Reduced tidal volume increases 'air hunger' at fixed PCO2 in ventilated quadriplegics.

The act of breathing diminishes the discomfort associated with hypercapnia and breath-holding. To investigate the mechanisms involved in this effect, we studied the effect of tidal volume (VT) on CO2-evoked air hunger in 5 high-level quadriplegic subjects whose ventilatory capacity was negligible, and who lacked sensory information from the chest wall. Subjects were ventilated at constant frequency with a hyperoxic gas mixture, and end-tidal PCO2 was maintained at a constant but elevated level. VT was varied between the subjects' normal VT and a smaller VT. Subjects used a category scale to rate their respiratory discomfort or 'air hunger' at 30-40 sec intervals. In 4 of 5 subjects there was a strong inverse relationship between breath size and air hunger ratings. The quality of the sensation associated with reduced VT was nearly identical to that previously experienced with CO2 alone. We conclude that afferent information from the lungs and upper airways is sufficient to modify the sensation of air hunger.     

Neuroimaging evidence implicating cerebellum in support of sensory/cognitive processes associated with thirst

Recent studies implicate the cerebellum, long considered strictly a motor control structure, in cognitive, sensory, and affective phenomenon. The cerebellum, a phylogenetically ancient structure, has reciprocal ancient connections to the hypothalamus, a structure important in vegetative functions. The present study investigated whether the cerebellum was involved in vegetative functions and the primal emotions engendered by them. Using positron emission tomography, we examined the effects on the cerebellum of the rise of plasma sodium concentration and the emergence of thirst in 10 healthy adults. The correlation of regional cerebral blood flow with subjects' ratings of thirst showed major activation in the vermal central lobule. During the development of thirst, the anterior and posterior quadrangular lobule, lingula, and the vermis were activated. At maximum thirst and then during irrigation of the mouth with water to alleviate dryness, the cerebellum was less activated. However, 3 min after drinking to satiation, the anterior quadrangular lobule and posterior cerebellum were highly activated. The increased cerebellar activity was not related to motor behavior as this did not occur. Instead, responses in ancient cerebellar regions (vermis, fastigal nucleus, archicerebellum) may be more directly related to vegetative and affective aspects of thirst experiences, whereas activity in neocerebellar (posterior) regions may be related to sensory and cognitive aspects. Moreover, the cerebellum is apparently not involved in the computation of thirst per se but rather is activated during changes in thirst/satiation state when the brain is "vigilant" and is monitoring its sensory systems. Some neocerebellar activity may also reflect an intentionality for gratification by drinking inherent in the consciousness of thirst.     

Respiratory patterns in anesthetised rats before and after anemic decerebration.

Experiments were undertaken to test the comparability of changes in respiratory frequency and tidal volume during hypoxia and hypercapnia in rats with and without intact peripheral chemoreceptors and with intact vagi. Neural organisation of respiratory control was perturbed by anemic decerebration, achieved by ligation of the common carotid and basilar arteries. Ischemia of the brain was produced as far candal as the rostral pontine nuclei involved in respiratory control but left the medulla well perfused. The dominant respiratory effect in animals breathing air or oxygen was polypnea with hypocapnia (mean PaCO2 when breathing air 24.7 mmHg, when breathing oxygen 29.6 mmHg). After decerebration the increase of ventilation produced by breathing 10% O2 in N2 was reduced compared with responses in the intact state but levels of ventilation (V1) in hypoxia were similar to those before decerebration. After decerebration, the increase of ventilation produced by breathing 5% CO2 was greatly reduced and the level of V1 in animals breathing CO2 was significantly less than in the intact state. Intermediate changes were seen in animals breathing 2-3% CO2 which converted the hypocapnia (PaCO2 30.9 mmHg) to eucapnia (PaCO2 46.4 mmHg). In the intact state, hypoxia dominantly caused increased frequency (f) and hypercapnia caused increased tidal volume (VT); after decerebration, hypoxia produced reduction of VT while hypercapnia produced reduction of f. Bilateral carotid sinus nerve section in decerebrate animals eliminated the ventilatory response to hypoxia but left the responses to hypercapnia unaltered. The results point to differences in the mechanisms by which hypoxia and hypercapnia influence respiration in both intact and decerebrate animals with carotid sinus and vagus nerves functional. The differences can now be interpreted in terms of specific neural features of respiratory control.     

