A Methodological Perspective on the Function and Assessment of Peripheral Chemoreceptors in Heart Failure: A Review of Data from Clinical Trials

Augmented peripheral chemoreceptor sensitivity (PChS) is a common feature of many sympathetically mediated diseases, among others, and it is an important mechanism of the pathophysiology of heart failure (HF). It is related not only to the greater severity of symptoms, especially to dyspnea and lower exercise tolerance but also to a greater prevalence of complications and poor prognosis. The causes, mechanisms, and impact of the enhanced activity of peripheral chemoreceptors (PChR) in the HF population are subject to intense research. Several methodologies have been established and utilized to assess the PChR function. Each of them presents certain advantages and limitations. Furthermore, numerous factors could influence and modulate the response from PChR in studied subjects. Nevertheless, even with the impressive number of studies conducted in this field, there are still some gaps in knowledge that require further research. We performed a review of all clinical trials in HF human patients, in which the function of PChR was evaluated. This review provides an extensive synthesis of studies evaluating PChR function in the HF human population, including methods used, factors potentially influencing the results, and predictors of increased PChS.

Assessing central and peripheral respiratory chemoreceptor interaction in humans

NEW FINDINGS

What is the central question of this study? We investigated the interaction between central and peripheral respiratory chemoreceptors in healthy, awake human participants by (a) using a background of step increases in steady-state normoxic fraction of inspired carbon dioxide to alter central chemoreceptor activation and (b) using the transient hypoxia test to target the peripheral chemoreceptors. What is the main finding and its importance? Our data suggests that the central-peripheral respiratory chemoreceptor interaction is additive in minute ventilation and respiratory rate, but hypoadditive in tidal volume. Our study adds important new data in reconciling chemoreceptor interaction in awake healthy humans, and is consistent with previous reports of simple addition in intact rodents and humans.

ABSTRACT

Arterial blood gas levels are maintained through respiratory chemoreflexes, mediated by central (CCR) in the CNS and peripheral (PCR) chemoreceptors located in the carotid bodies. The interaction between central and peripheral chemoreceptors is controversial, and few studies have investigated this interaction in awake healthy humans, in part due to methodological challenges. We investigated the interaction between the CCRs and PCRs in healthy humans using a transient hypoxia test (three consecutive breaths of 100% N₂ ; TT-HVR), which targets the stimulus and temporal domain specificity of the PCRs. TT-HVRs were superimposed upon three randomized background levels of steady-state inspired fraction of normoxic CO₂ (F_I CO₂ ; 0, 0.02 and 0.04). Chemostimuli (calculated oxygen saturation; ScO₂ ) and respiratory variable responses (respiratory rate, inspired tidal volume and ventilation; R_R , V_TI , V̇_I ), were averaged from all three TT-HVR trials at each F_I CO₂ level. Responses were assessed as (a) a change from BL (delta; ∆) and (b) indexed against ∆ScO₂ . Aside from a significantly lower ∆V_TI response in 0.04 F_I CO₂ (P = 0.01), the hypoxic rate responses (∆R_R or ∆R_R /∆ScO₂ ; P = 0.46, P = 0.81), hypoxic tidal volume response (∆V_TI /∆ScO₂ ; P = 0.08) and the hypoxic ventilatory responses (∆V̇_I and (∆V̇_I /∆ScO₂ ; P = 0.09 and P = 0.31) were not significantly different across F_I CO₂ trials. Our data suggests simple addition between central and peripheral chemoreceptors in V̇_I , which is mediated through simple addition in R_R responses, but hypo-addition in V_TI responses. Our study adds important new data in reconciling chemoreceptor interaction in awake healthy humans, and is consistent with previous reports of simple addition in intact rodents and humans. This article is protected by copyright. All rights reserved.

Peripheral chemoreflex activation induces expiratory but not inspiratory excitation of C1 pre-sympathetic neurones of rats

AIMS

Stimulation of peripheral chemoreceptors, as during hypoxia, increases breathing and respiratory-related sympathetic bursting. Activation of catecholaminergic C1 neurones induces sympathoexcitation, while its ablation reduces the chemoreflex sympathoexcitatory response. However, no study has determined the respiratory phase(s) in which the pre-sympathetic C1 neurones are recruited by peripheral chemoreceptor and whether C1 neurone activation affects all phases of respiratory modulation of sympathetic activity. We addressed these unknowns by testing the hypothesis that peripheral chemoreceptor activation excites pre-sympathetic C1 neurones during inspiration and expiration.

