Activity of bulbar respiratory neurones during cough and other respiratory tract reflexes in cats

Changes evoked by mechanical stimulation of the relevant parts of the respiratory tract in the activity of inspiratory and expiratory neurones in the ventral respiratory group of the medulla oblongata, and in pleural pressure and the diaphragmatic electromyogram, were determined during cough, sneeze and the aspiration and expiration reflexes in 17 anaesthetized (but not paralysed) cats. The results of 72 tests of elicitation of the given reflexes showed that: Compared with the control inspiration, both the mean and the maximum discharge frequency of spontaneously active inspiratory neurones rose during the inspiratory phase of cough, sneeze and the aspiration reflex. Regular recruitment of new inspiratory units was also observed in the inspiratory phase of cough and the aspiration reflex. Compared with the control expiration, both the mean and the maximum discharge frequency of spontaneously active expiratory neurones rose during the cough, sneeze and expiration reflex effort. Recruitment of latent expiratory neurones was always observed in the expulsive phase of the given respiratory processes. The recruitment of latent expiratory neurones was accompanied by reciprocal inhibition of the activity of inspiratory units and recruitment of latent inspiratory neurones by inhibition of the activity of expiratory units and recruitment of latent inspiratory neurones by inhibition of the activity of expiratory units. Regular recruitment of the same expiratory neurones in all expulsive respiratory processes, together with the similar incidence of inspiratory neurones in the inspiratory phase of sneeze and the aspiration reflex, indicates that they are "nonspecific" in character.     

Respiratory volume-time relationships during resistive loading in the cat.

The first-breath (neural) effects of graded resistive loads added separately during inspiration and expiration was studied in seven anesthetized cats before and after bilateral vagotomy. Additions of airflow resistance during inspiration reduced the volume inspired (VI) and increased inspiratory duration (TI). The duration of the ensuing unloaded expiration (TE) was unchanged. Vagotomy eliminated the TI modulation with inspiratory loads. Tracheal occlusion at the onset of inspiration yielded TI values similar to the fixed values observed following vagotomy. Resistive loads added during expiration produced similar results. Expired volume (VE) decreased and (TE) increased approaching the values obtained after vagotomy. Unlike the inspiratory resistive loads, loading during expiration results in an upward shift in the functional residual capacity (FRC). The FRC shift produces a time lag between the onset of diaphragmatic (EMG) activity and the initiation of airflow of the next (unloaded) inspiration. These studies suggest separate volume-time relationships for the inspiratory and expiratory phases of the breathing cycle. Both relationships are dependent upon vagally mediated volume feedback.     

Central inspiratory influence on abdominal expiratory nerve activity

Our purpose was to determine whether the intensity of abdominal expiratory nerve discharge is conditioned by the intensity of the preceding inspiratory phrenic discharge, independent of mechanical and chemical afferent influences. In decerebrate, paralyzed, vagotomized cats with bilateral pneumothoraxes, we recorded phrenic and abdominal (cranial iliohypogastric nerve, L1) nerve activities at hyperoxic normocapnia. We reduced the duration and intensity (i.e., integrated peak height) of phrenic nerve discharge for single cycles by stimulating the cut central end of the superior laryngeal nerve (SLN) during the central inspiratory phase (75 microA, 20-50 Hz, 0.2-ms pulse). Premature termination of inspiration consistently reduced expiratory duration (TE) and abdominal expiratory nerve activity (area of integrated neurogram), but the average reduction in TE was much less than the reduction in abdominal nerve activity (14 vs. 51%). Stimulation of the cut central end of the vagus nerve yielded similar results, as did spontaneous premature terminations of inspiration, which we observed in one cat. SLN stimulation during hyperoxic hypercapnia resulted in more variable responses, and higher stimulation frequencies were usually required to abort inspiration. SLN (or vagal) stimulation during expiration consistently increased abdominal expiratory nerve activity. We speculate that this facilitatory response is gated during inspiration, thereby allowing the inspiratory conditioning effect on the subsequent expiration to be expressed.(ABSTRACT TRUNCATED AT 250 WORDS)     

Postinspiratory activity of costal and crural diaphragm.

Because the first stage of expiration or "postinspiration" is an active neurorespiratory event, we expect some persistence of diaphragm electromyogram (EMG) after the cessation of inspiratory airflow, as postinspiratory inspiratory activity (PIIA). The costal and crural segments of the mammalian diaphragm have different mechanical and proprioceptive characteristics, so postinspiratory activity of these two portions may be different. In six canines, we implanted chronically EMG electrodes and sonomicrometer transducers and then sampled EMG activity and length of costal and crural diaphragm segments at 4 kHz, 10.2 days after implantation during wakeful, resting breathing. Costal and crural EMG were reviewed on-screen, and duration of PIIA was calculated for each breath. Crural PIIA was present in nearly every breath, with mean duration 16% of expiratory time, compared with costal PIIA with duration -2. 6% of expiratory time (P < 0.002). A linear regression model of crural centroid frequency vs. length, which was computed during the active shortening of inspiration, did not accurately predict crural EMG centroid frequency values at equivalent length during the controlled relaxation of postinspiration. This difference in activation of crural diaphragm in inspiration and postinspiration is consistent with a different pattern of motor unit recruitment during PIIA.     

