
W.: Relative roles of the aortic and carotid
sinus nerves in the rabbit in the control of
respiration and circulation during arterial
hypoxia and hypercapnia. J. Physiol. (London)
188: 435, 1967.
20.
CHALMERS, J. P., KORNER, P. I., AND WHITE, S.
W.: Local and reflex factors affecting the
distribution of the peripheral blood flow during
arterial hypoxia in the rabbit. J. Physiol.
(London) 192: 537, 1967.
21. KORNER, P. I., CHALMERS, J. P., AND WHITE, S.
Downloaded from http://circres.ahajournals.org/ by guest on June 18, 2017
W.: Some mechanisms of reflex control of the
circulation by the sympatho-adrenal system.
Circulation Res. 21 (suppl. I l l ) : III-157, 1967.
22.
CROCKER, E. F., JOHNSON, R. O., KORNER, P. I.,
UTHER, J. B., AND WHITE, S. W.: Effects of
hyperventilation on the circulatory response of
the rabbit to arterial hypoxia. J. Physiol.
(London) 199: 267, 1968.
23.
KORNER,
P. I.,
AND UTHER,
J. B.:
MONNIER,
M., AND GANCLOFF,
H.:
Atlas
34.
KORNER,
P.
I.,
AND EDWARDS,
A.
W.
29.
KORNER,
P.
I.,
AND EDWARDS,
A.
W.
T.:
30. COCHRAN, W. G., AND Cox, G. M.: Experimental
Designs. New York, Wiley, 1953.
31.
KORNER, P. I., LANGSFORD, C , STARR, D., UTHER,
J. B., WARD, W., AND WHITE, S. W.: Effects of
chloralose-urethane and sodium pentobarbitone
Circulation Research, Vol. XXIV. June 1969
V.,
AND BELEN'KII,
M.
L.:
of the Cardiovascular System. London,
Churchill, 1958, p. 197.
36. BROWN, A. M.: Cardiac sympathetic adrenergic
pathways in which synaptic transmission is
blocked by atropine sulphate. J. Physiol.
(London) 191: 271, 1967.
37.
CHALMERS, J. P., KORNER, P. I., AND WHITE, S.
W.: Distribution of peripheral blood flow in
primary tissue hypoxia induced by inhalation
of carbon monoxide. J. Physiol. (London)
192: 549, 1967.
38.
CASS, R., KUNTZMAN, R., AND BRODIE, B. B.:
Norepinephrine depletion as a possible mechanism of action of guanethidine (SU 5864), a
new hypotensive agent. Proc. Soc. Exptl. Biol.
Med. 103: 871, 1960.
39. CASS, R., AND SPRIGGS, T. L. B.: Tissue amine
levels and sympathetic blockade after guanethidine and bretylium. Brit. J. Pharmacol. 17:
442, 1961.
40.
ABERCROMHIE, G. F., AND D A VIES, B. N.: Action
of guanethidine with particular reference to the
sympathetic nervous system. Brit. J. Pharmacol.
20: 171, 1963.
41.
DALY, M. DE B., AND HAZZLEDINE, J. L.: Effects of
artificially induced hyperventilation on the
primary cardiac reflex response to stimulation
of the carotid bodies in the dog. J. Physiol.
(London) 168: 872, 1963.
T.:
Cardiac output during the steady-state in oxygen lack in the unanesthetized rabbit and its
relation to early circulatory response.
Quart. J. Exptl. Physiol. 45: 129, 1960.
S.
35. HEYMANS, C , AND NEIL, E.: Reflexogenic Areas
for
Immediate effects of acute hypoxia on the
heart rate, arterial pressure, cardiac output,
and ventilation of the unanesthetized rabbit.
Quart. J. Exptl. Physiol. 45: 113, 1960.
ANICHKOV,
Pharmacology of the carotid body chemoreceptors. Oxford, Pergamon, 1963, p. 67.
stereotaxis brain research on the conscious
rabbit. Amsterdam, Elsevier, 1961.
28.
GUYTON, A. C , Ross, T. M., CARRIER, O., JR.,
AND WALKER, T. R.: Evidence for tissue
oxygen demand as the major factor causing
autoregulation. Circulation Res. 15 (suppl. I ) :
1-60, 1964.
