Carbon dioxide effects on the ventilatory response to sustained hypoxia

We examined the interrelation between CO2 and the ventilatory response to moderate (80% arterial saturation) sustained hypoxia in normal young adults. On a background of continuous CO2-stimulated hyperventilation, hypoxia was introduced and sustained for 25 min. Initially, with the introduction of hypoxia onto hypercapnia, there was a brisk additional increase in inspiratory minute ventilation (VI) to 284% of resting VI, but the response was not sustained and hypoxic VI declined by 36% to a level intermediate between the initial increase and the preexisting hypercapnic hyperventilation. Through the continuous hypercapnia, the changes in hypoxic ventilation resulted from significant alterations in tidal volume (VT) and mean inspiratory flow (VT/TI) without changes in respiratory timing. In another experiment, sustained hypoxia was introduced on the usual background of room air, either with isocapnia or without maintenance of end-tidal CO2 (ETCO2) (poikilocapnic hypoxia). Regardless of the degree of maintenance of ETCO2, during 25 min of sustained hypoxia, VI showed an initial brisk increase and then declined by 35-40% of resting VI to a level intermediate between the initial response and resting room air VI. For both isocapnia and poikilocapnic conditions, the attenuation of VI was an expression of a diminished VT. Thus the decline in ventilation with sustained hypoxia occurred regardless of the background ETCO2, suggesting that the mechanism underlying the hypoxic decline is independent of CO2.     

Recovery of the ventilatory response to hypoxia in normal adults

Recovery of the initial ventilatory response to hypoxia was examined after the ventilatory response had declined during sustained hypoxia. Normal young adults were exposed to two consecutive 25-min periods of sustained isocapnic hypoxia (80% O2 saturation in arterial blood), separated by varying interludes of room air breathing or an increased inspired O2 fraction (FIO2). The decline in the hypoxic ventilatory response during the 1st 25 min of hypoxia was not restored after a 7-min interlude of room air breathing; inspired ventilation (VI) at the end of the first hypoxic period was not different from VI at the beginning and end of the second hypoxic period. After a 15-min interlude of room air breathing, the hypoxic ventilatory response had begun to recover. With a 60-min interlude of room air breathing, recovery was complete; VI during the second hypoxic exposure matched VI during the first hypoxic period. Ventilatory recovery was accelerated by breathing supplemental O2. With a 15-min interlude of 0.3 FIO2 or 7 min of 1.0 FIO2, VI of the first and second hypoxic periods were equivalent. Both the decline and recovery of the hypoxic ventilatory response were related to alterations in tidal volume and mean inspiratory flow (VT/TI), with little alteration in respiratory timing. We conclude that the mechanism of the decline in the ventilatory response with sustained hypoxia may require up to 1 h for complete reversal and that the restoration is O2 sensitive.     

Increased chemoreceptor output and ventilatory response to sustained hypoxia

In adult humans the ventilatory response to sustained hypoxia (VRSH) is biphasic, characterized by an initial brisk increase, due to peripheral chemoreceptor (PC) stimulation, followed by a decline attributed to central depressant action of hypoxia. To study the effects of selective stimulation of PC on the ventilatory response pattern to hypoxia, the VRSH was evaluated after pretreatment with almitrine (A), a PC stimulant. Eight subjects were pretreated with A (75 mg po) or placebo (P) on 2 days in a single-blind manner. Two hours after drug administration, they breathed, in succession, room air (10 min), O2 (5 min), room air (5 min), hypoxia [25 min, arterial O2 saturation (SaO2) = 80%], O2 (5 min), and room air (5 min). End-tidal CO2 was kept constant at the normoxic base-line values. Inspiratory minute ventilation (VI) and breathing patterns were measured over the last 2 min of each period and during minutes 3-5 of hypoxia, and nadirs in VI were assessed just before and after O2 exposure. Independent of the day, the VRSH was biphasic. With P and A pretreatment, early hypoxia increased VI 4.6 +/- 1 and 14.2 +/- 1 (SE) l/min, respectively, from values obtained during the preceding room-air period. On A day the hypoxic ventilatory decline was significantly larger than that on P day, and on both days the decline was a constant fraction of the acute hypoxic response.(ABSTRACT TRUNCATED AT 250 WORDS)     