'Air hunger' from increased PCO2 persists after complete neuromuscular block in humans

The tolerance of totally curarized subjects for prolonged breath hold is viewed by many as evidence that respiratory muscle contraction is essential to generate the sensation of breathlessness. Although conflicting evidence exists, none of it was obtained during total neuromuscular block. We completely paralyzed four normal, unsedated subjects with vecuronium (a non-depolarizing neuromuscular blocker). Subjects were mechanically ventilated with hyperoxic gas mixtures at fixed rate and tidal volume. End-expiratory PCO2 (PETCO2) was varied surreptitiously by changing inspired PCO2. Subjects rated their respiratory discomfort or 'air hunger' every 45 sec. At low PETCO2 (median 35 Torr) they felt little or no air hunger. When PETCO2 was raised (median 44 Torr) all subjects reported severe air hunger. They had reported the same degree of air hunger at essentially the same PETCO2 before paralysis. When questioned afterwards all subjects said the sensation could be described by the terms 'air hunger', 'urge to breathe', and 'shortness of breath', and that is was like breath holding. They reported no fundamental difference in the sensation before and after paralysis. We conclude that respiratory muscle contraction is not important in the genesis of air hunger evoked by hypercapnia.     

The perception of respiratory work and effort can be independent of the perception of air hunger

Dyspnea in patients could arise from both an urge to breathe and increased effort of breathing. Two qualitatively different sensations, "air hunger" and "respiratory work and effort," arising from different afferent sources are hypothesized. In the laboratory, breathing below the spontaneous level may produce an uncomfortable sensation of air hunger, and breathing above it a sensation of work or effort. Measurement of a single sensory dimension cannot distinguish these as separate sensations; we therefore measured two sensory dimensions and attempted to vary them independently. In five normal subjects we obtained simultaneous ratings of air hunger and of work and effort while independently varying PCO(2) or the level of targeted voluntary breathing. We found a difference in response to the two stimulus dimensions: air hunger ratings changed more steeply when PCO(2) was altered and ventilation was constant; work or effort ratings changed more steeply when ventilation was altered and PCO(2) was constant. We conclude that "air hunger" is qualitatively different from "work and effort" and arises from different afferent sources.     

'Air hunger' arising from increased PCO2 in mechanically ventilated quadriplegics

A number of investigators have proposed that the sense of respiratory discomfort accompanying hypercapnia depends on respiratory mechanoreceptors which inform the sensory cortex of reflex increases in breathing. To test this hypothesis, we studied subjects whose respiratory muscles were paralyzed, and who were thus unable to increase breathing in response to hypercapnia. We gradually elevated inspired PCO2 in four tracheostomized quadriplegic subjects supported by constant mechanical ventilation. These subjects reported sensations of 'air hunger' (e.g., "short of breath", "air-starved") when end-tidal PCO2 increased 10 Torr (mean) above their resting levels. In a second experiment we used the forced-choice technique to determine the ability of three of these subjects to detect repeated changes of end-tidal PCO2. Two detected 7 Torr changes, the third detected 11 Torr changes. These data suggest that changes in breathing are not necessary to evoke the sense of 'air hunger'. We conclude that the likely mechanisms are (1) projection of chemoreceptor afferent traffic to the sensory cortex, and (2) projection of corollary discharge from brainstem respiratory centers to the sensory cortex.     

Brain responses associated with consciousness of breathlessness (air hunger)

Little is known about the physiological mechanisms subserving the experience of air hunger and the affective control of breathing in humans. Acute hunger for air after inhalation of CO(2) was studied in nine healthy volunteers with positron emission tomography. Subjective breathlessness was manipulated while end-tidal CO(2-) was held constant. Subjects experienced a significantly greater sense of air hunger breathing through a face mask than through a mouthpiece. The statistical contrast between the two conditions delineated a distributed network of primarily limbic/paralimbic brain regions, including multiple foci in dorsal anterior and middle cingulate gyrus, insula/claustrum, amygdala/periamygdala, lingual and middle temporal gyrus, hypothalamus, pulvinar, and midbrain. This pattern of activations was confirmed by a correlational analysis with breathlessness ratings. The commonality of regions of mesencephalon, diencephalon and limbic/paralimbic areas involved in primal emotions engendered by the basic vegetative systems including hunger for air, thirst, hunger, pain, micturition, and sleep, is discussed with particular reference to the cingulate gyrus. A theory that the phylogenetic origin of consciousness came from primal emotions engendered by immediate threat to the existence of the organism is discussed along with an alternative hypothesis by Edelman that primary awareness emerged with processes of ongoing perceptual categorization giving rise to a scene [Edelman, G. M. (1992) Bright Air, Brilliant Fire (Penguin, London)].     

