JOURNAL
OF APPLIED
Vol. 40, No. 1, January
PHYSIOLOGY
1976.
Printed
in U.S.A.
Central neural stimulation of respiration in unanesthetized decerebrate cats FREDERIC L. ELDRIDGE Departments of Medicine and Physiology, University of North School of Medicine, Chapel Hill, North Carolina 27514
Carolina
ELDRIDGE, FREDERIC L. Cent& neural stimuZation of respiration in unanesthetized decerebrate cats. J. Appl. Physiol. 40( 1): 23-28. 1976.-A previously reported central neural respiratory control process was restudied in unanesthetized decerebrate cats during spontaneous breathing, and during conditions of constant chemical stimulation where phrenic nerve activity was used to quantitate respiratory output. Respiration was increased by carotid sinus nerve stimulation. The pattern of respiration was examined at the cessation of such stimulation. In spontaneously breathing animals, active hyperventilation (HV) was followed by hyperpaea for up to 30 s and never by apnea. Passive HV was always followed by apnea. In animals with controlled chemical conditions, the transient at the end of stimulation consisted of two components, the first an immediate decrease in respiratory output and the second a slow decrease with a period of over 5 m. It is suggested that a facilitatory feedback process, probably located in the reticular activating system, maintains respiratory output for some time after cessation of a stimulus. This study duplicates the results of previous studies and shows that no area of the brain above the pons is required for the mechanism’s operation.
the total, was associated with a process which decayed exponentially over a period of 3-5 min. -Merent inputs from the carotid bodies, vagi, or thorax were not causal, CO, levels did not sign&an tlY affect the function, and circulatory changes did not explain the findings. It was concluded that the neural mechanism involved was located in the brain and that it probably acted through the medium of neural positivefeedback circuits which caused an afterdischarge of the respiratory control system. It was suggested that these circuits could be in the respiratory control system itself or in the reticular activ eating system, but h.igher centers of the brain were not ruled out. The present study which repeats both types of experiments noted above was carried out in unanesthetized decerebrate cats and shows
regulation of respiration; neural control; nonchemical respiratory stimuli; neural feedback; afterdischarge; open-loop technique; hyperventilation; poststimulation respiration; reticular activating system
CATS, following hyperventilation (HV) induced by carotid sinus nerve (CSN) stimulation or calf squeezing, do not develop post-HV apnea despite the development of significant hypocapnia; they instead exhibit post-HV hyperpnea which declines on a regular pattern during the first minute of recovery (2). Awake human beings show similar post-HV breathing patterns after voluntary hyperventilation (8). The findings have suggested that a neural process associated with active, i.e., neurally generated, breathing maintains respiration at an increased level for some time after cessation of a disturbing stimulus. The process was further documented by studies in anesthetized, paralyzed cats where chemical stimulation was kept constant by means of a constant-volume ventilator and the respiratory output monitored by means of integrated phrenic nerve activity (3). At the cessation of several kinds of respiratory stimulation,
Healthy, adult cats were first anes thetized with ether. The trachea was cannulated and both common carotid arteries ligated. A craniotomy was performed and midcollicular decerebration accomplished by means of suction. A femoral artery was cannulated for measurement of arterial pressure and for obtaining samples for blood gas and pH analysis. Continuous sampling of airway Pco2 was accomplished through a catheter placed in the airway and analysis by an infrared CO, analyzer (Beckman LB-2). Temperature was monitored by a rectal thermistor and maintained at 37-38OC by means of a heating lamp. Following these preparations both vagosympathetic trunks were isolated in the neck and surrounded by loose ties for later sectioning. Further dissection exposed both carotid sinus nerves, one of which was cut; the other was crushed distally and placed on bipolar platinum electrodes in a pool of mineral oil for electrical stimulation of afferents to the brain stem. Two types of studies were performed. In all cases, at least 2 h were allowed to elapse between the last administration of ether and the beginning of any experimental run. Spontaneous breathing. A group of seven cats was
there was an immediate decrease in respiratory output
studied during spontaneous respiration.
but not to prestimulation levels. The remainder of the decrease, which accounted for approximately 50-70% of
was measured by means of a miniature Krogh spirometer, from which CO, was absorbed by soda lime and 0,
that areas of the brain above the pons are not necessary
for the function mechanism.
of this important
neural
respiratory
METHODS
ANESTHETIZED
Tidal volume
23
Downloaded from www.physiology.org/journal/jappl at Lunds Univ Medicinska Fak Biblio (130.235.066.010) on February 14, 2019.
