Airway CO2 Research Articles

High flow nasal cannula: Influence of gas type and flow rate on airway pressure and CO2 clearance in adult nasal airway replicas

BackgroundHigh flow nasal cannula therapy is a form of respiratory support which delivers high flow rates of heated, humidified gas to the nares via specialized cannula. Two primary mechanisms of action attributed to the therapy are the provision of positive airway pressure as well as clearance of CO2-rich exhaled gas from the upper airways. MethodsPhysiologically accurate nose-throat airway replicas were connected at the trachea to a lung simulator, where CO2 was supplied to mimic the CO2 content in exhaled gas. Cannula delivered either air, oxygen or heliox (80/20%volume helium/oxygen) to the replicas at flow rates ranging from 0 to 60 l/min. Five replicas and three cannulas were compared. Tracheal pressure and CO2 concentration were continuously measured. The lung simulator provided breaths with tidal volume of 500 ml and frequency of 18 breaths/min. Additional clearance measurements were conducted for tidal volume and breathing frequency of 750 ml and 27 breaths/min, respectively. FindingsCannula flow rate was the dominant factor governing CO2 concentration. Average CO2 concentration decreased with increasing cannula flow rate, but above 30 L/min this effect was less pronounced. Tracheal positive end-expiratory pressure increased with flow rate and was lower for heliox than for air or oxygen. A predictive correlation was developed and used to predict positive end-expiratory pressure for a given cannula size as a function of supplied flow rate and occlusion of the nares. InterpretationCompared with administration of air or oxygen, administration of heliox is expected to result in similar CO2 clearance from the upper airway, but markedly lower airway pressure.

Effect of Face Mask Design and Bias Flow on Rebreathing During Noninvasive Ventilation

Noninvasive ventilation (NIV) is used to treat respiratory failure because it reduces the risks of endotracheal intubation and postextubation respiratory failure. A wide range of different interfaces is available, but concerns exist about rebreathing. This study evaluated a total face mask with a 2-limb ventilation circuit and separate access for inflow and outflow gas, which was developed to reduce rebreathing. In a bench test, a standard total face mask (with a single connector to the ventilation circuit) and the modified total face mask were applied to a mannequin connected to an active breathing simulator. A known CO2 flow (V̇CO2 ) was delivered to the mannequin's trachea. We tested the following settings: CPAP with the mechanical PEEP valve set at 8 cm H2O (with 60 and 90 L/min continuous flow) and pressure support of 6 and 12 cm H2O (with 2 and 15 L/min bias flow). The settings were tested at simulated breathing frequencies of 15 and 30 breaths/min and with V̇CO2 of 200 and 300 mL/min. The active simulator generated a tidal volume of 500 mL. Airway pressure, air flow, CO2 concentration, and CO2 flow as the product of air flow and CO2 were recorded. The mean volume of CO2 rebreathed and the minimum CO2 inspiratory concentration were significantly lower with the modified mask than with the standard mask. The 15 L/min bias flow significantly decreased rebreathing with the DiMax0 mask, whereas it had no effect with the traditional mask. A face mask with a two-limb ventilation circuit and separate access for inflow and outflow gas reduces rebreathing during NIV. The addition of bias flow enhances this effect. Further studies are required to verify the clinical relevance.

Effects on the pulmonary hemodynamics and gas exchange with a speed modulated right ventricular assist rotary blood pump: a numerical study

Rotary blood pumps (RBPs) are the newest generation of ventricular assist devices. Although their continuous flow characteristics have been accepted widely, more and more research has focused on the pulsatile modulation of RBPs in an attempt to provide better perfusion. In this study, we investigated the effects of an axial RBP serving as the right ventricular assist device on pulmonary hemodynamics and gas exchange using a numerical method with a complete cardiovascular model along with airway mechanics and a gas exchange model. The RBP runs in both constant speed and synchronized pulsatile modes using speed modulation. Hemodynamics and airway O2 and CO2 partial pressures were obtained under normal physiological conditions, and right ventricle failure conditions with or without RBP. Our results showed that the pulsatile mode of the RBP could support right ventricular assist to restore most hemodynamics. Using speed modulation, both pulmonary arterial pressure and flow pulsatility were increased, while there was only very little effect on alveolar O2 and CO2 partial pressures. This study could provide basic insight into the influence of pulmonary hemodynamics and gas exchange with speed modulated right ventricular assist RBPs, which is concerned when designing their pulsatile control methods.