The impact of hypercapnia on retinal capillary blood flow assessed by scanning laser Doppler flowmetry

AIM: To determine the effect of hypercapnia on retinal capillary blood flow using scanning laser Doppler flowmetry (SLDF). METHODS: One randomly selected eye of each of 10 normal healthy subjects (mean age 25 years, SD 2.3) was studied. Subjects breathed unrestricted air for 15 min before (baseline) and after raising fractional (percent) end-tidal concentration of CO2 (FETCO2) for 15 min by adding low flows of CO2 to air entering a sequential gas delivery circuit attached to a nasal mask. Five good quality baseline SLDF images were acquired both of the optic nerve head (ONH) and of the macula. Subsequently, a minimum of 7 sequential images were acquired during hypercapnia. Five further images were acquired of the ONH, or of the macula, after returning to unlimited air breathing. The respiratory parameters of subjects were continually monitored. RESULTS: The group mean increase in end-tidal CO2 was 14.13% (SD 4.10) relative to baseline. The nasal macula (P = 0.028) and foveal (P = 0.042) areas showed a significant increase in retinal capillary blood flow in response to hypercapnia while no significant change was noted in the ONH or temporal macula areas. Change in blood flow significantly correlated with change of FETCO2 and/or end-tidal PO2 for 3 of the 4 locations. CONCLUSIONS: Hypercapnia provoked a significant increase in retinal capillary blood flow in 2 of 4 retinal locations. Hypercapnia also induced a change in respiratory parameters that significantly correlated with change in retinal capillary blood flow in 3 of the 4 locations.     

Pulmonary vagal modulation of ventilation in toads (Bufo marinus)

This study examined the role of pulmonary vagal feedback on hypercapnic chemosensitivity and breathing pattern formation in cane toads (Bufo marinus). Decerebrate, paralysed toads were uni-directionally ventilated with air, 2.5% CO(2) or 5.0% CO(2) with the lungs inflated or deflated, before and after pulmonary vagotomy. Motor output from the mandibular branch of the trigeminal nerve served as an index of fictive breathing. As respiratory drive was increased, breathing frequency increased and breaths were clustered into discrete episodes separated by periods of apnea. Lung deflation tended to enhance episodic breathing while inflation biased the system towards apnea at low levels of respiratory drive and a pattern of continuous, small breaths at higher levels of respiratory drive. Following bilateral pulmonary vagotomy there was no increase in ventilation during hypercapnia and lung inflation/deflation had no effect on breathing pattern. In isolated brainstem-spinal cord preparations from the same animals, all variables associated with fictive breathing were unaffected by changes in superfusate pH from 8.0 to 7.6. The breathing pattern from the in vitro preparations was highly variable. This study demonstrates a crucial role for vagal feedback in modulating respiration and the respiratory responses to hypercapnia in B. marinus.     

Hypercapnia during weaning. A complication of nutritional support

Excess carbohydrate calories in total parenteral nutrition (TPN) solutions can precipitate acute hypercapnic respiratory failure in patients with chronic lung disease secondary to increased carbon dioxide (CO2) production. Two young patients recovering from the adult respiratory distress syndrome experienced hypercapnia during weaning as a result of nutritionally related increased CO2 production. As carbohydrate calories were decreased, CO2 production diminished and hypercapnia resolved. Hypercapnia as a complication of nutritional support during weaning can occur in patients without chronic lung disease and is corrected by decreasing carbohydrate calories.     

Neuroimaging of cerebral activations and deactivations associated with hypercapnia and hunger for air

There are defined medullary, mesencephalic, hypothalamic, and thalamic functions in regulation of respiration, but knowledge of cortical control and the elements subserving the consciousness of breathlessness and air hunger is limited. In nine young adults, air hunger was produced acutely by CO(2) inhalation. Comparisons were made with inhalation of a N(2)/O(2) gas mixture with the same apparatus, and also with paced breathing, and with eyes closed rest. A network of activations in pons, midbrain (mesencephalic tegmentum, parabrachial nucleus, and periaqueductal gray), hypothalamus, limbic and paralimbic areas (amygdala and periamygdalar region) cingulate, parahippocampal and fusiform gyrus, and anterior insula were seen along with caudate nuclei and pulvinar activations. Strong deactivations were seen in dorsal cingulate, posterior cingulate, and prefrontal cortex. The striking response of limbic and paralimbic regions points to these structures having a singular role in the affective sequelae entrained by disturbance of basic respiratory control whereby a process of which we are normally unaware becomes a salient element of consciousness. These activations and deactivations include phylogenetically ancient areas of allocortex and transitional cortex that together with the amygdalar/periamygdalar region may subserve functions of emotional representation and regulation of breathing.     