24
F. L. ELDRIDGE
replaced to keep the volu .me consta nt. The experiments were carried ou .t under hyperoxic conditions. After a
recording of stable control conditions, the animal was caused to hyperventilate by electrical stimulation of the CSN at a frequency of 25 Hz. The duration of active HV in the 32 experimental runs was 43 s (t 17 SD). After cessation of the stimulus, respiration was followed for at least 3.5 min. Tidal volume, airway Pcoz, arterial pressure, and a
stimulus marker were observed on an Gcilloscope and recorded on magnetic tape for later processing. Analysis was focu sed on the first sev*era1 posthyperventilation minutes. Breath-by-breath minute ventilation was calculated from the tidal volume and respiratory rate. These values, end-tidal Pco2, and arterial pressure were averaged for the control period and during the final 15 20 s of active HV. The first poststimulus recovery breath was calculated separately; the remainder of the recovery period was divided into segments of 10 s for the first 0.5 min, 15 s for up to 2 min, 20 s to 3 min, and 30 s for times over 3 min. Experimental
runs were performed both before and
after bila teral vagotomies. In several cats, hyperventilation. was also passively induced by means of a mechanical ventilator and the post-HV period recorded; this procedure was, however, not systematically compared with active HV as it had been in the previous report (2). Conditions of constant chemical stimulation. A group of eight cats was studied under conditions which allowed stable blood gas and pH levels during an experimental run. In nine runs, large openings were made in the chest walls and bilateral pneumothoraces induced to keep respi ratory muscle contrac tion from affecting ventilation. In 17 runs, in addition to the pneumothoraces, gallamine triethiodide-indu .ced paralysis was used to open the chemical feedback loop. In two runs succinylcholine was used as the paralyzing agent. In all cases, ventilation was maintained by means of a volume respirator and the end-tidal Pcoz monitored to assure constaticy. The peak rate of rectified integrated phrenic nerve activity was used to quantitate the neural respiratory output in the manner previously described (3, 4).
respiration was calculated separately; the remainder of the recovery period was divided into segments similar to those used in the spontaneous breathing experiments. More complete details of the equipment and procedures are given in the previous reports on spontaneously breathing cats (2) and paralyzed cats (3) whose brains were intact. RESULTS
Spontaneous breathing. Stimulation of the carotid sinus nerve led to an increase in ventilation and concomitant decrease in the end-tidal Pco,. At the end of the period of stimulation ventilation fell with the first recovery breath, but never to control level; it then decreased slowly over the next 15-20 s to or below control (Fig. IA). Apnea never occurred after active HV. On the other hand, passive HV which was associated with the same or less decrease in PcoZ always led to a period of post-HV apnea, as shown in the example in Fig. U3. The post-HV hyperpnea was consistently present after active HV whether the vagus nerves were intact (Fig. 2A) or cut (Fig. 12B). Control, hyperventilation and recovery data are shown in Fig. 3. Figure 3-A includes 25 runs in seven cats with intact vagi. Control Pco, was 29.5 mmHg and fell to 17.4 mmHg during hyperventilation. The postHV hyperpnea was due to increases in both respiratory frequency and tidal volume. Figure 3B includes seven runs in four cats with cut vagi, Control PcoZ was 30.3 mmHg and fell to 21.3 mmHg during hyperventilation. All of the post-HV hyperpnea was due to an increase in tidal volume, for respiratory rate fell below control during HV and early recovery. The period of post-HV hy-
After the recording of stable control conditions, in-
creased respiratory output was induced by CSN stimulation as before, but for 100 s. Following withdrawal of the stimulus recovery was recorded for at 1.east 5 m, but in some cases still 1.onger periods w ‘ere required before return of respiratory output to control levels. Eleven runs were performed with the vagi intact, 17 with the vagi cut. In 12 runs air was breathed, while in 16 hyperoxic conditions obtained. Phrenic activity, airway PcoZ, arterial pressure, and the stimulus marker were recorded on magnetic tape for subsequent breath-by-breath analysis. The neural respiratory (tidal) output for each breath was determined from the peak 0.1-s rate of phrenic nerve activity. Neural minute output was calculated as the product of this value and respiratory frequency. For purposes of comparing various animals, percentage responses of neural tidal and minute outputs were calculated as before (3), using a value of 100% for the mean respiratorv outnut during stimulation. The first no&stimulus
FIG. 1. Arterial pressure (API, tidal volume (VT), and airway Pco, during quiet breathing, hyperventilation, and recovery from hyperventilation in a spontaneously breathing cat with intact vagi. A: active hyperventilation induced by CSN stimulation of 40 s duration. Despite marked hypocapnia hyperpnea is present during the first 15 s of recovery and apnea does not occur. 3; passive hyperventilation of 62 s duration. Tidal volume was not recorded during HV. Although hypocapnia is less severe than inpaneE A, there is apnea for the first 30 s of the recovery period. (Note that the hyperventilation periods have been arranged to show onlv the first and last 20 9.1
Downloaded from www.physiology.org/journal/jappl at Lunds Univ Medicinska Fak Biblio (130.235.066.010) on February 14, 2019.