Breathing during sleep with mild hypoxia

To investigate ventilatory response to mild hypoxia during non-rapid-eye-movement sleep, we administered approximately 16% O2 (which corresponds to concentrations found in commercial high altitude air craft) to 12 normal subjects by using a Venturi mask, which did not alter the breathing pattern during this study. Under mild hypoxia, inspiratory minute ventilation during sleep showed an initial rapid increase (P less than 0.001) but then declined significantly (P less than 0.001) and stabilized. Stable levels differed among individuals and, compared with those measured before hypoxia, were significantly lower in some subjects, higher in one, and essentially unchanged in the others. The initial rapid increase in minute ventilation after mild hypoxia during sleep correlated with the respective values of hypoxic ventilatory response during the awake state (P less than 0.01), but the final lowered levels did not. We conclude that the ventilatory response after mild hypoxia during sleep is biphasic and hypoxic depression exerts considerable influence on ventilation under mild hypoxia during sleep. So we should take hypoxic depression into consideration to evaluate the response to hypoxia during sleep.     

Biphasic ventilatory response to hypoxia in unanesthetized rats

To determine the role of postinspiratory inspiratory activity of the diaphragm in the biphasic ventilatory response to hypoxia in unanesthetized rats, we examined diaphragmatic activity at its peak (DI), at the end of expiration (DE), and ventilation in adult unanesthetized rats during poikilocapnic hypoxia (10 % O2) sustained for 20 min. Hypoxia induced an initial increase in ventilation followed by a consistent decline. Tidal volume (VT), frequency of breathing (fR), DI and DE at first increased, then VT and DE decreased, while fR and DI remained enhanced. Phasic activation of the diaphragm (DI-DE) increased significantly at 10, 15 and 20 min of hypoxia. These results indicate that 1) the ventilatory response of unanesthetized rats to sustained hypoxia has a typical biphasic character and 2) the increased end-expiratory activity of the diaphragm limits its phasic inspiratory activation, but this increase cannot explain the secondary decline in tidal volume and ventilation.     

Aminophylline effects on ventilatory response to hypoxia and hyperoxia in normal adults

In 10 normal young adults, ventilation was evaluated with and without pretreatment with aminophylline, an adenosine blocker, while they breathed pure O2 1) after breathing room air and 2) after 25 min of isocapnic hypoxia (arterial O2 saturation 80%). With and without aminophylline, 5 min of hyperoxia significantly increased inspiratory minute ventilation (VI) from the normoxic base line. In control experiments, with hypoxia, VI initially increased and then declined to levels that were slightly above the normoxic base line. Pretreatment with aminophylline significantly attenuated the hypoxic ventilatory decline. During transitions to pure O2 (cessation of carotid bodies' output), VI and breathing patterns were analyzed breath by breath with a moving-average technique, searching for nadirs before and after hyperoxia. On placebo days, at the end of hypoxia, hyperoxia produced nadirs that were significantly lower than those observed with room-air breathing and also significantly lower than when hyperoxia followed normoxia, averaging, respectively, 6.41 +/- 0.52, 8.07 +/- 0.32, and 8.04 +/- 0.39 (SE) l/min. This hypoxic depression was due to significant decrease in tidal volume and prolongation of expiratory time. Aminophylline partly prevented these alterations in breathing pattern; significant posthypoxic ventilatory depression was not observed. We conclude that aminophylline attenuated hypoxic central depression of ventilation, although it does not affect hyperoxic steady-state hyperventilation. Adenosine may play a modulatory role in hypoxic but not in hyperoxic ventilation.     

Effect of 100% O2 on hypoxic eucapnic ventilation

We measured ventilation in nine young adults while they breathed pure O2 after breathing room air and after 5 and 25 min of hypoxia. With isocapnic hypoxia (arterial O2 saturation 80 +/- 2%) mean ventilation increased at 5 min and then declined, so that at 25 min values did not differ from those on room air. After 3 min of O2 breathing, ventilation was greater than that on room air or after 25 min of isocapnic hypoxia, whether the hyperoxia had been preceded by hypoxia or normoxia. During transitions to pure O2 breathing, ventilation was analyzed breath by breath with a moving average technique, searching for nadirs before and after increases in PO2. After both 5 and 25 min of hypoxia, O2 breathing was associated with transient depressions of ventilation, which were greater after 25 min than after 5 min. Significant depressions were not observed when hyperoxia followed room air breathing, and O2-induced nadirs after hypoxia were lower than those observed during room air breathing. O2 transiently depressed ventilation after hypoxia but not after room air breathing. These results suggest that the normal ventilatory response to isocapnic hypoxia has two components, an excitatory one from peripheral chemoreceptors, which is turned off by O2 breathing, and a slower inhibitory one, probably of central origin, which is affected less promptly by O2 breathing.     