Hyperoxic-induced hypercapnia in stable chronic obstructive pulmonary disease

We investigated the mechanism of hyperoxic-induced hypercapnia in 17 stable patients with moderate to severe chronic obstructive pulmonary disease (mean FEV1 = 0.95 L and FVC = 2.43 L). Ventilatory and mouth occlusion pressure (P0.1) responses to hypercapnia and hypoxia were measured with standard rebreathing techniques. In a randomized, single-blind fashion, we studied the effect of 15 min of hyperoxia or air on transcutaneous carbon dioxide (PtcCO2), CO2 production (VCO2), total minute ventilation (VE), and calculated dead space to tidal volume ratio (VD/VT). With O2, the PtcCO2 (p less than 0.01) and VD/VT (p less than 0.02) increased. The change in PtcCO2 with O2 was not significantly related to the indices of respiratory drive, nor to the baseline PtcCO2 or SaO2, but was related to the FEV1 (p less than 0.05). The O2 caused a slight decrease in mean VE and mean VCO2, but the effects in individual patients were variable. Both substantial increases or decreases in VE (delta VE) occurred, but these were accompanied by changes in VCO2 (delta VCO2) in the same direction. The effect of changes in VE on PaCO2 is shown to be almost completely cancelled by the concomitant changes in VCO2. Thus, the major portion of the change in PaCO2 was due to changes in VD/VT. We conclude that hyperoxic-induced hypercapnia is primarily due to impairment in gas exchange rather than to depression of ventilation. A reduced FEV1 appears to be a significant risk factor, whereas indices of respiratory drive are not likely to play a major role.     

Acid-base and ventilatory adaptation in conscious dogs during chronic hypercapnia

Ventilation and cisternal cerebrospinal fluid (CSF) and arterial acid-base balance were measured in awake dogs during air control and from 1 h to 26 days of breathing 5% CO2 in air. Ventilation increased 4-fold during acute hypercapnia and then declined to a minimum at 5-10 days. Between 1-3 days and 16-26 days of hypercapnia ventilation was relatively stable at 2.5 times control. [HCO3-]CSF increased rapidly by 12 h of hypercapnia and in the steady-state [HCO3-]CSF was correlated with PCSFCO2. Between 1 h and 1.5 days of hypercapnia, increase in [HCO3-]CSF was also correlated with increase in [NH3]CSF. Despite increase in [HCO3-]CSF, there was no compensation of [H+]CSF throughout 26 days of hypercapnia. Hydrogen ion may have contributed to the control of ventilation during chronic hypercapnia since ventilation was correlated with [HCO3-]a and [HCO3-]CSF. However, a relationship between ventilation and [H+] of arterial blood and CSF during chronic hypercapnia was relatively poor or absent. Ventilatory adaptation to chronic hypercapnia could not be related to metabolism or to [NH3]CSF. The mechanism(s) by which the increase in PCO2 during chronic respiratory acidosis results in sustained elevation of ventilation remains to be resolved.     

Intravascular membrane oxygenation and carbon dioxide removal--a new application for permissive hypercapnia?

Pressure limited ventilation or "lung rest" may prevent further exacerbation of acute lung injury from high airway pressures. A therapeutic goal of an intracorporeal oxygenation and carbon dioxide removal device (IVOX) is reduction of airway pressures. We noted increased IVOX CO2 removal as mixed venous CO2 increased in experimental animals. However, we recognize the limited clinical utility of removing approximately 30% of venous CO2. Therefore, intentional hypoventilation to limit airway pressures (mild permissive hypercapnia) was used in 5 patients with respiratory failure, and again we noted improved CO2 removal with increasing mixed venous CO2 concentrations. Preliminary calculations demonstrate that a CO2 gradient of approximately 70 mm Hg is needed to remove 100 ml CO2/min. The use of more aggressive permissive hypercapnia protocols with IVOX may permit further reduction in airway pressure without problems of severe respiratory acidosis.     