NEURAL
STIMULATION
25
OF RESPIRATION
change the inspiratory and Fig. 2).
perpnea was longer in this group than in the animals with intact vagi, but it should be noted that the degree of hypocapnia was also less. In some of the cats with intact vagi, deep sighing breaths occurred at regular intervals despite both hypocapnia and bilaterally cut carotid sinus nerves. Some of these sighs occurred during the early recovery period after cessation of stimulation; a striking finding was thatthe very deep inspiration did not significantly inter-
recovery pattern
(cf., Fig. lA
Conditions of constant chemical stimulation. Following an increase in respiratory output due to CSN stimulation, the transient at the end of stimulation consisted of two components, one an immediate (first-breath) de-
crease in respiratory output and the second a slow decrease that occurred over a period of several minutes.
fere with the slowly declining recovery pattern of respi-
Examples of these findings are shown in Fig. 5 in a single unanesthetized decerebrate cat with intact vagi
ration. In the example in Fig. 4, the large sigh is 3 times the magnitude of adjacent breaths. While it does appear to have some effect on end-expiratory level, it does not
and not paralyzed (Fig. SA), with intact vagi but paralyzed with gallamine (Fig. 5.B), and with vagi cut and paralyzed with gallamine (Fig. 5C).
FIG. 2. Two experimental runs in one spontaneously breathing cat, with intact vagi (A) and cut vagi (B). Both show posthyperventilation hyperpnea and no apnea. Symbols are the same as in Fig. 1.
QUIET lACTiVE 1 do 45 6@0 SEC HV I’ RECOVERY FIG. 4. Control, hyperventilation produced by carotid sinus nerve stimulation, and recovery in cat with intact vagi. This cat had regularly occurring deep sighing inspirations, one of which occurred at the third breath (*) of recovery. Note that while there is some effect on end-expiratory level, the smoothly declining inspiratory pattern is not significantly affected by the sigh. AP is arterial pressure.
‘
arr -=-I68 =mx pz %E b? OE I-E pp%Lla 8 s-g
‘-.. El40 %I2
I
I
I
I$;
30-
$ II
20‘
FIG. 3. A; findings in 25 experimental runs in 7 spontaneously breathing cats with intact vagi during control, hyperventilation produced by carotid sinus nerve stimulation, and recovery. Values are means + SE. (When SE brackets are not present, SE is smaller than size of point symbol.) B: findings in 7 experimental runs in 4 spontaneously breathing vagotomized cats during control, hyperventilation produced by carotid sinus nerve stimulation, and recovery. Values are means + SE.
20 IO 120
25~~~~~ 100 80
60 40 l600-
5 G -I
I
1
I5
~
I
1400 -
p .c 1200 $E ‘T r 2
r
IOOO-
800-
1 j I I
L CONTROL
HV
40 RECOVERY,
80 120 SECONDS
160
200
CONTROL
HV
40
RECOVERY,
80
I20
160
200
SECONDS
Downloaded from www.physiology.org/journal/jappl at Lunds Univ Medicinska Fak Biblio (130.235.066.010) on February 14, 2019.
F. L. ELDRIDGE
26
STIMULATION
CONTROL+
1 ,STlMULATlON
{RECOVERY
160sec
I240
set
1360
Downloaded from www.physiology.org/journal/jappl at Lunds Univ Medicinska Fak Biblio (130.235.066.010) on February 14, 2019.
NEURAL
STIMULATION
OF
27
RESPIRATION
_
rLW
=
~
-
t-+ !-
1 Control
“HV”
I ._-L--L40
80
I 120
I 160
I 200
I 240
1 280 Control RECOV ‘ERY, seconds
OF APPLIED
Vol. 40, No. 1, January
PHYSIOLOGY
1976.
