Search

This technical note is presenting and discussing a severe limitation of the breath-by-breath (BBB) determination of pulmonary gas exchange routinely used as a surrogate for the metabolic gas exchange rate at rest and during exercise. We are presenting the view that continuous airway gas sampling at the mouth used for the determination of O2 and CO2 content is inaccurate in the range of a low-to-medium expiratory flows, due to the discrepancy between the cross section surfaces of the sampling line and the expiratory tube (mask or mouth piece). This difference results in the sampling of a mixed exhaled gas at low expiratory flow, for which any temporal relationship between the instantaneous expired CO2 and O2 signals and instantaneous expiratory flow is lost. Further analysis of this mechanism points to the difference between the exhalation (ve) and sampling (vs) gas velocities, where the ratio of ve/vs must be equal or higher than 1 to enable proper analysis of the respiratory gas concentrations during exhalation. Moreover, the above requirement is particularly crucial in monitoring the tidal concentration of the respiratory gases in experiments with small animals.

to the editor: I read the Viewpoint article by Haouzi ([1][1]) entitled “Precedence and autocracy in breathing control” with great interest. The author makes a strong and informative case for the complexity of the mechanisms involved in the control of breathing. But he makes conventional error

This review examines the evidence that skeletal muscles can sense the status of the peripheral vascular network through group III and IV muscle afferent fibers. The anatomic and neurophysiological basis for such a mechanism is the following: 1) a significant portion of group III and IV afferent fibers have been found in the vicinity and the adventitia of the arterioles and the venules; 2) both of these groups of afferent fibers can respond to mechanical stimuli; 3) a population of group III and IV fibers stimulated during muscle contraction has been found to be inhibited to various degrees by arterial occlusion; and 4) more recently, direct evidence has been obtained showing that a part of the group IV muscle afferent fibers is stimulated by venous occlusion and by injection of vasodilatory agents. The physiological relevance of sensing local distension of the vascular network at venular level in the muscles is clearly different from that of the large veins, since the former can directly monitor the degree of tissue perfusion. The possible involvement of this sensing mechanism in respiratory control is discussed mainly in the light of the ventilatory effects of peripheral vascular occlusions during and after muscular exercise. It is proposed that this regulatory system anticipates the chemical changes that would occur in the arterial blood during increased metabolic load and attempts to minimize them by adjusting the level of ventilation to the level of muscle perfusion, thus matching the magnitudes of the peripheral and pulmonary gas exchange.

The influence of impeding muscle perfusion on the time course of ventilatory decline during recovery from electrically induced hindlimb contractions has been studied in 14 anesthetized dogs. When intravascular balloons, placed in abdominal aorta and inferior vena cava just rostral to the iliac bifurcation, were inflated at the cessation of contraction bout, minute ventilation (VE) was significantly reduced during recovery compared with control. The subsequent restoration of iliac circulation rapidly augmented VE, which peaked at the fifth breath after release, by an average of +4.97 L.min-1; VE then returned exponentially to resting (pre-contraction) level. Breathing 100% O2 did not affect the VE recovery pattern neither during iliac occlusion nor immediately after its release (the peak average delta VE = +4.42 L.min-1). When a local anesthetic (5% Lidocaine) was applied bilaterally to the regions of carotid bifurcation, systemic blood pressure was significantly increased and the VE response to both iliac occlusion and release were nearly abolished. The VE response to inhalation of 5% CO2 in air was not affected by this procedure, whereas the stimulation of VE with 2 mg i.v. bolus of NaCN was attenuated. When the local anesthetic was thoroughly washed out (and systemic blood pressure had returned to control level) the previously observed VE responses to iliac occlusion and release were restored. These results and analysis of the VE response timing (transits and latencies) suggest that the vascular rather than humoral effects or tissue 'metaboreception' modulate ventilatory recovery from muscular contractions; baroreception appears to be important in this process.