Comparative study of the activity of medullary respiratory neurones during polypnea triggered by hypothalamic electrical stimulation or by hypothalamic heating (author's transl)

We have compared in "encéphale isolé bas" cats the activity of medullary respiratory neurones during polypnea triggered by electrical stimulation (PSt) or by heating (PTh) of the hypothalamus. The medullary respiratory neurones are classified according to:--their anatomical localization (dorsal or ventral respiratory nucleus);--their axon destination (spinal : bulbo-spinal respiratory neurones; non spinal : propriobulbar neurones);--their discharge pattern;--the correlation coefficient between the number of spikes delivered in each burst and the duration of the corresponding respiratory phase (HILAIRE et MONTEAU, 1975). 1. During the two polypneas (PSt and PTh), we observe:--a reduction of activity that preferentially affects some groups of neurones (propriobulbar neurones) (fig. 3);--an inversion of the discharge firing rate, which increases during inspiration in normopnea and decreases in polypnea (fig. 1; fig. 6);--a decrease of the maximal discharge firing rate for the neurones of different groups (Table V). 2. However, two differences exist : during PSt, the maximal discharge firing rate increases for the inspiratory bulbo-spinal neurones of the dorsal nucleus and for the early-burster inspiratory propriobulbar neurones. The recruitment of the bulbo-spinal inspiratory neurones seems to be different; they are activated earlier during PSt than during PTh (Table VI). 3. Some of the observed differences are probably quantitative and we think that polypnea triggered by hypothalamic electrical stimulation is a good model for thermal polypnea.     

Functional Tests of Pulmonary Respiration

The duration of voluntary holding of external breathing (VHEB) on inspiration or expiration was measured in 300 18- to 22-year-old subjects of both sexes at rest and during bicycle ergometer exercise, in a sitting position or in the float posture, on land or immersed in water. The length of expiratory VHEB was found to depend on the residual lung volume, whereas inspiratory VHEB correlated with the functional residual lung capacity. Physical exercise increased the oxygen uptake and accelerated the consumption of lung oxygen reserves available on inspiration or expiration and decreased the duration of VHEB. Special formulas were proposed for calculating the theoretical length of VHEB and for its comparing the result with corresponding experimental values, as well as for determination of the lung volumes and oxygen consumption.     

Effects of hypercapnia on inspiratory and expiratory muscle activity during expiration

Persistence of inspiratory muscle activity during the early phase of expiratory airflow slows the rate of lung deflation, whereas heightened expiratory muscle activity produces the opposite effect. To examine the influence of increased chemoreceptor drive and the role of vagal afferent activity on these processes, the effects of progressive hypercapnia were evaluated in 12 anesthetized tracheotomized dogs before and after vagotomy. Postinspiratory activity of inspiratory muscles (PIIA) and the activity of expiratory muscles were studied. During resting breathing, the duration of PIIA correlated with the duration of inspiration but not with expiration. Parasternal intercostal PIIA was directly related to that of the diaphragm. Based on their PIIA, dogs could be divided into two groups: one with prolonged PIIA (mean 0.57 s) and the other with brief PIIA (mean 0.16 s). Hypercapnia caused progressive shortening of the PIIA in the dogs with prolonged PIIA during resting breathing. The electrical activity of the external oblique and internal intercostal muscles increased gradually during CO2 rebreathing in all dogs both pre- and postvagotomy. After vagotomy, abdominal activity continued to increase with hypercapnia but was less at all levels of PCO2. The internal intercostal response to hypercapnia was not affected by vagotomy. The combination of shorter PIIA and augmented expiratory activity with hypercapnia might, in addition to changes in lung recoil pressure and airway resistance, hasten exhalation.     

Coordination of swallowing and respiration in unconscious subjects

We investigated the coordination of swallowing and breathing in 11 unconscious patients with an endotracheal tube in place during the recovery period from general anesthesia. Swallows occurred during both the inspiratory and expiratory phases with no preponderant occurrence during either phase. When a swallow occurred during inspiration, the inspiration was interrupted immediately and was followed by expiration, but the durations of both inspiration and expiration were progressively increased as the time from the onset of inspiration to the onset of swallowing was progressively delayed. A swallow coinciding with the expiratory phase progressively prolonged the duration of the expiration that had been interrupted as the timing of swallowing was progressively delayed. Repeated swallows invariably and in a predictable manner caused changes in the breathing pattern. Thus when the frequency of regularly repeated swallows was relatively high, the breathing pattern was characterized by regular, shallow, and rapid breaths. When the frequency of regularly repeated swallows was relatively low, the breathing pattern was characterized by regular, deep, and slow breaths. When the frequency of repeated swallows was irregular, the breathing patterns were characterized by inconsistent changes in tidal volume and respiratory frequency. Our results indicate that, in unconscious subjects, some mechanisms integrating respiration and swallowing are operative and responsible for changes in breathing patterns during swallowing.     