Pulmonary receptor chemosensitivity and the ventilatory response to inhaled CO2 in the turtle.

Ventilatory responses of unanesthetized turtles to changes in the intrapulmonary CO2 content of a vascularly isolated and an intact lung were measured during spontaneous breathing. The hyperpnea associated with inhalation of CO2 by the vascularly isolated lung was 19% of that associated with inhalation of CO2 by the intact lung. Transection of the vagus nerve supplying the isolated lung abolished this response. We conclude that both inhibition of pulmonary stretch receptor discharge with increasing levels of FICO2 and a functional increase in central inspiratory volume threshold during hypercapnia contribute to tidal volume increases following CO2 inhalation in normal animals. The major component of the ventilatory response of intact turtles to increasing levels of FICO2, however, was an increase in respiratory frequency. When CO2 was inspired only by the vascularly isolated lung the increase in respiratory frequency was only 21% of that recorded when the same levels of CO2 were inspired by the intact lung. Thus the ventilatory response of turtles to increasing levels of FICO2 is primarily dependent upon concomitant hypercapnia.     

The effects of hypercapnia and hypoxia on single hypoglossal nerve fiber activity

The respiratory related modulation of hypoglossal nerve activity has been studied at the single fiber level in cats under hyperoxic hypercapnia and hypoxic conditions and their conduction velocities determined. Changes in fiber activity were compared to simultaneous changes occurring in phrenic activity. Three different kinds of discharge patterns were observed: (a) inspiratory, (b) phasic activity during both inspiration and expiration, and (c) continuous random activity with no respiratory modulation. These fibers could be grouped into three categories according to their pattern of discharge during CO2 breathing. Type I fibers, mean conduction velocity of 30.0 m/sec, exhibited only an inspiratory phasic discharge during 100% O2 breathing. Their discharge frequency increased rapidly with higher levels of CO2 and hypoxia. Type II fibers, mean conduction velocity of 36.7 m/sec, had three different kinds of inspiratory-expiratory discharge patterns during 100% O2 breathing. With increasing hypercapnia or hypoxia fibers of this group discharged phasically during inspiration and discharge at low frequency during expiration. Type III fibers had a non phasic discharge pattern at 100% O2 breathing and at all levels of CO2 tested (up to 10%). Discharge frequency rose during CO2 rebreathing and hypoxia, but the rate of increase was much less than Type I and Type II fibers. Their mean conduction velocity was 41.3 m/sec. The inspiratory activity of Type I and II fibers increased their activity more than the phrenic during hypercapnia and hypoxia. Type II and Type III fibers are responsible at least in part for the tonic activity of the nerve.     

The effects of hypercapnia and cooling the ventral medullary surface on capsaicin induced respiratory reflexes

The effect of right atrial (RA) injection of 3 micrograms/kg capsaicin on phrenic, hypoglossal and recurrent laryngeal activities was studied in chloralose anesthetized, paralyzed and artificially ventilated cats. Within 2 sec following capsaicin injection, the phrenic and hypoglossal activities completely disappeared (apnea), while the recurrent laryngeal activity markedly increased. Similar responses were also obtained with RA injection of phenyldiguinide (PDG), suggesting that the respiratory responses of both drugs are essentially similar. Sino-aortic denervation did not affect the capsaicin induced respiratory responses. Bilateral vagotomy abolished the responses, suggesting that vagal sensory receptors are responsible for the reflex effects. Hyperoxic hypercapnia (3 and 7% CO2 in O2) reduced the apneic duration of phrenic and hypoglossal nerves. The magnitude of the recurrent laryngeal excitation was decreased during CO2 breathing. Graded focal cooling of the intermediate area (Is area) of the ventral medullary surface (to inhibit central chemoreceptor activity) significantly prolonged capsaicin induced apneic duration of hypoglossal nerves more than the phrenic. The recurrent laryngeal responses, however, were unaffected by cooling of the ventral medullary surface. The results show that capsaicin and PDG, presumably by stimulating C fibers, affect cranial nerves as well as the phrenic. The reflex responses to C fiber stimulation seem to be altered by intervention which stimulate (hypercapnia) or depress (Is cooling) 'central chemoreceptors.'     