METHODS

Using the in situ preparation of rat, we made intracellular recordings from baroreceptive pre-sympathetic C1 neurones during peripheral chemoreflex stimulation. We optogenetically activated C1 neurones selectively and compared any respiratory-phase-related increases in sympathetic activity with that which occurs following stimulation of the peripheral chemoreflex.

RESULTS

Activation of peripheral chemoreceptors using cytotoxic hypoxia (potassium cyanide) increased the firing frequency of C1 neurones and both the frequency and amplitude of their excitatory post-synaptic currents during the phase of expiration only. In contrast, optogenetic stimulation of C1 neurones activates inspiratory neurones, which secondarily inhibit expiratory neurones, but produced comparable increases in sympathetic activity across all phases of respiration.

CONCLUSION

Our data reveal that the peripheral chemoreceptor-mediated expiratory-related sympathoexcitation is mediated through excitation of expiratory neurones antecedent to C1 pre-sympathetic neurones; these may be found in the Kölliker-Fuse nucleus. Despite peripheral chemoreceptor excitation of inspiratory neurones, these do not trigger C1 neurone-mediated increases in sympathetic activity. These studies provide compelling novel insights into the functional organization of respiratory-sympathetic neural networks.

Induction of Day-Time Periodic Breathing is Associated With Augmented Reflex Response From Peripheral Chemoreceptors in Male Patients With Systolic Heart Failure

Spontaneous day-time periodic breathing (sPB) constitutes a common phenomenon in systolic heart failure (HF). However, it is unclear whether PB during wakefulness could be easily induced and what are the physiological and clinical correlates of patients with HF in whom PB induction is possible. Fifty male HF patients (age 60.8 ± 9.8 years, left ventricle ejection fraction 28.0 ± 7.4%) were prospectively screened and 46 enrolled. After exclusion of patients with sPB the remaining underwent trial of PB induction using mild hypoxia (stepwise addition of nitrogen gas to breathing mixture) which resulted in identification of inducible (iPB) in 51%. All patients underwent assessment of hypoxic ventilatory response (HVR) using transient hypoxia and of hypercapnic ventilatory response (HCVR) employing Read’s rebreathing method. The induction trial did not result in any adverse events and minimal SpO₂ during nitrogen administration was ∼85%. The iPB group (vs. non-inducible PB group, nPB) was characterized by greater HVR (0.90 ± 0.47 vs. 0.50 ± 0.26 L/min/%; p <0.05) but comparable HCVR (0.88 ± 0.54 vs. 0.67 ± 0.68 L/min/mmHg; p = NS) and by worse clinical and neurohormonal profile. Mean SpO₂ which induced first cycle of PB was 88.9 ± 3.7%, while in sPB mean SpO₂ preceding first spontaneous cycle of PB was 96.0 ± 2.5%. There was a reverse relationship between HVR and the relative variation of SpO₂ during induced PB (r = −0.49, p = 0.04). In summary, PB induction is feasible and safe in HF population using simple and standardized protocol employing incremental, mild hypoxia. Pathophysiology of iPB differs from sPB, as it relies mostly on overactive peripheral chemoreceptors. At the same time enhanced HVR might play a protective role against profound hypoxia during iPB.

A5 noradrenergic-projecting C1 neurons activate sympathetic and breathing outputs in anaesthetized rats

NEW FINDINGS

What is the central question of this study? C1 neurons innervate pontine noradrenergic cell groups, including the A5 region: do A5 noradrenergic neurons contribute to the activation of sympathetic and respiratory responses produced by selective activation of the C1 group of neurons. What is the main finding and its importance? The increase in sympathetic and respiratory activities elicited by selective stimulation of C1 neurons is reduced after blockade of excitatory amino acid within the A5 region, suggesting that the C1-A5 pathway might be important for sympathetic-respiratory control.