1. In response to an acute exercise-induced metabolic acidosis, the fall of arterial pH is constrained by the magnitude of the compensatory hyperventilation. To determine the role of the carotid bodies in this regulatory process, subjects performed prolonged (24 min) square-wave cycle ergometry from a background of unloaded cycling at inspired oxygen fractions (FI,O2) of 0.12 O2 (high carotid body gain), 0.21 O2 (normal carotid body gain) and 0.80 O2 (low carotid body gain). The work rates were selected to provide the same exercise intensity, despite the different inspirates; i.e. resulting in a constant increase in arterial blood [lactate] (delta [L-] approximately 4 mequiv l-1. 2. Ventilatory and pulmonary gas exchange variables were computed breath-by-breath and arterial blood was sampled at intervals throughout the tests and analysed subsequently for [lactate], [pyruvate], arterial partial pressures of oxygen and carbon dioxide (PO2, PCO2), pH, [bicarbonate] and [potassium]. 3. Hypoxia markedly reduced, and hyperoxia magnified, the transient decrease in arterial pH following exercise onset. However, there was a slow acid-base compensatory component, even when carotid chemosensitivity was suppressed by hyperoxia. We therefore conclude that, in humans, carotid body chemosensitivity plays a dominant role in constraining variations of arterial pH in response to the acute metabolic acidosis of heavy exercise, but that secondary-presumably central chemosensory-mechanisms subserve a slower compensatory role.

To the Editor : We have read with a lot of interest the paper by Oelberg et al. ([9][1]) in which the ventilatory effects of inflating proximal thigh cuffs to 45 Torr during isotonic quadriceps exercise were analyzed in 12 healthy subjects. In the last sentence of this paper, the authors conclude

To determine the role of cardiac reflexes in mediating exercise hyperpnea, we investigated ventilatory responses to treadmill exercise in seven calves with artificial hearts and seven controls. In both groups, the ventilatory responses were adequate for the metabolic demands of the exercise; this resulted in regulation of arterial PCO2 and pH despite the absence of cardiac output increase in the implanted group. In this group, there was a small but significant reduction of arterial PO2 by 4 +/- 3 Torr and a rise of blood lactate by 1.1 +/- 1 mmol/l. When cardiac output was experimentally increased in the implanted calves to a level commensurate with that spontaneously occurring in the control calves, ventilation was not affected. However, experimental reductions of cardiac output led to an immediate augmentation of exercise hyperpnea by 4.56 +/- 4.3 l/min and a further significant lactate increase of 1.2 +/- 1.22 mmol/l that was associated with a significant decrease in the exercise O2 consumption (0.32 +/- 0.13 l/min). These observations indicate that neither cardiac nor hemodynamic effects of increased cardiac output constitute an obligatory cause of exercise hyperpnea in the calf.

We have developed a method for simulating respiratory gas exchange for on-line calibration of metabolic measurement systems. It utilizes a pump which intakes a mixture of atmospheric air and a known flow of precision-analysed calibration gas (21% CO2, 79% N2). It expels the resulting mixture with flow wave form and profiles of gas concentration which closely resemble those of normal expiration. Control of the calibration mixture's inflow allows the investigator to set any desired metabolic rate regardless of the minute ventilatory rate. This separation of metabolic from ventilatory rates provides a stringent test of the computational performance of the respiratory gas exchange measurement systems. The apparatus can reproduce any range of respiratory and metabolic performance (currently ranging from 0.2-5 l.min-1 O2 uptake and CO2 output) with accuracy +/- 2%.

To test the hypothesis, previously suggested by Huszczuk et al. (1993), that distention of the peripheral microvascular network could, per se, stimulate ventilation, the ventilatory effects of papaverine-induced muscular vasodilation were studied in ten anaesthetized sheep. Because systemic action of papaverine may involve the arterial baro- and chemoreceptors, the animals were surgically prepared for a reversible isolation of the hindlimb circulation. Papaverine injection (1-2 mg/kg) into the arterial inflow of the isolated limbs provoked a 13 +/- 6 sec-delayed increase in VE by 1.8 +/- 0.2 L min-1 (p < 0.01) with a concomitant decrease in peripheral vascular resistance and no decrease in the systemic arterial blood pressure. Identical control injection into a jugular vein prior to the hindlimb circulatory separation yielded an increase of VE by 4.95 +/- 0.58 L min-1 with a latency of 21 +/- 2 sec and a coinciding moderate decrease of the systemic arterial pressure. The present data suggest that papaverine injection into the hindlimb circulation can stimulate ventilation independently of its possible effects on the arterial baro- or chemoreceptors, supporting the hypothesis that muscular vasodilation could contribute to the control of breathing through a neural link.