The influence of renal sympathetic nerves on renal hemodynamic and renin responses during hypercapnia in dogs

Studies were conducted in anesthetized dogs to examine the influence of the renal sympathetic nerves on renal hemodynamic and renin responses during controlled hypercapnia. The dogs were subjected to unilateral denervation and tested for their responses to hypercapnia induced by inhalation of 15% CO2 in air. Simultaneous measurements of the responses from both the denervated and innervated kidneys allowed an assessment of the influence of the renal nerves on the responses during acute hypercapnia. The data indicate that reductions in renal blood flow and glomerular filtration rate and increases in renin of the renal vein during respiratory acidosis are dependent, in part, on the presence of intact renal nerves. Other factors, however, are probably also present.     

The respiratory neuromuscular response to hypoxia, hypercapnia, and obstruction to airflow in asthma.

In chronic obstructive pulmonary disease (COPD), the neuromuscular response to an acute increase in airflow produced by external flow resistive loads (FRL) is impaired. The present study compared the response to FRL of 15 subjects with airway obstruction due to asthma and that of 15 normal subjects. FRL were applied during progressive hypercapnia and isocapnic hypoxia produced by rebreathing techniques to permit the response to be assessed at the same degree of CO2 or O2 drive. The neuromuscular response to FRL was assessed from the airway occlusion pressure developed 100 msec after the onset of inspiration (P100), as well as ventilation. During control rebreathing, ventilatory responses to hypercapnia (ratio of change in minute ventilation to change in PCO2, delta VE/delta PCO2) and hypoxia (ratio of change in VE to the change in percentage of O2 saturation, delta VE/deltaSO2) were the same in asthmatic and normal subjects despite differences in the mechanics of breathing. The P100 response to hypercapnia delta P100/delta PCO2) and hypoxia (delta P100/delta SO2) as well as absolute P100 at any given degree of O2 and CO2 drive was greater during control rebreathing in asthmatics than in normal subjects (P less than 0.05). FRL values of 9 and 18 cm H2O per L per sec applied during either hypercapnia or hypoxia increased the occlusion pressure to a greater extent in asthmatics than in normal subjects. Methacholine-induced bronchoconstriction was used to test the effect of acute airway obstruction on the response to FRL. Bronchoconstriction was associated with an increase in the P100 response to hypercapnia and to FRL, despite increases in lung volume and decreases in inspiratory muscle force. We conclude that: (1) asthmatics with airway dysfunction have an increased nonchemical drive to breathe mediated at least in part by sensory receptors in the airways; (2) asthmatics with airway obstruction respond supernormally to acute changes in resistance to airflow, unlike subjects with COPD. The failure of COPD subjects with prolonged airway obstruction to respond to FRL may be due to adaptation of the sensory mechanisms that respond to changes in airway resistance.     

Blood flow to the respiratory muscles during hypercapnic hyperpnoea in the newborn lamb

The blood flow to the diaphragm, external and internal intercostal muscles, abdominal oblique muscles, and other rib-cage and abdominal muscles was measured, using radio-labelled microspheres, in 6 newborn lambs quietly breathing in air and during 3 different levels of CO2 induced hypercapnic hyperpnoea (inspired gas containing 4%, 5.5%, or 7% CO2). We also calculated the oxygen uptake of the diaphragm (VO2di). While the lambs were breathing air diaphragmatic blood flow (Qdi, 38.2 +/- 4.0 SEM ml.min-1.100 g-1) was similar to external intercostal muscle blood flow (Qei, 37.1 +/- 8.1 ml.min-1.100 g-1), and both were greater than internal intercostal muscle blood flow (Qii, 24.8 +/- 6.1 ml.min-1.100 g-1; P less than 0.05). During hyperpnoea Qdi, Qei, and Qii were augmented with Qdi equal to 200.1 +/- 12.2 ml.min-1.100 g-1 in 7% CO2 and Qei equal to 88.4 +/- 14.1 ml.min-1.100 g-1 in 7% CO2 (Qdi was greater than Qei, P less than 0.01). Qii was 40.7 +/- 5.6 ml.min-1.100 g-1 in 7% CO2 being less than Qdi (P less than 0.01) and Qei (P less than 0.05). The abdominal oblique muscles also had augmented flow in response to hyperpnoea. The level of hypercapnia that resulted in an augmentation of Qdi (5.5% inspired CO2) was lower than that which augmented Qei and Qii (7% inspired CO2). VO2di was linearly related to Qdi (r = 0.98). Our results suggest that in the newborn lamb the diaphragm is the dominant respiratory muscle in response to hypercapnia.