33. JOHNSON, P. C : Autoregulatory responses of cat
mesenteric arterioles measured in vivo. Circulation Res. 22: 199, 1968.
KORNER, P. I., UTHER, J. B., AND WHITE, S. W.:
Circulatory effects of chlorlose-urethane and
sodium pentobarbitone anesthesia in the rabbit.
J. Physiol. (London) 199: 253, 1968.
27.
32.
Dynamic
characteristics of the cardiovascular autonomic
effects during severe arterial hypoxia in the
unanesthetized rabbit. Circulation Res. 24:
671, 1969.
24. KORNER, P. I.: Effect of section of the carotid
sinus and aortic nerves on the cardiac output
of the rabbit. J. Physiol. (London) 180: 266,
1965.
25. WHITE, S. W.: Adrenalectomy in the rabbit.
Australian J. Exptl. Biol. Med. Sci. 44: 447,
1966.
26.
anesthesia on the local and autonomic components of the circulatory response to arterial
hypoxia. J. Physiol. (London) 199: 283,
1968.
CHALMERS, J. P., ISBISTEH, J. P., KORNER, P. I.,
42.
DALY, M. DE B., AND ROBINSON, B. H.: Analysis
of the reflex systemic vasodilator response
elicited by lung-inflation in the dog. J. Physiol.
(London) 195: 387, 1968.
43. HOKFELT, B.: Noradrenaline and adrenaline in
mammalian tissue. Acta Physiol. Scand. 25
(suppl. 92): 1, 1951.
44.
FERGUSON, R. W., FOLKOW, B., MITTS, M. G.,
AND HOFF, E. C : Effect of cortical stimulation
upon epinephrine activity. J. Neurophysiol.
20: 329, 1957.
45. ECCLES, J. C : Physiology of Synapses. Berlin,
Springer, 1964.
46. SCOTT, M. J.: Reflex effects of carotid body
KORNER, UTHER, WHITE
776
chemoreceptor stimulation on the heart rate of
the rabbit. Australian J. Exptl. Biol. Med. Sci.
44: 393, 1966.
47.
BAILEY, P., AND BREMER, F.: A sensory cortical
representation of the vagus nerve. J. Neurophysiol. 1: 405, 1938.
48. SCHOLANBER, P. F.: Animals in aquatic environments: Diving mammals and birds. In Hand
book of Physiology, sec. 4, Adaptation to the
Environment, edited by D. B. Dill, E. F.
Adolph, and C. G. Wilbur. Washington, D. C ,
American Physiological Society, 1964, p.
729.
49. ANDERSEN, H. T.: Factors determining circulatory
adjustments to diving: II. Asphyxia. Acta
Physiol. Scand. 58: 186, 1963.
Downloaded from http://circres.ahajournals.org/ by guest on June 18, 2017
50.
HOIAENBERG, N. K., AND UVNAS, B.: Role of the
cardiovascular response in the resistance to
asphyxia in avian divers. Acta Physiol. Scand.
58: 150, 1963.
51. KELLOCG, R. H.: Central chemical regulation of
respiration. In Handbook of Physiology, sec. 3,
vol. 1, Respiration, edited by W. O. Fenn and
H. Rahn. Washington, D. C , American Physiological Society, 1964, p. 507.
52.
COMBOE, J. H., JR.: Peripheral chemoreceptors.
In Handbook of Physiology, sec. 3, vol. 1,
Respiration, edited by W. O. Fenn and H.
Rahn. Washington, D. C , American Physiological Society, 1964, p. 557.
53.
DALY, M. DE B., AND SCOTT, M. J.: Analysis of
the primary cardiovascular reflex effects of
stimulation of the carotid body chemoreceptors
of the dog. J. Physiol. (London) 162: 555,
1962.
Circulation Research, Vol. XXIV, June 1969
Central Nervous Integration of the Circulatory and Respiratory Responses to Arterial
Hypoxemia in the Rabbit
PAUL I. KORNER, John B. Uther and Saxon W. White
Downloaded from http://circres.ahajournals.org/ by guest on June 18, 2017
Circ Res. 1969;24:757-776
doi: 10.1161/01.RES.24.6.757
Circulation Research is published by the American Heart Association, 7272 Greenville Avenue, Dallas, TX 75231
Copyright © 1969 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/24/6/757
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/