Versatile breath sampler for online gas sensor analysis

Breath analysis is an emerging field in gas sensor technology aiming for the next generation of hand-held and non-invasive medical diagnostic and monitoring devices. Often unrecognized, however, remains the importance of correct and standardized breath sampling, a prerequisite for meaningful breath analysis. Here, a simple and modular sampler design is presented that enables monitored breath extraction for reproducible and online gas sensor analysis. It consists of an inlet to monitor airway pressure and CO2 concentration online during exhalation and a tube to capture and buffer (20–40 s) end-tidal breath to meet prolonged exposure requirements of some gas sensors. Demonstrated on the example of acetone, a breath marker for fat burn, this sampler is suitable for continuous (refreshment ∼100 s) and reproducible online breath analysis, as validated by state-of-the-art proton transfer reaction time-of-flight mass spectrometry (PTR-TOF-MS). Most interestingly, even short breath pulses down to 1 s result in reliable determination of end-tidal acetone concentrations while it is severely underestimated when sampling is not done properly. Finally, this sampler is applied to detect breath acetone with solid-state Si-doped WO3 sensors in humans. The proposed sampler can be connected to various types of gas sensors and applied flexibly for other breath markers (including isoprene, ethanol and methanol).

Evaluation of a method for isocapnic hyperventilation: a clinical pilot trial.

Isocapnic hyperventilation (IHV) is a method that shortens time to extubation after inhalation anaesthesia using hyperventilation (HV) without lowering airway CO2 . In a clinical trial on patients undergoing long-duration sevoflurane anaesthesia for major ear-nose-throat (ENT) surgery, we evaluated the utility of a technique for CO2 delivery (DCO2 ) to the inspiratory limb of a closed breathing circuit, during HV, to achieve isocapnia. Fifteen adult ASA 1-3 patients were included. After end of surgery, mechanical HV was started by doubling baseline minute ventilation. Simultaneously, CO2 was delivered and dosed using a nomogram developed in a previous experimental study. Time to extubation and eye opening was recorded. Inspired (FICO2 ) and expired (FETCO2 ) CO2 and arterial CO2 levels were monitored during IHV. Cognition was tested pre-operatively and at 20, 40 and 60 min after surgery. A DCO2 of 285 ± 45 ml/min provided stable isocapnia during HV (13.5 ± 4.1 l/min). The corresponding FICO2 level was 3.0 ± 0.3%. Time from turning off the vaporizer (1.3 ± 0.1 MACage) to extubation (0.2 ± 0.1 MACage) was 11.3 ± 1.8 min after 342 ± 131 min of anaesthesia. PaCO2 and FETCO2 remained at normal levels during and after IHV. In 85% of the patients, post-operative cognition returned to pre-operative values within 60 min. In this cohort of patients, a DCO2 nomogram for IHV was validated. The patients were safely extubated shortly after discontinuing long-term sevoflurane anaesthesia. Perioperatively, there were no adverse effects on arterial blood gases or post-operative cognition. This technique for IHV can potentially be used to decrease emergence time from inhalation anaesthesia.

Nasal highflow eliminates CO2 from lower airways

IntroductionNasal highflow (NHF) has a growing evidence in treatment of hypoxemic respiratory failure. There are preliminary data available about use in acute and chronic type-II-respiratory failure. Unfortunately underlying mechanisms of NHF are not well understood. Increase in airway pressure seems too small to explain the observed reduction in hypercapnia. Most interesting effects are wash-out of upper airways and reduction of functional dead space. There are no data available about a wash-out of lower airways. MethodsWe established a sheep lung model to evaluate CO2-wash-out in lower airways. Therefore we placed measuring and insufflation catheter in maximal expanded lung. The lung was not ventilated in order to minimize influence of CO2 rebreathing and increase in airway pressure. Airway pressure and CO2 values were measured in lower airways and in tracheal space. ResultsCO2 was decreased by NHF in lower airways and in tracheal space. Changes in CO2 were flow dependent. There was also an increase in airway pressure in these settings. ConclusionsNHF is able to decrease CO2 in lower airways in a flow-dependent manner. This effect is independent of an increase in airway pressure and CO2-rebreathing. So wash-out is an important reason for efficiency of NHF in decreasing hypercapnia.