Ventilation and cardiac output subside gradually following cessation of exercise, which is commonly linked to the slow wash-out of materials from the recovering muscles. The effect of hindering the removal of the metabolic products of heavy cycle exercise on the kinetics of ventilation and gas exchange was studied in 5 subjects by occluding the femoral circulation with cuffs during the first 2 min of recovery (15 tests). Fifteen undisturbed recoveries served as controls. Compared to spontaneous recovery, circulatory obstruction induced an immediate (from the first breath) decrease in minute ventilation (VE), while end-tidal CO2 (PETCO2) as well as lactate and K+ in venous blood at forearm did not change significantly. A ventilatory deficit of 27 +/- 9 L was observed from the 2 min of occlusion. Following cuff deflation, VE rose 2-3 breaths after PETCO2 began to increase in every subject. The mechanisms of the normocapnic reduction of VE during occlusion, as well as the rise of ventilation following cuff release, are still unclear. However, these results argue against any significant role for hyperpnea-inducing intramuscular chemoreception, or point to muscular perfusion as a prerequisite of such a mechanism to operate.

Ventilatory responses to changes in right ventricular (RV) load were studied in spontaneous breathing anesthetized dogs. Moving average RV pressure leads to (PRV) was used as an index of the RV strain. RV load was changed in two ways: 1) cardiac output (Q) was increased by infusion of isoproterenol (0.7–1.2 micrograms/min) and reduced by infusion of vasopressin (0.3–0.5 U/min); and 2) RV pressure was increased independently on Q by partial balloon obstruction of the RV outflow. When Q was changed by drug infusion there was a linear correlation between leads to PRV and Q (avg r = 0.04). Well-correlated linear relationships were found between expired minute ventilation (VE) and leads to PRV (avg r greater than 0.03), the slopes and intercepts of which were not significantly different whether leads to PRV was changed by altering Q, partial obstruction of RV outflow, or combining both procedures. Bilateral vagotomy did not alter the VE/leads to PRV slope resulting from RV balloon inflations. It is suggested that the RV strain may act as a controller of ventilation and provide a link between Q and VE.

Brachial intra-arterial blood pressure [systolic (AS) and diastolic (AD)] and cuff blood pressure [systolic (CS) and fourth- and fifth-phase diastolic (CD)] were simultaneously measured by a single observer in 13 middle-aged men during 1-min incremental cycle exercise. On the average, the mean AS exceeded the mean CS by 10 to 11 mm Hg, while the mean AD exceeded the average fourth and fifth CD by 5 and 13 mm Hg, respectively. During incremental exercise, AS, CS, AD, and fourth-phase CD increased, while fifth-phase CD decreased. We also measured intra-arterial blood pressure in nine young adult men smokers during 1-min incremental cycle exercise. In both groups, the average intra-arterial blood pressures increased in a relatively linear fashion from rest to maximal exercise: AS change = 74 +/- 5 mm Hg (SE) and AD change = 28 +/- 3 mm Hg for young men; AS change = 59 +/- 5 mm Hg and AD change = 12 +/- 3 mm Hg for middle-aged men. In this population of middle-aged smokers, intra-arterial mean blood pressure during exercise approximated diastolic plus 2/5 pulse pressure for intra-arterial measures or diastolic plus 1/2 pulse pressure for cuff measures rather than the traditional formula of diastolic plus 1/3 pulse pressure.

Two groups of dogs were anaesthetised, paralysed and artificially ventilated using a respirator driven by the phrenic motoneurone output, electrically processed to resemble transpulmonary pressure. In one group, blood gases were maintained constant with closed-chest cardiopulmonary bypass; the second group were studied without ‘bypass’. Therefore it was possible to determine the relative contributions of vagal and chemial feedback to the effect of altering the depth of an inspiration on the associated phrenic motoneurone output. Mono-exponential regressions between change of respirator gain and changes in both inspiratory time and peak amplitude of the processed phrenic signal were found for all dogs. The rate of rise of the processed phrenic signal usually changed in the direction of the change in respirator gain, suggesting the presence of vagal positive feedback during eupnoeic breathing. After vagotomy, all responses were absent in the ‘bypass’ group and small in the ‘non-bypass’ group. These experiments quantitate the role of pulmonary vagal afferent discharge, in phase with inspiration, in the regulation of phrenic motoneurone output in a closed-loop situation.