Ventilatory adjustments during sustained resistive unloading in exercising humans

Effect of He-O2-breathing (79.1%:20.9%) compared to air-breathing on inspiratory ventilation (VI) and its different components [tidal volume (VT), the duration of the phases of each respiratory cycle (tI, tTOT)] as well as on inspiratory mouth occlusion pressure (P0.1) were studied in six normal men at rest and during 72 constant-load exercises (90 W) over a much longer period than in previous studies. Results showed that, irrespective of the order of administration of the two gases (7 min air----7 min He-O2 or vice versa): at rest, P0.1 decreased during He-O2 inhalation but no changes in VI and breathing pattern were detectable; during exercise, sustained He-induced hyperventilation was observed without any change in the absolute value of P0.1; increase in P0.1 between the resting period and exercise (delta P0.1) was significantly higher during He-O2-breathing than during air breathing; this He-induced hyperventilation was associated with a sustained increase in VT/tI, but with constant tI/tTOT. Helium-breathing during exercise cannot be a simple situation of resistance unloading, as has been suggested. We conclude that He-O2-breathing, after the initial compensation period, induces reflex changes in ventilatory control with an increase in inspiratory neural drive. Moreover, it appears that exercise P0.1 is not a legitimate index of inspiratory neural drive whenever rest P0.1 changes according to the nature of the inhaled gas mixture.     

Time course of air hunger mirrors the biphasic ventilatory response to hypoxia.

Determining response dynamics of hypoxic air hunger may provide information of use in clinical practice and will improve understanding of basic dyspnea mechanisms. It is hypothesized that air hunger arises from projection of reflex brain stem ventilatory drive ("corollary discharge") to forebrain centers. If perceptual response dynamics are unmodified by events between brain stem and cortical awareness, this hypothesis predicts that air hunger will exactly track ventilatory response. Thus, during sustained hypoxia, initial increase in air hunger would be followed by a progressive decline reflecting biphasic reflex ventilatory drive. To test this prediction, we applied a sharp-onset 20-min step of normocapnic hypoxia and compared dynamic response characteristics of air hunger with that of ventilation in 10 healthy subjects. Air hunger was measured during mechanical ventilation (minute ventilation = 9 +/- 1.4 l/min; end-tidal Pco(2) = 37 +/- 2 Torr; end-tidal Po(2) = 45 +/- 7 Torr); ventilatory response was measured during separate free-breathing trials in the same subjects. Discomfort caused by "urge to breathe" was rated every 30 s on a visual analog scale. Both ventilatory and air hunger responses were modeled as delayed double exponentials corresponding to a simple linear first-order response but with a separate first-order adaptation. These models provided adequate fits to both ventilatory and air hunger data (r(2) = 0.88 and 0.66). Mean time constant and time-to-peak response for the average perceptual response (0.36 min(-1) and 3.3 min, respectively) closely matched corresponding values for the average ventilatory response (0.39 min(-1) and 3.1 min). Air hunger response to sustained hypoxia tracked ventilatory drive with a delay of approximately 30 s. Our data provide further support for the corollary discharge hypothesis for air hunger.     

Naloxone reduces decrease in ventilation induced by hypoxia in newborn infants

The mechanism responsible for the decrease in ventilation during breathing of low fractional concentration of inspired O2 in the newborn infant is poorly understood. The present study tested the hypothesis that endogenous opiates account for this ventilatory decrease. Eleven healthy newborn infants breathed 15% O2, balance N2 for 5 min following an injection of saline and following an injection of naloxone. Neither injection caused a change in minute ventilation (VE) or ventilatory pattern when the infants were breathing room air. However, the decreased ventilation during hypoxia following naloxone was significantly less than that following saline. VE dropped about 14% following saline but only about 4% following naloxone. However, the adult ventilatory response to hypoxemia, i.e., a relatively sustained increase in VE, was not attained. Naloxone had no influence on the occurrence of periodic breathing during hypoxemia. Thus in the healthy full-term newborn infant, endogenous opiates account only for a part of the decreased ventilation during hypoxemia.     