An appropriate inspiratory flow pattern can enhance CO2 exchange, facilitating protective ventilation of healthy lungs

BackgroundIn acute lung injury, CO2 exchange is enhanced by prolonging the volume-weighted mean time for fresh gas to mix with resident alveolar gas, denoted mean distribution time (MDT), and by increasing the flow rate immediately before inspiratory flow interruption, end-inspiratory flow (EIF). The objective was to study these effects in human subjects without lung disease and to analyse the results with respect to lung-protective ventilation of healthy lungs. MethodsDuring preparation for intracranial surgery, the lungs of eight subjects were ventilated with a computer-controlled ventilator, allowing breath-by-breath modification of the inspiratory flow pattern. The durations of inspiration (TI) and postinspiratory pause (TP) were modified, as was the profile of the inspiratory flow wave (i.e. constant, increasing, or decreasing). The single-breath test for CO2 was used to quantify airway dead space (VDaw) and CO2 exchange. ResultsA long MDT and a high EIF augment CO2 elimination by reducing VDaw and promoting mixing of tidal gas with resident alveolar gas. A heat and moisture exchanger had no other effect than enlarging VDaw. A change of TI from 33 to 15% and of TP from 10 to 28%, leaving the time for expiration unchanged, would augment tidal elimination of CO2 by 14%, allowing a 10% lower tidal volume. ConclusionsIn anaesthetized human subjects without lung disease, CO2 exchange is enhanced by a long MDT and a high EIF. A short TI and a long TP allow significant reduction of tidal volume when lung-protective ventilation is required. Clinical trial registrationNCT01686984.

Isocapnic augmented exercise ventilation in congestive heart failure: effect of apparent (real‐feel) metabolic CO2 load in respiratory chemosensing

CO2 breathing introduces an airway CO2 load that clogs the mechanism of pulmonary CO2 elimination, inevitably causing hypercapnia when the inspired PCO2 is high enough to prohibit isocapnia regardless of the ventilation level. The control law (Comroe's law) evidencing such self‐imposed permissive hypercapnia has been dubbed “CO2 tolerance curve” (rather than CO2 response curve) by Fenn and Craig (1963). By contrast, muscular exercise introduces a metabolic CO2 load that is readily countered with proportionate increase in ventilation without CO2 retention. Isocapnia from rest to exercise is maintained by the controller with the exercise ventilatory response being well adjusted for corresponding decrease in anatomical VD/VT, resulting in a small positive Y‐intercept in the ventilation vs. metabolic CO2 output relationship (Whipp's law). Remarkably, patients with congestive heart failure compensate for their abnormally large alveolar VD/VT with augmented ventilation, maintaining isocapnia at rest and during exercise without CO2 retention despite their decreased exercise tolerance. Thus, the control law governing isocapnic exercise ventilatory response is not merely determined by metabolic CO2 output per se but is responsive to an apparent (real‐feel) metabolic CO2 load that also accounts for the adverse effect of physiological VD/VT on pulmonary CO2 elimination. (Supported by HL067966 and RR028241)

Hypercapnic augmented exercise ventilation in dead space loading: effect of virtual (illusory) airway CO2 load in respiratory chemosensing

Patients with congestive heart failure (CHF) suffer increased alveolar dead space, yet they demonstrate compensatory augmentation of ventilation such that arterial PCO2 remains remarkably normal (isocapnic) from rest to maximal exercise despite their decreased exercise tolerance. This paradoxical effect suggests that the control law governing isocapnic exercise ventilatory response is not merely determined by metabolic CO2 output per se but is responsive to an apparent (real‐feel) metabolic CO2 load that also accounts for the adverse effect of physiological VD/VT on pulmonary CO2 elimination. By contrast, healthy individuals subjected to dead space loading with increased series dead space also experience augmented ventilation at rest and during exercise as with increased alveolar dead space in CHF, but the resultant response is hypercapnic instead of isocapnic, as with CO2 breathing. The ventilatory effects of dead space loading are therefore similar to those of increased alveolar dead space and CO2 breathing combined. These observations are consistent with the hypothesis that the increase in total VD/VT in dead space loading adds to the apparent metabolic CO2 load as with increased alveolar dead space in CHF, but this is through rebreathing of CO2 in dead space gas thus creating a virtual (illusory) airway CO2 load in respiratory chemosensing that mimics CO2 breathing. (Supported by HL067966 and RR028241)

Two potassium‐channel blockers differently affect afferent signal of avian intrapulmonary chemoreceptors