ABSTRACT

Adrenergic C1 neurons innervate and excite pontine noradrenergic cell groups, including the ventrolateral pontine noradrenergic region (A5). Here, we tested the hypothesis that C1 activates A5 neurons through the release of glutamate and this effect is important for sympathetic and respiratory control. Using selective tools, we restricted the expression of channelrhodopsin2 under the control of the artificial promoter PRSx8 to C1 neurons (69%). Transduced catecholaminergic terminals within the A5 region are in contact with noradrenergic A5 neurons and the C1 terminals within the A5 region are predominantly glutamatergic. In a different group of animals, we performed retrograde lesion of C1 adrenergic neurons projecting to the A5 region with unilateral injection of the immunotoxin anti-dopamine β-hydroxylase-saporin (anti-DβH-SAP) directly into the A5 region during the hypoxic condition. As expected, hypoxia (8% O₂ , 3 h) induced a robust increase in fos expression within the catecholaminergic C1 and A5 regions of the brainstem. Depletion of C1 cells projecting to the A5 regions reduced fos immunoreactivity induced by hypoxia within the C1 region. Physiological experiments showed that bilateral injection of kynurenic acid (100 mM) into the A5 region reduced the rise in mean arterial pressure, and sympathetic and phrenic nerve activities produced by optogenetic stimulation of C1 cells. In conclusion, the C1 neurons activate the ventrolateral pontine noradrenergic neurons (A5 region) possibly via the release of glutamate and might be important for sympathetic and respiratory outputs in anaesthetized rats.

Carotid chemoreceptors have a limited role in mediating the hyperthermia-induced hyperventilation in exercising humans

Hyperthermia causes hyperventilation at rest and during exercise. We previously reported that carotid chemoreceptors partly contribute to the hyperthermia-induced hyperventilation at rest. However, given that a hyperthermia-induced hyperventilation markedly differs between rest and exercise, the results obtained at rest may not be representative of the response in exercise. Therefore, we evaluated whether carotid chemoreceptors contribute to hyperthermia-induced hyperventilation in exercising humans. Eleven healthy young men (23 ± 2 yr) cycled in the heat (37°C) at a fixed submaximal workload equal to ~55% of the individual's predetermined peak oxygen uptake (moderate intensity). To suppress carotid chemoreceptor activity, 30-s hyperoxia breathing (100% O₂) was performed at rest (before exercise) and during exercise at increasing levels of hyperthermia as defined by an increase in esophageal temperature of 0.5°C (low), 1.0°C (moderate), 1.5°C (high), and 2.0°C (severe) above resting levels. Ventilation during exercise gradually increased as esophageal temperature increased (all P ≤ 0.05), indicating that hyperthermia-induced hyperventilation occurred. Hyperoxia breathing suppressed ventilation in a greater manner during exercise (-9 to -13 l/min) than at rest (-2 ± 1 l/min); however, the magnitude of reduction during exercise did not differ at low (0.5°C) to severe (2.0°C) increases in esophageal temperature (all P > 0.05). Similarly, hyperoxia-induced changes in ventilation during exercise as assessed by percent change from prehyperoxic levels were not different at all levels of hyperthermia (~15-20%, all P > 0.05). We show that in young men carotid chemoreceptor contribution to hyperthermia-induced hyperventilation is relatively small at low-to-severe increases in body core temperature induced by moderate-intensity exercise in the heat. NEW & NOTEWORTHY Exercise-induced increases in hyperthermia cause a progressive increase in ventilation in humans. However, the mechanisms underpinning this response remain unresolved. We showed that in young men hyperventilation associated with exercise-induced hyperthermia is not predominantly mediated by carotid chemoreceptors. This study provides important new insights into the mechanism(s) underpinning the regulation of hyperthermia-induced hyperventilation in humans and suggests that factor(s) other than carotid chemoreceptors play a more important role in mediating this response.

The hypoxic test in preterm neonates reinvestigated

AIM

We currently lack a suitable gold-standard method for implementation on modern equipment to assess peripheral chemoreceptor sensitivity. The aim of the present study was to develop an accurate and reproducible method for assessing peripheral chemoreceptors sensitivity in sleeping preterm neonates.

METHODS

A poïkilocapnic hypoxic test was performed twice during rapid eye movement sleep (REM sleep) and non-rapid eye movement sleep (nonREM sleep). The infant breathed hypoxic gas (15% O₂ ) for 60 s. The ventilatory response to hypoxia was assessed by comparing minute ventilation during the control period (21% O₂ ) with successive 4-cycles sequences during hypoxia. We detected the first statistically significant increase in minute ventilation and recorded the corresponding response time.