Ventilatory effects and interactions with change in PaO2 in awake goats.

We utilized selective carotid body (CB) perfusion while changing inspired O2 fraction in arterial isocapnia to characterize the non-CB chemoreceptor ventilatory response to changes in arterial PO2 (PaO2) in awake goats and to define the effect of varying levels of CB PO2 on this response. Systemic hyperoxia (PaO2 greater than 400 Torr) significantly increased inspired ventilation (VI) and tidal volume (VT) in goats during CB normoxia, and systemic hypoxia (PaO2 = 29 Torr) significantly increased VI and respiratory frequency in these goats. CB hypoxia (CB PO2 = 34 Torr) in systemic normoxia significantly increased VI, VT, and VT/TI; the ventilatory effects of CB hypoxia were not significantly altered by varying systemic PaO2. We conclude that ventilation is stimulated by systemic hypoxia and hyperoxia in CB normoxia and that this ventilatory response to changes in systemic O2 affects the CB O2 response in an additive manner.     

Relationship between lung volume and tracheal area as assessed by acoustic reflection

In nine normal subjects we measured the ventilatory response to isocapnic hypoxia with and without an intravenous infusion of 1 mg of somatostatin. Arterial O2 saturation was rapidly lowered to 80 +/- 2% in 2 min and maintained for 30 min. During control experiments, ventilation increased immediately (3-5 min) and then declined so that at 25 min of hypoxia ventilation was little above that in room air. Somatostatin was associated with a small decrease in ventilation while the subjects breathed room air. With hypoxia there was no immediate increase in ventilation for the group as a whole, although an increase was observed in one subject. With somatostatin, after 25 min of hypoxia, mean ventilation was lower than at any other time in the study; as hypoxia was discontinued ventilation increased slightly. Somatostatin causes profound depression of the ventilatory response to hypoxia by a mechanism that is not known but may be central. With somatostatin hypoxia of 25-min duration tends to depress ventilation.     

Effect of a dopamine antagonist on ventilation during sustained hypoxia in mice

To test the hypothesis that dopamine accumulated in the carotid body limits hyperventilation during acclimatization to sustained hypoxia, we administered the dopamine antagonist droperidol to mice undergoing acclimatization to an inspired O2 fraction (FIo2) of 0.1. Twelve mice were exposed to hypoxia for 10 days and ventilation in 10% O2 and in 7% CO2 in air were measured daily by a plethysmographic method. Under both conditions ventilation increased during acclimatization to hypoxia: ventilation in 10% O2 increased from 39.4 +/- 3.8 (mean +/- SE) ml/min before exposure to sustained hypoxia to 72.2 +/- 4.2 ml/min after 3 days of continuous hypoxia, and ventilation in 7% CO2 in air at the same time increased from 113.2 +/- 5.4 ml/min to 140.0 +/- 5.6 ml/min. Twelve mice were exposed to FIo2 of 0.1 for 10 days and received droperidol (300 micrograms/kg intraperitoneally) before exposure to sustained hypoxia and on the 2nd, 4th, and 8th days of continuous hypoxia. Before exposure to sustained hypoxia, droperidol increased ventilation in 10% O2 from 40.1 +/- 2.5 ml/min to 72.5 +/- 5.2 ml/min, but after 2, 4, and 8 days of continuous hypoxia droperidol caused an acute fall in ventilation (ventilation in 10% O2 after droperidol on day 2: 49.1 +/- 3.1 ml/min, on day 4: 44.4 +/- 3.7 ml/min, and on day 8: 27.8 +/- 3.4 ml/min). Two days after the animals were returned to room air, ventilation in 10% O2 again increased in response to droperidol. We conclude that dopamine in the carotid body does not limit ventilatory responses to hypoxia during acclimatization to sustained hypoxia.     