Resume Les experiences ont porte sur cinquante-cinq lapins soumis a l'aide d'un melange de chloraloseurethane a une narcose tres legere. Les animaux ont ete immobilises au Flaxedil et ventiles artificiellement. Les activites des neurones respiratoires du bulbe et les activites respiratoires des fibres isolees efferentes des nerfs vague et phrenique ont ete enregistrees simultanement. Les effets de la vagotomie bilaterale et d'injections d'histamine ont ete etudies. Nous avons constate que: (1) les neurones respiratoires sont groupes dans la region de l'obex, sans division anatomique en parties inspiratoire et expiratoire; (2) les types d'activite des neurones respiratoires et des fibres du nerf vague et du nerf phrenique ainsi que le type de reaction aux stimuli utilises dependent dans une large mesure du niveau de la narcose; (3) la vagotomie entraine l'acceleration du rythme respiratoire; (4) une injection d'histamine accelere le rythme respiratoire, et determine des changements dans les types d'activite; la vagotomie diminue considerablement l'intensite de cette reaction a l'histamine.

Nasal passages constitute the physiologic airway. Impaired nasal breathing leads to various disease states. History taking and rhinoscopic examination are not sufficient for the complete evaluation of nasal respiratory patency. The authors present a case for introduction of posterior rhinomanometry (own modification) as an objective method in studying nasal patency in children, 127 healthy children (aged 5-10 years) were examined. Nasal resistance to air flow was measured. Posterior rhinomanometry was used in 198 children with impaired nasal patency of various causes. The results obtained in this study are reproducible and in accord with data in the literature. Children between 5 and 10 years of age cooperate well with the examiner. The authors find posterior rhinomanometry suitable for examination of children within this age group.

The "exercise hyperpnoea" problem was studied in 44 anaesthetized rabbits by increasing the cardiac output by: a) tilting by 12 degrees the hindlimbs up; b) injecting into the right heart 15 ml of dextran; c) normocapnic or d) hypocapnic blood. In all cases there was an increase in minute ventilation, VE, independently of the presence or absence of an increase in CO2 flux to the lungs. We decreased the cardiac output by tilting up the head by 10 degrees; this elicited a decrease in VE accompanied by an increase in the fractional end-tidal CO2. For elucidating the role played by carotid body chemoreceptors in eleven rabbits the legs were tilted up before and while the animals breathed pure oxygen. The results were very similar. When the hindlimbs were tilted up after cervical bilateral vagotomy the VE response diminished about fifty per cent. We conclude that: a) hyperpnoea and hypopnoea elicited by a change in cardiac output are due to stimulation or depression of mechanoreceptors localized in the pulmonary circulation, b) carotid body chemoreceptors are not required for these responses, c) fifty per cent of the VE response depends on the integrity of the vagus nerves, d) it is probable that the increase in venous return plays an important role in the hyperpnoea at the onset of muscular exercise.

This study reinvestigates the problems of ventilatory drive linked to the rate of CO2 flow to the pulmonary circulation, and the respiratory response to the changes in cardiac output, which are both known to be capable of matching ventilation to perfusion so that the arterial blood remains isocapnic. Taking advantage of the known influence of posture on venous return, 21 healthy humans were subjected to the experimental maneuver which consisted of lifting the legs from the lower to uppermost position. This maneuver increased venous return to the right heart by bringing stagnant blood from the legs carrying more CO2. In all, 34 tests were performed. The results showed a rapid increase in minute ventilation of 12.3 percent. There were also increases in end-tidal CO2, CO2 production and cardiac output. All of the changes were fully reversible and reproducible. The results support the concept that a higher gain CO2 receptor system is operating during increased CO2 delivery to the lungs via the circulation. The possibility of reflex hyperpnea elicited from pressoreceptors localized in pulmonary circulation is also considered.