We investigated the roles of IA and IDR potassium channels in action potential generation of avian intrapulmonary chemoreceptors (IPC). IPC are CO2‐sensing vagal afferents in avian lungs, with fast and inverse response to airway CO2 partial pressure (PCO2). Discharge frequency of a single cell IPC was recorded in 11 anesthesized, ventilated ducks. FICO2 was electronically controlled. Intravenous injection of voltage‐gated potassium‐channel blockers 4‐aminopyridine (4AP) or tetraethylammonium (TEA) disrupted the response to CO2. 4AP blocks the transient K current (IA), and TEA blocks the delayed rectifier current (IDR). A dose of 7 mg/kg of 4AP eliminated both phasic and tonic IPC sensitivities to FICO2, resulting in constant high‐frequency AP firing (p < 0.05). Low dosages of TEA had no effect on that sensitivity, although at high TEA doses (>; 20 mg/kg) the phasic (p < 0.05) but not tonic response to a step change in FICO2 disappeared. Our results suggest that transient voltage gated K channels (affected at low dosages of 4AP and high dosages of TEA) are involved in setting resting membrane potential in IPC. If IA in IPC is increased by CO2/H+ as it is in slowly adapting mammalian pulmonary stretch receptors, IA channel modulation by CO2/H+ may also underlie CO2 signal transduction in avian IPC. Support: NIH 2R15HL087269

Novel Portable Device Measures Preoperative Patient Metabolic Gas Exchange

Indirect calorimetry (IC), the measurement of airway CO2 elimination (VCO2), O2 [corrected] uptake (VO2) [corrected], and respiratory exchange ratio (RER = VCO2/VO2), is a noninvasive modality for the assessment of body metabolism. In anesthesia, IC can signal critical events and onset of acute metabolic derangements. We have previously demonstrated the accuracy and precision of a new IC measurement system designed for mechanically ventilated patients, comprised of a new clinical bymixer, fast response humidity and temperature sensor, and a flowmeter. However, measurement of IC during spontaneous breathing is challenging because of unstable tidal volume, frequency, and functional residual capacity (FRC). A new device for IC measurements, designed specifically for spontaneous breathing, was validated against a metabolic lung simulator bench setup. In a second study, the same device was used to conduct preoperative measurements of VCO2 and VO2 in 15 patients. Our measurements showed excellent correlation and agreement with metabolic lung simulator values: The average (+/-SD) percent error for airway VCO2 was -4.7% +/- 3.31%; the average (+/-SD) percent error for airway VO2 was -0.30% +/- 5.25%. Average values of VCO2 and VO2 in the patient study (3.01 +/- 0.56 and 3.44 +/- 0.69 mL x kg(-1) x min(-1), respectively) were in agreement with previously reported values. We have shown that the new, portable bymixer-flow device, using a bymixer and a fast response humidity sensor, provided accurate and convenient bedside measurement of VCO2 and VO2. We believe that it can contribute in the future to preoperative assessment and baseline reference value for perioperative management.

A learned component of the ventilatory response to exercise in man.

The ventilatory response to mild-to-moderate exercise in humans is isocapnic, or 'error-free'. It has been suggested that this response is learned over many repetitions of exercise through the process of minimising any deviations from normal in the blood gas tensions, as sensed by the chemoreceptors. However, relatively limited training programmes have failed to produce any convincing evidence in humans that forcibly altering the blood gas tensions during repeated periods of exercise alters the subsequent steady-state ventilatory response to exercise. In this study, eight healthy young subjects were exposed, over a 7 day training period, to a total of 70 repeated bouts of exercise paired with a simultaneous airway CO2 load to stimulate the chemoreceptors (protocol EX + CO2). The ventilatory response to exercise was measured before and after training to determine whether it had been modified. Two further training protocols were undertaken as controls. One employed repeated exercise without an airway CO2 load, and the other employed repeated airway CO2 loading without exercise. On the 1st and 2nd days following training with protocol EX + CO2, end-tidal PCO2 was regulated at a lower level during steady-state exercise than following training with the control protocols and than before training (mean +/- S.E.M. reduction in end-tidal PCO2 = 1.32 +/- 0.36 Torr, ANOVA, P < 0.05). In contrast to previous studies, this finding demonstrates that the steady-state ventilatory response to exercise can be modified by a prior period of altered chemoreception during exercise. This suggests that ventilation is matched to metabolic rate during exercise by a mechanism that involves learning and memory.

Effects of positive end-expiratory pressure increments can be predicted by computer simulation based on a physiological profile in acute respiratory failure.