RESULTS

During normoxia, minute ventilation was higher during REM sleep than in nonREM sleep (428.1 mL · min^-1  · kg^-1 [307.7-633.6]; 388.8 mL · min^-1  · kg^-1 [264.7-608.0], respectively; P = 0.001). After hypoxia, minute ventilation increased in both REM and nonREM sleep. The response was significantly higher in REM than in nonREM (25.3% [10.8-80.0] and 16.8% [7.5-33.2], respectively; P = 0.005). The intraclass correlation coefficients for all respiratory parameters were above 0.90.

CONCLUSION

We have developed a highly reliable method for assessing peripheral chemoreceptors sensitivity at the response time to hypoxia. In the future, researchers could use this method to assess the involvement of peripheral chemoreceptors in infants who experience chronic hypoxia (e.g. in bronchopulmonary dysplasia and recurrent apnea).

Spinal cord injury is associated with enhanced peripheral chemoreflex sensitivity

Sleep‐disordered breathing (SDB) is prevalent in individuals with chronic spinal cord injury (SCI), but the exact mechanism is unknown. The aim of this study was to investigate whether peripheral chemoreceptors activity is enhanced in individuals with chronic SCI compared to abled‐bodied control subjects using CO₂ and O₂ chemical tests. In protocol (1) 30 subjects (8 cervical [cSCI], 7 thoracic [tSCI] and 15 able‐bodied [AB]) were studied to determine the ventilatory response to hyperoxia during wakefulness in the supine position. In protocol (2) 24 subjects (6 cSCI, 6 tSCI, and 12 AB subjects) were studied to determine the ventilatory response to a single breath of CO₂ (SBCO₂). The chemoreflex response to SBCO₂ was calculated as ∆V_E/∆CO₂ (L/min/mmHg). The ventilatory response to hyperoxia was defined as the % change in V_T following acute hyperoxia compared to preceding baseline. During hyperoxia SCI subjects had a significant decrease in V_T and V_E (63.4 ± 21.7% and 63.1 ± 23.0% baseline, respectively, P < 0.05) compared to AB (V_T: 87.1 ± 14.3% and V_E: 91.38 ± 15.1% baseline, respectively, P < 0.05). There was no significant difference between cSCI and tSCI in the V_T or V_E during hyperoxia (P = NS). There was no significant correlation between AHI and V_E% baseline (r = −0.28) in SCI and AB (n = 30). SCI participants had a greater ventilatory response to an SBCO₂ than AB (0.78 ± 0.42 L/min/mmHg vs. 0.26 ± 0.10 L/min/mmHg, respectively, P < 0.05). Peripheral ventilatory chemoresponsiveness is elevated in individuals with chronic SCI compared to able‐bodied individuals.

Gene expression analyses reveal metabolic specifications in acute O2 -sensing chemoreceptor cells

KEY POINTS

Glomus cells in the carotid body (CB) and chromaffin cells in the adrenal medulla (AM) are essential for reflex cardiorespiratory adaptation to hypoxia. However, the mechanisms whereby these cells detect changes in O₂ tension are poorly understood. The metabolic properties of acute O₂ -sensing cells have been investigated by comparing the transcriptomes of CB and AM cells, which are O₂ -sensitive, with superior cervical ganglion neurons, which are practically O₂ -insensitive. In O₂ -sensitive cells, we found a characteristic prolyl hydroxylase 3 down-regulation and hypoxia inducible factor 2α up-regulation, as well as overexpression of genes coding for three atypical mitochondrial electron transport subunits and pyruvate carboxylase, an enzyme that replenishes tricarboxylic acid cycle intermediates. In agreement with this observation, the inhibition of succinate dehydrogenase impairs CB acute O₂ sensing. The responsiveness of peripheral chemoreceptor cells to acute hypoxia depends on a 'signature metabolic profile'.