Interaction of hypoxic and hypercapnic stimuli on breathing pattern in the newborn rat

We aimed to investigate whether newborn rats respond to acute hypoxia with a biphasic pattern as other newborn species, the characteristics of their ventilatory response to hypercapnia, and the ventilatory response to combined hypoxic and hypercapnic stimuli. First, we established that newborn unanesthetized rats (2-4 days old) exposed to 10% O2 respond as other species. Their ventilation (VE), measured by flow plethysmography, immediately increased by 30%, then dropped and remained around normoxic values within 5 min. The drop was due to a decrease in tidal volume, while frequency remained elevated. Hence, alveolar ventilation was about 10% below normoxic value. At the same time O2 consumption, measured manometrically, dropped (-23%), possibly indicating a mechanism to protect vital organs. Ten percent CO2 in O2 breathing determined a substantial increase in VE (+47%), indicating that the respiratory pump is capable of a marked sustained hyperventilation. When CO2 was added to the hypoxic mixture, VE increased by about 85%, significantly more than without the concurrent hypoxic stimulus. Thus, even during the drop in VE of the biphasic response to hypoxia, the respiratory control system can respond with excitation to a further increase in chemical drive. Analysis of the breathing patterns suggests that in the newborn rat in hypoxia the inspiratory drive is decreased but the inspiratory on-switch mechanism is stimulated, hypercapnia increases ventilation mainly through an increase in respiratory drive, and moderate asphyxia induces the most powerful ventilatory response by combining the stimulatory action of hypercapnia and hypoxia.     

Hypoxic exposure and activation of the afterdischarge mechanism in conscious humans

After voluntary hyperventilation, normal humans do not develop a significant ventilatory depression despite low arterial CO2 tension, a phenomenon attributed to activation of a brain stem mechanism referred to as the "afterdischarge." Afterdischarge is one of the factors that promote ventilatory stability. It is not known whether physiological stimuli, such as hypoxia, are able to activate the afterdischarge in humans. To test this, breath-by-breath ventilation (VI) was measured in nine young adults during and immediately after a brief period (35-51 s) of acute hypoxia (end-tidal O2 tension 55 Torr). Hypoxia was terminated by switching to 100% O2 (end-tidal O2 tension of first posthypoxic breath greater than 100 Torr). Brief hypoxia increased VI and decreased end-tidal CO2 tension. In all subjects, termination of hypoxia was followed by a gradual ventilatory decay; hyperoxic VI remained higher than the normoxic baseline for several breaths and, despite the negative chemical stimulus of hyperoxia and hypocapnia, reached a new steady state without an apparent undershoot. We conclude that brief hypoxia is able to activate the afterdischarge mechanism in conscious humans. This contrasts sharply with the ventilatory undershoot that follows relief of sustained hypoxia, thereby suggesting that sustained hypoxia inactivates the afterdischarge mechanism. The present findings are of relevance to the pathogenesis of periodic breathing in a hypoxic environment. Furthermore, brief exposure to hypoxia might be useful for evaluation of the role of afterdischarge in other disorders associated with unstable breathing.     

Ventilatory response to sustained hypoxia after pretreatment with aminophylline

During sustained hypoxia the decline in ventilation that occurs in normal adult humans may be related to central accumulation of a neurochemical with net inhibitory effect. Recent investigations have shown that the putative neurotransmitter adenosine can effect a prolonged respiratory inhibition. Therefore we evaluated the possible role of adenosine in the hypoxia ventilatory decline by employing aminophylline as an adenosine blocker. We evaluated the ventilatory response to 25 min of sustained hypoxia (80% arterial O2 saturation), in eight young adults after pretreatment with either intravenous saline or aminophylline. With a mean serum aminophylline level of 15.7 mg/l, over 25 min of sustained hypoxia, peak hypoxic ventilation decreased by only 12.8% compared with 24.8% with saline, a significant difference. However, the ventilatory decline during sustained hypoxia was not abolished by the aminophylline pretreatment. Unlike the usual tidal volume-dependent attenuation of hypoxic ventilation exhibited after saline, after aminophylline the ventilatory decline was achieved predominantly through alterations in respiratory timing. Thus aminophylline pretreatment did alleviate the hypoxic ventilatory decline, although the associated alterations in breathing pattern were uncharacteristic. We conclude that adenosine may play a contributing role in the hypoxic ventilatory decline.     