We examined whether computer simulation predicts airway pressures after increments of positive end-expiratory pressure (PEEP) in acute respiratory failure. Prospective, nonrandomized comparative trial in an intensive care unit of a university hospital. Twelve consecutive acute respiratory failure patients. INTERVENTIONS. PEEP increments from 0 to 2.5, 5, 7.5, 10, and 15 cm H(2)O. A physiological profile comprising values for compliance, respiratory resistance and CO(2 )elimination as a function of tidal volume was established from a recording of ordinary breaths prior to increments of PEEP. Airway pressures and CO(2 )elimination were measured 30 s after resetting, pressures also after 10 min. Values from simulation of the resetting, based on the profile, were compared to measured values. The profiles indicated vast differences in physiology between the 12 subjects. Errors of simulation of airway pressures were nonsignificant or trivial up to PEEP levels of 10 cm H(2)O (95% of errors <3 cm H(2)O). After 10 min plateau pressure averaged 1.5 cm H(2)O lower than 30 s after resetting. At increments to PEEP 7.5, 10, and 15, CO(2 )elimination fell by on average 4%, 8%, and 11%, respectively. As tidal volume and respiratory rate was unchanged this was not predicted. On the basis of a simple lung model, simulation predicted effects of moderate increments of PEEP on airway pressures in patients with complex physiology.

Computer-aided ventilator resetting is feasible on the basis of a physiological profile.

Ventilator resetting is frequently needed to adjust tidal volume, pressure and gas exchange. The system comprising lungs and ventilator is so complex that a trial and error strategy is often applied. Comprehensive characterization of lung physiology is feasible by monitoring. The hypothesis that the effect of ventilator resetting could be predicted by computer simulation based on a physiological profile was tested in healthy pigs. Flow, pressure and CO2 signals were recorded in 7 ventilated pigs. Elastic recoil pressure was measured at postinspiratory and post-expiratory pauses. Inspiratory and expiratory resistance as a function of volume and compliance were calculated. CO2 elimination per breath was expressed as a function of tidal volume. Calculating pressure and flow moment by moment simulated the effect of ventilator action, when respiratory rate was varied between 10 and 30 min(-1) and minute volume was changed so as to maintain PaCO2. Predicted values of peak airway pressure, plateau pressure, and CO2 elimination were compared to values measured after resetting. With 95% confidence, predicted pressures and CO2 elimination deviated from measured values with < 1 cm H2O and < 6%, respectively. It is feasible to predict effects of ventilator resetting on the basis of a physiological profile at least in health.

Laryngeal-receptor responses to phasic CO2 in anesthetized cats.

We compared the effects of CO2 applied continuously and during expiration on laryngeal-receptor activity in paralyzed, artificially ventilated and nonparalyzed, spontaneously breathing cats by using an isolated larynx, artificially ventilated to approximate a normal respiratory cycle. The majority of quiescent negative-pressure and all cold receptors were excited by 5 and 9% CO2 applied both continuously and during expiration. In general, quiescent positive-pressure, tonic negative-pressure, and tonic positive-pressure receptors were inhibited by 5 and 9% CO2 applied continuously and during expiration. There were no significant differences between responses to 5 and 9% CO2 or to continuous and expired CO2 or between paralyzed and nonparalyzed preparations. In conclusion, laryngeal receptors respond to changes in CO2 concentration occurring during a normal respiratory cycle. Because laryngeal-receptor stimulation exerts reflex effects on ventilation and upper airway muscle activity, these results suggest that airway CO2 plays a role in reflex regulation of breathing and upper airway patency.

Mechanical model testing of rebreathing potential in infant bedding materials

Rebreathing of expired air may be a lethal hazard for prone sleeping infants. This paper describes a mechanical model to simulate infant breathing, and examines the effects of bedding on...

Chest vibration redistributes intra-airway CO sub 2 during tracheal insufflation in ventilatory failure