ABSTRACT

Acute O₂ sensing is a fundamental property of cells in the peripheral chemoreceptors, e.g. glomus cells in the carotid body (CB) and chromaffin cells in the adrenal medulla (AM), and is necessary for adaptation to hypoxia. These cells contain O₂ -sensitive ion channels, which mediate membrane depolarization and transmitter release upon exposure to hypoxia. However, the mechanisms underlying the detection of changes in O₂ tension by cells are still poorly understood. Recently, we suggested that CB glomus cells have specific metabolic features that favour the accumulation of reduced quinone and the production of mitochondrial NADH and reactive oxygen species during hypoxia. These signals alter membrane ion channel activity. To investigate the metabolic profile characteristic of acute O₂ -sensing cells, we used adult mice to compare the transcriptomes of three cell types derived from common sympathoadrenal progenitors, but exhibiting variable responsiveness to acute hypoxia: CB and AM cells, which are O₂ -sensitive (glomus cells > chromaffin cells), and superior cervical ganglion neurons, which are practically O₂ -insensitive. In the O₂ -sensitive cells, we found a characteristic mRNA expression pattern of prolyl hydroxylase 3/hypoxia inducible factor 2α and up-regulation of several genes, in particular three atypical mitochondrial electron transport subunits and some ion channels. In addition, we found that pyruvate carboxylase, an enzyme fundamental to tricarboxylic acid cycle anaplerosis, is overexpressed in CB glomus cells. We also observed that the inhibition of succinate dehydrogenase impairs CB acute O₂ sensing. Our data suggest that responsiveness to acute hypoxia depends on a 'signature metabolic profile' in chemoreceptor cells.

Interindividual variability in the dose-specific effect of dopamine on carotid chemoreceptor sensitivity to hypoxia

Human studies use varying levels of low-dose (1-4 μg·kg(-1)·min(-1)) dopamine to examine peripheral chemosensitivity, based on its known ability to blunt carotid body responsiveness to hypoxia. However, the effect of dopamine on the ventilatory responses to hypoxia is highly variable between individuals. Thus we sought to determine 1) the dose response relationship between dopamine and peripheral chemosensitivity as assessed by the ventilatory response to hypoxia in a cohort of healthy adults, and 2) potential confounding cardiovascular responses at variable low doses of dopamine. Young, healthy adults (n = 30, age = 32 ± 1, 24 male/6 female) were given intravenous (iv) saline and a range of iv dopamine doses (1-4 μg·kg(-1)·min(-1)) prior to and throughout five hypoxic ventilatory response (HVR) tests. Subjects initially received iv saline, and after each HVR the dopamine infusion rate was increased by 1 μg·kg(-1)·min(-1). Tidal volume, respiratory rate, heart rate, blood pressure, and oxygen saturation were continuously measured. Dopamine significantly reduced HVR at all doses (P < 0.05). When subjects were divided into high (n = 13) and low (n = 17) baseline chemosensitivity, dopamine infusion (when assessed by dose) reduced HVR in the high group only (P < 0.01), with no effect of dopamine on HVR in the low group (P > 0.05). Dopamine infusion also resulted in a reduction in blood pressure (3 μg·kg(-1)·min(-1)) and total peripheral resistance (1-4 μg·kg(-1)·min(-1)), driven primarily by subjects with low baseline chemosensitivity. In conclusion, we did not find a single dose of dopamine that elicited a nadir HVR in all subjects. Additionally, potential confounding cardiovascular responses occur with dopamine infusion, which may limit its usage.

Peripheral chemoreceptors tune inspiratory drive via tonic expiratory neuron hubs in the medullary ventral respiratory column network

Models of brain stem ventral respiratory column (VRC) circuits typically emphasize populations of neurons, each active during a particular phase of the respiratory cycle. We have proposed that "tonic" pericolumnar expiratory (t-E) neurons tune breathing during baroreceptor-evoked reductions and central chemoreceptor-evoked enhancements of inspiratory (I) drive. The aims of this study were to further characterize the coordinated activity of t-E neurons and test the hypothesis that peripheral chemoreceptors also modulate drive via inhibition of t-E neurons and disinhibition of their inspiratory neuron targets. Spike trains of 828 VRC neurons were acquired by multielectrode arrays along with phrenic nerve signals from 22 decerebrate, vagotomized, neuromuscularly blocked, artificially ventilated adult cats. Forty-eight of 191 t-E neurons fired synchronously with another t-E neuron as indicated by cross-correlogram central peaks; 32 of the 39 synchronous pairs were elements of groups with mutual pairwise correlations. Gravitational clustering identified fluctuations in t-E neuron synchrony. A network model supported the prediction that inhibitory populations with spike synchrony reduce target neuron firing probabilities, resulting in offset or central correlogram troughs. In five animals, stimulation of carotid chemoreceptors evoked changes in the firing rates of 179 of 240 neurons. Thirty-two neuron pairs had correlogram troughs consistent with convergent and divergent t-E inhibition of I cells and disinhibitory enhancement of drive. Four of 10 t-E neurons that responded to sequential stimulation of peripheral and central chemoreceptors triggered 25 cross-correlograms with offset features. The results support the hypothesis that multiple afferent systems dynamically tune inspiratory drive in part via coordinated t-E neurons.