Chemoreflex modulation of ventilatory dynamics during exercise in humans

The precision of arterial blood gas homeostasis following a change of work rate depends on the response kinetics of ventilation. The carotid bodies (CB's) have been proposed as modulators of these kinetics. The present investigation was undertaken to determine whether the effect is specific to CB activation or whether other factors that augment the exercise hyperpnea would produce a similar response. We therefore established the effects of increased CB and central (C) chemoreflex activation on the inspired ventilatory (VI) dynamics for moderate-intensity cycling. Work tests were separately performed with air, 12% O2 to increase CB activity, 100% O2 to "abolish" CB activity, and CO2 in O2 to increase C activity. The time constant of the VI response was substantially shortened by hypoxia (40 s) compared with air breathing (58 s) and increased by 100% O2 (92 s) and, even more so, by CO2 in O2 (101 s). We conclude that increased carotid (but not central) chemoreflex responsiveness speeds the kinetics of the exercise hyperpnea by a process that is not merely the consequence of increased ventilatory drive.     

Effects of CO2 breathing on ventilatory response to sustained hypoxia in normal adults

The relationship between CO2 and ventilatory response to sustained hypoxia was examined in nine normal young adults. At three different levels of end-tidal partial pressure of CO2 (PETCO2, approximately 35, 41.8, and 44.3 Torr), isocapnic hypoxia was induced for 25 min and after 7 min of breathing 21% O2, isocapnic hypoxia was reinduced for 5 min. Regardless of PETCO2 levels, the ventilatory response to sustained hypoxia was biphasic, characterized by an initial increase (acute hypoxic response, AHR), followed by a decline (hypoxic depression). The biphasic response pattern was due to alteration in tidal volume, which at all CO2 levels decreased significantly (P less than 0.05), without a significant change in breathing frequency. The magnitude of the hypoxic depression, independent of CO2, correlated significantly (r = 0.78, P less than 0.001) with the AHR, but not with the ventilatory response to CO2. The decline of minute ventilation was not significantly affected by PETCO2 [averaged 2.3 +/- 0.6, 3.8 +/- 1.3, and 4.5 +/- 2.2 (SE) 1/min for PETCO2 35, 41.8, and 44.3 Torr, respectively]. This decay was significant for PETCO2 35 and 41.8 Torr but not for 44.3 Torr. The second exposure to hypoxia failed to elicit the same AHR as the first exposure; at all CO2 levels the AHR was significantly greater (P less than 0.05) during the first hypoxic exposure than during the second. We conclude that hypoxia exhibits a long-lasting inhibitory effect on ventilation that is independent of CO2, at least in the range of PETCO2 studied, but is related to hypoxic ventilatory sensitivity.     

Effect of hypoxia on immediate ventilatory load response in dogs.

The effective elastance of the respiratory system (which has been previously shown to provide an index of the ability of the respiratory musculature to compensate rapidly for transient mechanical ventilatory loads) was measured in six hypoxic dogs to determine whether hypoxia hindered immediate load-compensatory mechanisms. The effective elastance value was computed from measurements of control tidal volume and the pressure developed at the airway opening during the first inspiratory effort following airway occlusion at FRC. The mean effective elastance was 197 cmH2O/l while the animals were breathing room air and did not change significantly when the animals were rendered hypoxic by reducing the inspired oxygen concentration, in five dogs, or by controlled hemorrhage, in two dogs. It was concluded that inasmuch as effective elastance measurements remain constant during hypoxia, the stability of ventilation is not significantly impaired in this situation.     

Increased normoxic ventilation induced by repetitive hypoxia in conscious dogs.

To determine if a long-lasting increase in normoxic ventilatory drive is induced in conscious animals by repetitive hypoxia, we examined the normoxic [arterial O2 saturation (SaO2) > 93%] ventilatory response following successive episodes of 2-min eucapnic hypoxic challenges (SaO2 = 80%) in awake tracheotomized dogs. End-tidal CO2 was maintained at the resting level during and after repetitive hypoxia. The experimental protocol was performed twice in each of five dogs on separate days. To determine if changes in normoxic ventilation occurred between episodes of repetitive hypoxia, data were compared from six periods (epochs) for all experiments. The mean minute ventilation (VI) during three normoxic periods between episodes of intermittent hypoxia was 135, 154, and 169% of control (P < 0.05). VI during a 30-min recovery period was still higher at 183 and 172% of control (P < 0.05). Normoxic VI between hypoxic and recovery periods was significantly higher than the corresponding values in sham experiments. Our results indicate that a long-lasting increase in normoxic ventilation can be evoked in an awake unanesthetized dog by a short exposure to repetitive hypoxia.