Printed
in U.S.A.
Central neural stimulation of respiration in unanesthetized decerebrate cats FREDERIC L. ELDRIDGE Departments of Medicine and Physiology, University of North School of Medicine, Chapel Hill, North Carolina 27514
Carolina
ELDRIDGE, FREDERIC L. Cent& neural stimuZation of respiration in unanesthetized decerebrate cats. J. Appl. Physiol. 40( 1): 23-28. 1976.-A previously reported central neural respiratory control process was restudied in unanesthetized decerebrate cats during spontaneous breathing, and during conditions of constant chemical stimulation where phrenic nerve activity was used to quantitate respiratory output. Respiration was increased by carotid sinus nerve stimulation. The pattern of respiration was examined at the cessation of such stimulation. In spontaneously breathing animals, active hyperventilation (HV) was followed by hyperpaea for up to 30 s and never by apnea. Passive HV was always followed by apnea. In animals with controlled chemical conditions, the transient at the end of stimulation consisted of two components, the first an immediate decrease in respiratory output and the second a slow decrease with a period of over 5 m. It is suggested that a facilitatory feedback process, probably located in the reticular activating system, maintains respiratory output for some time after cessation of a stimulus. This study duplicates the results of previous studies and shows that no area of the brain above the pons is required for the mechanism’s operation.
the total, was associated with a process which decayed exponentially over a period of 3-5 min. -Merent inputs from the carotid bodies, vagi, or thorax were not causal, CO, levels did not sign&an tlY affect the function, and circulatory changes did not explain the findings. It was concluded that the neural mechanism involved was located in the brain and that it probably acted through the medium of neural positivefeedback circuits which caused an afterdischarge of the respiratory control system. It was suggested that these circuits could be in the respiratory control system itself or in the reticular activ eating system, but h.igher centers of the brain were not ruled out. The present study which repeats both types of experiments noted above was carried out in unanesthetized decerebrate cats and shows
regulation of respiration; neural control; nonchemical respiratory stimuli; neural feedback; afterdischarge; open-loop technique; hyperventilation; poststimulation respiration; reticular activating system
CATS, following hyperventilation (HV) induced by carotid sinus nerve (CSN) stimulation or calf squeezing, do not develop post-HV apnea despite the development of significant hypocapnia; they instead exhibit post-HV hyperpnea which declines on a regular pattern during the first minute of recovery (2). Awake human beings show similar post-HV breathing patterns after voluntary hyperventilation (8). The findings have suggested that a neural process associated with active, i.e., neurally generated, breathing maintains respiration at an increased level for some time after cessation of a disturbing stimulus. The process was further documented by studies in anesthetized, paralyzed cats where chemical stimulation was kept constant by means of a constant-volume ventilator and the respiratory output monitored by means of integrated phrenic nerve activity (3). At the cessation of several kinds of respiratory stimulation,
Healthy, adult cats were first anes thetized with ether. The trachea was cannulated and both common carotid arteries ligated. A craniotomy was performed and midcollicular decerebration accomplished by means of suction. A femoral artery was cannulated for measurement of arterial pressure and for obtaining samples for blood gas and pH analysis. Continuous sampling of airway Pco2 was accomplished through a catheter placed in the airway and analysis by an infrared CO, analyzer (Beckman LB-2). Temperature was monitored by a rectal thermistor and maintained at 37-38OC by means of a heating lamp. Following these preparations both vagosympathetic trunks were isolated in the neck and surrounded by loose ties for later sectioning. Further dissection exposed both carotid sinus nerves, one of which was cut; the other was crushed distally and placed on bipolar platinum electrodes in a pool of mineral oil for electrical stimulation of afferents to the brain stem. Two types of studies were performed. In all cases, at least 2 h were allowed to elapse between the last administration of ether and the beginning of any experimental run. Spontaneous breathing. A group of seven cats was
there was an immediate decrease in respiratory output
studied during spontaneous respiration.
but not to prestimulation levels. The remainder of the decrease, which accounted for approximately 50-70% of
was measured by means of a miniature Krogh spirometer, from which CO, was absorbed by soda lime and 0,
that areas of the brain above the pons are not necessary
for the function mechanism.
of this important
neural
respiratory
METHODS
ANESTHETIZED
Tidal volume
23
Downloaded from www.physiology.org/journal/jappl at Lunds Univ Medicinska Fak Biblio (130.235.066.010) on February 14, 2019.