Ventilatory response to high inspired carbon dioxide concentrations in anesthetized dogs

Background:

The ventilation ( ) response to inspired CO₂ has been extensively studied, but rarely with concentrations >10%.

Aims:

These experiments were performed to determine whether would increase correspondingly to higher concentrations and according to conventional chemoreceptor time delays.

Materials and Methods:

We exposed anesthetized dogs acutely, with and without vagotomy and electrical stimulation of the right vagus, to 20-100% CO₂-balance O₂ and to 0 and 10% O₂-balance N₂.

Results:

The time delays decreased and response magnitude increased with increasing concentrations (p<0.01), but at higher concentrations the time delays were shorter than expected, i.e., 0.5 s to double at 100% CO₂, with the response to 0% O₂ being ~3 s slower. Right vagotomy significantly reduced baseline breathing frequency (fR), increased tidal volume (VT) and increased the time delay by ~3 s. Bilateral vagotomy further reduced baseline fR and , and reduced the response to CO₂ and increased the time delay by ~12 s. Electro-stimulation of the peripheral right vagus while inspiring CO₂ caused a 13 s asystole and further reduced and delayed the response, especially after bilateral vagotomy, shifting the mode from VT to fR.

Conclusions:

Results indicate that airway or lung receptors responded to the rapid increase in lung H⁺ and that vagal afferents and unimpaired circulation seem necessary for the initial rapid response to high CO₂ concentrations by receptors upstream from the aortic bodies.

Carotid body function in aged rats: responses to hypoxia, ischemia, dopamine, and adenosine

The carotid body (CB) is the main arterial chemoreceptor with a low threshold to hypoxia. CB activity is augmented by A(2)-adenosine receptors stimulation and attenuated by D(2)-dopamine receptors. The effect of aging on ventilatory responses mediated by the CB to hypoxia, ischemia, and to adenosine and dopamine administration is almost unknown. This study aims to investigate the ventilatory response to ischemia and to adenosine, dopamine, and their antagonists in old rats, as well as the effect of hypoxia on adenosine 3',5'-cyclic monophosphate (cAMP) accumulation in the aged CB. In vivo experiments were performed on young and aged rats anesthetized with pentobarbitone and breathing spontaneously. CB ischemia was induced by bilateral common carotid occlusions. cAMP content was measured in CB incubated with different oxygen concentrations. Hyperoxia caused a decrease in cAMP in the CB at all ages, but no differences were found between normoxia and hypoxia or between young and old animals. The endogenous dopaminergic inhibitory tonus is slightly reduced. However, both the ventilation decrease caused by exogenous dopamine and the increase mediated by A(2A)-adenosine receptors are not impaired in aged animals. The bradycardia induced by adenosine is attenuated in old rats. The CB's peripheral control of ventilation is preserved during aging. Concerns have also arisen regarding the clinical usage of adenosine to revert supraventricular tachycardia and the use of dopamine in critical care situations involving elderly people.

Breathing at high altitude

Acclimatization to long-term hypoxia takes place at high altitude and allows gradual improvement of the ability to tolerate the hypoxic environment. An important component of this process is the hypoxic ventilatory acclimatization (HVA) that develops over several days. HVA reveals profound cellular and neurochemical re-organization occurring both in the peripheral chemoreceptors and in the central nervous system (in brainstem respiratory groups). These changes lead to an enhanced activity of peripheral chemoreceptor and re-inforce the central translation of peripheral inputs to efficient respiratory motor activity under the steady low O(2) pressure. We will review the cellular processes underlying these changes with a particular emphasis on changes of neurotransmitter function and ion channel properties in peripheral chemoreceptors, and present evidence that low O(2) level acts directly on brainstem nuclei to induce cellular changes contributing to maintain a high tonic respiratory drive under chronic hypoxia.