To determine if high-frequency external chest wall vibration added to low flow intratracheal fresh gas insufflation alters the intra-airway CO2 distribution and the resistance to CO2 transport from the lungs. Prospective study. Experimental laboratory. Six adult anesthesized and paralyzed mongrel dogs (mean weight 24.3+/- 4.4 kg). Dogs were ventilated by three methods: a) intermittent positive pressure ventilation; b) intermittent positive pressure ventilation with tracheal insufflation of fresh gas (FIO2 of 0.4) flowing at 0.15 L/kg/min through a catheter positioned at the carina; and c) intermittent positive pressure ventilation with tracheal insufflation and with external high-frequency chest wall vibration of the dependent hemithorax. We measured arterial blood gas values as an index of global gas exchange, and intrapulmonary airway CO2 concentrations as an index of local gas exchange. Intra-airway CO2 concentrations along the axis of the airways were measured via a sampling catheter. Airway axial concentration profiles were constructed and resistances to gas transport were calculated from the measured data. Vibration increased intraluminal CO2 concentrations from 1.1% to 2.5% mouthward of the insufflation catheter tip. Peak resistance to CO2 transport decreased by 65% during vibration relative to the insufflation-only value. Vibration displaced peak transport resistance from second- to fourth-generation airways. Global gas exchange improves during ventilation by chest wall vibration with low flow insufflation. Local gas exchange in the central airways is also improved due to increased intraluminal mixing and CO2 elimination. This ventilation technique may confer therapeutic advantages over conventional mechanical ventilation in the treatment of ventilatory failure.

Ventilatory and upper airway muscle responses to upper airway CO2 in anaesthetized neonatal guinea-pigs.

In adult animals, CO2 confined to the isolated upper airway (UA) lumen reflexly inhibits breathing and excites genioglossus muscle activity (Nolan et al, 1990) by affecting the activity of CO2-sensitive receptors in the superior laryngeal nerves (Bradford et al, 1993). These effects may be important in the regulation of breathing and UA patency since the genioglossus is a major UA dilator muscle. There is some evidence that superior laryngeal nerve (SLN) reflexes are more potent in neonates than in adults (Al-Shway & Mortola, 1982; Boggs & Bartlett Jr, 1982). Therefore, the present investigation examines the effects of UA CO2 on ventilation and geniohyoid (GH) muscle activity in neonatal Guinea-pigs. We chose to study the GH muscle because of its ease of access and because, like the genioglossus, it exerts an important influence on UA patency. Contraction of the GH causes anterior displacement of the hyoid bone and stiffening of the hypopharynx (Brouillette & Thach, 1979), a site which is vulnerable to collapse (Shepard Jr et al, 1991).

CO2-sensitive olfactory and pulmonary receptor modulation of episodic breathing in bullfrogs.

Breathing was monitored during normocarbia, hypercarbia (6% CO2 in air), and the period immediately after the return to normocarbic conditions in intact, olfactory-denervated, and vagotomized bullfrogs. In intact frogs, ventilation increased during hypercarbia, but the breathing pattern remained episodic. Immediately upon return to air, there was a further paradoxical increase in breathing frequency, and breathing became continuous in most frogs. Results obtained from animals after olfactory receptor denervation indicate that tonic stimulation of olfactory receptors by airway CO2 inhibited breathing during hypercarbia. Measurements of the kinetics of changes in airway and arterial blood CO2 levels support the suggestion that the sudden release of this inhibition on the return to normocarbic conditions was responsible for the posthypercarbic hyperpnea. Vagotomy increased ventilation during normocarbia. Hypercarbia now caused a change in breathing pattern but had no net effect on total ventilation, suggesting that pulmonary vagal feedback inhibited ventilation during normocarbia but stimulated ventilation during hypercarbia. Although olfactory and pulmonary receptor feed-back shape the breathing pattern, they were not responsible for initiating or terminating the episodes of breathing.

Intrapulmonary CO2-rise time and ventilation in ducks.

Many investigators have reported finding intrapulmonary chemoreceptors (IPCs) in several species of birds and reptiles; however, the role of IPCs in ventilatory control in birds has not been identified. This study was undertaken to assess the role of intrapulmonary CO2 dynamics on the control of breathing. Ducks were unidirectionally ventilated, and their breathing was monitored with a pneumotachograph connected to a personal computer. A gas-mixing system controlled by the computer adjusted the rate of rise of airway CO2 concentration ([CO2]). On inspiration, the computer removed CO2 from the ventilating gas for 1.5 s, followed by a controlled [CO2] rise. Breathing frequency was directly related to the rate of rise of airway [CO2]. Tidal volume, however, was not correlated with CO2-rise time but was related to the peak airway [CO2]. This response is likely mediated by IPCs because preventing airway [CO2] from falling during inspiration immediately altered that breath. An increase in CO2 production (as in exercise) will lead to an increase in the rate of CO2 excretion into the lung. The resulting alteration of breathing frequency would thus maintain acid-base balance. The observed response, therefore, may represent a link between ventilation and metabolism.