The effect of haemorrhage and transfusion of blood on minute ventilation, VE, in eight anaesthetized pigs was studied by bleeding the animals from the jugular vein and retransfusion of the same quantity of blood. Bleeding the pig from the jugular vein at approximately 50 ml/min, during 2 min, decreased V/ by 8.4 +/- 5.6 per cent of the pre-bleeding control valve (p = 0.0004) and when the same quantity of blood was retransfused the VE increased by 11.8 +/- 12.4 per cent (p=0.03). The possible mechanisms are discussed. We conclude that: a) mechanoreceptors localized in pulmonary circulation play an important role in the VE response during the first seconds of venous haemorrhage and venous transfusion, b) autotransfusion from moving legs and its effect on pulmonary mechanoreceptors may explain in part hyperpnoea at the onset of muscular exercise.

To investigate the contribution of vascular and metabolic stimuli to the sustained hyperpnea after exercise, the respiratory effects of obstructing and then releasing the femoral blood flow were recorded in 15 normal volunteers during recovery from steady-state cycle exercise (80 W). Obstruction was achieved using cuffs around the upper thighs, inflated for the first 2 min of recovery to a pressure of 200 mmHg. Cuff inflation significantly reduced ventilation during recovery compared with control (P less than 0.001); the subsequent release of pressure was accompanied by an increase in ventilation (averaging 3.2 l/min), which began on the first breath after release. This preceded a rise in end-tidal CO2 (maximum 8.3 Torr increase), which first became significant on the fourth breath after release and led to a further rise in ventilation. The first-breath increase in ventilation after cuff release persisted, although slightly attenuated (averaging 2.5 l/min), in additional experiments with inspired O2 fraction of 1.0. The pattern of ventilatory response was also similar when the experiments were performed with 5% CO2 in air as the inspirate. The immediate rise in ventilation on cuff release, together with the persistent response on 100% O2, suggests that the vascular changes resulting from cuff release exert an influence on ventilation independent of the effects of released metabolites on the known chemoreceptors. The persistence of the response on 5% CO2 indicates that CO2-sensitive lung afferents do not have a major role in these responses.

A prostacyclin-like substance was detected by bioassay in the blood of dogs and cats during hypoventilation and increased end-expiratory pressure. This biologically active material, most likely originating from lungs, relaxed isolated vascular strips and disaggregated platelets. Its release was not prevented by indomethacin or aspirin. Biological activity was not abolished by 10 min incubation of blood at 38 degrees C. Although the identity of the substance has not been established the release of a biologically active prostacyclin-like material might play a role in circulatory adaptation to disturbed ventilatory function.

Anaesthetized rabbits were given 200 ppm sulphur dioxide to breathe for 10 min. This abolished activity in 23 of 26 pulmonary stretch receptors, while leaving that of lung irritant receptors unimpaired. The Breuer-Hering reflex was abolished and breathing became deeper and slower. Inspiratory time (tI) was increased and expiratory time (tE) decreased. Subsequent vagotomy increased tidal volume (VT), tI and tE. In animals with stretch receptors blocked, injections of phenyl diguanide and histamine still increased breathing frequency and decreased VT, indicating that reflexes from lung irritant and J-receptors were intact. Inhalation of 8% CO2 caused a bigger increase in frequency and tidal volume in rabbits with stretch receptor block compared with controls or those after vagotomy. Induction of pneumothorax with stretch receptor block transiently prolonged tI and shortened tE; removal of the pneumothorax also transiently shortened tE and usually also decreased tI. The results suggest that lung irritant receptors reflexly shorten tE in all our experimental conditions, but have various effects on tI which may depend on the timing of the irritant receptor discharge and refractoriness of the inspiratory response.

The level of integrated phrenic nerve activity (C3 root) has been studied under various forms of impairment of respiratory muscle function, such as paralysis (gallamine), pneumothorax and phrenectomy. Experiments were performed in two groups of rabbits artificially ventilated by means of the conventional respirator and the phrenic nerve driven respirator. It was found that both paralysis and bilateral pneumothorax were very strong stimuli exciting central inspiratory activity despite the constancy of ventilation. Under the same conditions phrenectomy caused only slight persistent excitation of the activity studied. It is worth noting that activity in the C3 root contralateral to the cut phrenic nerve trunk was as a rule more strongly stimulated. As artificial ventilation preceded by administration of gallamime is commonly applied in physiological experiments, the possibility of its effect on central respiratory activity will be discussed.