24
F. L. ELDRIDGE
replaced to keep the volu .me consta nt. The experiments were carried ou .t under hyperoxic conditions. After a
recording of stable control conditions, the animal was caused to hyperventilate by electrical stimulation of the CSN at a frequency of 25 Hz. The duration of active HV in the 32 experimental runs was 43 s (t 17 SD). After cessation of the stimulus, respiration was followed for at least 3.5 min. Tidal volume, airway Pcoz, arterial pressure, and a
stimulus marker were observed on an Gcilloscope and recorded on magnetic tape for later processing. Analysis was focu sed on the first sev*era1 posthyperventilation minutes. Breath-by-breath minute ventilation was calculated from the tidal volume and respiratory rate. These values, end-tidal Pco2, and arterial pressure were averaged for the control period and during the final 15 20 s of active HV. The first poststimulus recovery breath was calculated separately; the remainder of the recovery period was divided into segments of 10 s for the first 0.5 min, 15 s for up to 2 min, 20 s to 3 min, and 30 s for times over 3 min. Experimental
runs were performed both before and
after bila teral vagotomies. In several cats, hyperventilation. was also passively induced by means of a mechanical ventilator and the post-HV period recorded; this procedure was, however, not systematically compared with active HV as it had been in the previous report (2). Conditions of constant chemical stimulation. A group of eight cats was studied under conditions which allowed stable blood gas and pH levels during an experimental run. In nine runs, large openings were made in the chest walls and bilateral pneumothoraces induced to keep respi ratory muscle contrac tion from affecting ventilation. In 17 runs, in addition to the pneumothoraces, gallamine triethiodide-indu .ced paralysis was used to open the chemical feedback loop. In two runs succinylcholine was used as the paralyzing agent. In all cases, ventilation was maintained by means of a volume respirator and the end-tidal Pcoz monitored to assure constaticy. The peak rate of rectified integrated phrenic nerve activity was used to quantitate the neural respiratory output in the manner previously described (3, 4).
respiration was calculated separately; the remainder of the recovery period was divided into segments similar to those used in the spontaneous breathing experiments. More complete details of the equipment and procedures are given in the previous reports on spontaneously breathing cats (2) and paralyzed cats (3) whose brains were intact. RESULTS
Spontaneous breathing. Stimulation of the carotid sinus nerve led to an increase in ventilation and concomitant decrease in the end-tidal Pco,. At the end of the period of stimulation ventilation fell with the first recovery breath, but never to control level; it then decreased slowly over the next 15-20 s to or below control (Fig. IA). Apnea never occurred after active HV. On the other hand, passive HV which was associated with the same or less decrease in PcoZ always led to a period of post-HV apnea, as shown in the example in Fig. U3. The post-HV hyperpnea was consistently present after active HV whether the vagus nerves were intact (Fig. 2A) or cut (Fig. 12B). Control, hyperventilation and recovery data are shown in Fig. 3. Figure 3-A includes 25 runs in seven cats with intact vagi. Control Pco, was 29.5 mmHg and fell to 17.4 mmHg during hyperventilation. The postHV hyperpnea was due to increases in both respiratory frequency and tidal volume. Figure 3B includes seven runs in four cats with cut vagi, Control PcoZ was 30.3 mmHg and fell to 21.3 mmHg during hyperventilation. All of the post-HV hyperpnea was due to an increase in tidal volume, for respiratory rate fell below control during HV and early recovery. The period of post-HV hy-
After the recording of stable control conditions, in-
creased respiratory output was induced by CSN stimulation as before, but for 100 s. Following withdrawal of the stimulus recovery was recorded for at 1.east 5 m, but in some cases still 1.onger periods w ‘ere required before return of respiratory output to control levels. Eleven runs were performed with the vagi intact, 17 with the vagi cut. In 12 runs air was breathed, while in 16 hyperoxic conditions obtained. Phrenic activity, airway PcoZ, arterial pressure, and the stimulus marker were recorded on magnetic tape for subsequent breath-by-breath analysis. The neural respiratory (tidal) output for each breath was determined from the peak 0.1-s rate of phrenic nerve activity. Neural minute output was calculated as the product of this value and respiratory frequency. For purposes of comparing various animals, percentage responses of neural tidal and minute outputs were calculated as before (3), using a value of 100% for the mean respiratorv outnut during stimulation. The first no&stimulus
FIG. 1. Arterial pressure (API, tidal volume (VT), and airway Pco, during quiet breathing, hyperventilation, and recovery from hyperventilation in a spontaneously breathing cat with intact vagi. A: active hyperventilation induced by CSN stimulation of 40 s duration. Despite marked hypocapnia hyperpnea is present during the first 15 s of recovery and apnea does not occur. 3; passive hyperventilation of 62 s duration. Tidal volume was not recorded during HV. Although hypocapnia is less severe than inpaneE A, there is apnea for the first 30 s of the recovery period. (Note that the hyperventilation periods have been arranged to show onlv the first and last 20 9.1
Downloaded from www.physiology.org/journal/jappl at Lunds Univ Medicinska Fak Biblio (130.235.066.010) on February 14, 2019.
NEURAL
STIMULATION
25
OF RESPIRATION
change the inspiratory and Fig. 2).
perpnea was longer in this group than in the animals with intact vagi, but it should be noted that the degree of hypocapnia was also less. In some of the cats with intact vagi, deep sighing breaths occurred at regular intervals despite both hypocapnia and bilaterally cut carotid sinus nerves. Some of these sighs occurred during the early recovery period after cessation of stimulation; a striking finding was thatthe very deep inspiration did not significantly inter-
recovery pattern
(cf., Fig. lA
Conditions of constant chemical stimulation. Following an increase in respiratory output due to CSN stimulation, the transient at the end of stimulation consisted of two components, one an immediate (first-breath) de-
crease in respiratory output and the second a slow decrease that occurred over a period of several minutes.
fere with the slowly declining recovery pattern of respi-
Examples of these findings are shown in Fig. 5 in a single unanesthetized decerebrate cat with intact vagi
ration. In the example in Fig. 4, the large sigh is 3 times the magnitude of adjacent breaths. While it does appear to have some effect on end-expiratory level, it does not
and not paralyzed (Fig. SA), with intact vagi but paralyzed with gallamine (Fig. 5.B), and with vagi cut and paralyzed with gallamine (Fig. 5C).
FIG. 2. Two experimental runs in one spontaneously breathing cat, with intact vagi (A) and cut vagi (B). Both show posthyperventilation hyperpnea and no apnea. Symbols are the same as in Fig. 1.
QUIET lACTiVE 1 do 45 6@0 SEC HV I’ RECOVERY FIG. 4. Control, hyperventilation produced by carotid sinus nerve stimulation, and recovery in cat with intact vagi. This cat had regularly occurring deep sighing inspirations, one of which occurred at the third breath (*) of recovery. Note that while there is some effect on end-expiratory level, the smoothly declining inspiratory pattern is not significantly affected by the sigh. AP is arterial pressure.
‘
arr -=-I68 =mx pz %E b? OE I-E pp%Lla 8 s-g
‘-.. El40 %I2
I
I
I
I$;
30-
$ II
20‘
FIG. 3. A; findings in 25 experimental runs in 7 spontaneously breathing cats with intact vagi during control, hyperventilation produced by carotid sinus nerve stimulation, and recovery. Values are means + SE. (When SE brackets are not present, SE is smaller than size of point symbol.) B: findings in 7 experimental runs in 4 spontaneously breathing vagotomized cats during control, hyperventilation produced by carotid sinus nerve stimulation, and recovery. Values are means + SE.
20 IO 120
25~~~~~ 100 80
60 40 l600-
5 G -I
I
1
I5
~
I
1400 -
p .c 1200 $E ‘T r 2
r
IOOO-
800-
1 j I I
L CONTROL
HV
40 RECOVERY,
80 120 SECONDS
160
200
CONTROL
HV
40
RECOVERY,
80
I20
160
200
SECONDS
Downloaded from www.physiology.org/journal/jappl at Lunds Univ Medicinska Fak Biblio (130.235.066.010) on February 14, 2019.
F. L. ELDRIDGE
26
STIMULATION
CONTROL+
1 ,STlMULATlON
{RECOVERY
160sec
I240
set
1360
Downloaded from www.physiology.org/journal/jappl at Lunds Univ Medicinska Fak Biblio (130.235.066.010) on February 14, 2019.
NEURAL
STIMULATION
OF
27
RESPIRATION
_
rLW
=
~
-
t-+ !-
1 Control
“HV”
I ._-L--L40
80
I 120
I 160
I 200
I 240
1 280 Control RECOV ‘ERY, seconds