Role of costal and crural diaphragm and parasternal intercostals during coughing in cats

The inspiratory phase of coughs often consists of large inspired volumes and increased motor discharge to the costal diaphragm. Furthermore, diaphragm electrical activity may persist into the early expiratory portion of coughs. To examine the role of other inspiratory muscles during coughing, electromyograms (EMG) recorded from the crural diaphragm (Dcr) and parasternal intercostal (PSIC) muscles were compared to EMG of the costal diaphragm (Dco) in anesthetized cats. Tracheal or laryngeal stimulation typically produced a series of coughs, with variable increases in peak inspiratory EMGs of all three muscles. On average, peak inspiratory EMG of Dco increased to 346 +/- 60% of control (P less than 0.001), Dcr to 514 +/- 82% of control (P less than 0.0002), and PSIC to 574 +/- 61% of control (P less than 0.0005). Augmentations of Dcr and PSIC EMG were both significantly greater than of Dco EMG (P less than 0.05 and P less than 0.002, respectively). In most animals, EMG of Dco correlated significantly with EMG of Dcr and of PSIC during different size coughs. Electrical activity of all three muscles persisted into the expiratory portions of many (but not all) coughs. The duration of expiratory activity lasted on average 0.17 +/- 0.03 s for Dco, 0.25 +/- 0.06 s for Dcr, and 0.31 +/- 0.09 s for PSIC. These results suggest that multiple respiratory muscles are recruited during inspiration of coughs, and that the persistence of electrical activity into expiration of coughs is not unique to the costal diaphragm.     

Assessment of upper airway stabilizing forces with the use of phrenic nerve stimulation in conscious humans.

Phrenic nerve stimulation (PNS) applied at end-expiration allows the investigation of passive upper airway (UA) dynamic during wakefulness. Assuming that phasic UA dilating/stabilizing forces should modify the UA properties when twitches are applied during inspiration, we compared the UA dynamic responses to expiratory and inspiratory twitches (2 s and 200 ms after expiratory and inspiratory onset, respectively) in nine men (mean age 28 yr). This procedure was repeated with a 2-cm mouth opening provided with a closed mouthpiece. The percentage of flow-limited (FL) twitches was significantly higher when PNS was realized during expiration than during inspiration. Maximal inspiratory flow (Vi(max)) of FL twitches was significantly higher for inspiratory twitches (1,383 +/- 42 and 1,185 +/- 40 ml/s). With mouth aperture, Vi(max) decreased with an increase in the corresponding pharyngeal resistance values, and the percentage of twitch with a FL regimen increased but only for inspiratory twitches. We conclude that 1) UA dynamics are significantly influenced by the inspiratory/expiratory timing at which PNS is applied, 2) the improvement in UA dynamic properties observed from expiratory to inspiratory PNS characterizes the overall inspiratory stabilizing effects, and 3) mouth aperture alters the stability of UA structures during inspiration.     

Respiratory neuronal activity during apnea and other breathing patterns induced by laryngeal stimulation

Respiration cycles through three distinct phases (inspiration, postinspiration, and expiration) each having corresponding medullary cells that are excited during one phase and inhibited during the other two. Laryngeal stimulation is known to induce apnea in newborn animals, but the cellular mechanisms underlying this effect are not known. Intracellular recording of ventral respiratory group neurons was accomplished in intact anesthetized, paralyzed, and mechanically ventilated piglets. Apnea was induced by insufflation of the larynx with ammonia-saturated air, smoke, or water. Laryngeal insufflation induced phrenic nerve apnea, stimulation of postinspiratory neurons, and stable membrane potentials in inspiratory and expiratory cells consistent with postinspiratory inhibition. Usually the membrane potential of each neuronal type cycled through an expiratory level before onset of the first recovery breath. Variants of the apnea response, probably reflecting the aspiration reflex or sniffing, sneezing, coughing, and swallowing, were also observed. These latter patterns showed oscillation between inspiration and postinspiration without an apparent intervening stage II expiratory phase. However, stage II expiratory activity always preceded onset of the first ramp inspiration after such a pattern. These findings suggest that activation of postinspiratory mechanisms causes profound alterations in the respiratory pattern and that stage II expiration importantly modulates recovery of ramp inspiratory activity. The mechanism of this latter effect may be inhibition of early inspiratory neurons with consequent postinhibitory rebound.     

Effect of expiratory loading on glottic dimensions in humans

We examined the effects of external mechanical loading on glottic dimensions in 13 normal subjects. When flow-resistive loads of 7, 27, and 48 cmH2O X l-1 X s, measured at 0.2 l/s, were applied during expiration, glottic width at the mid-tidal volume point in expiration (dge) was 2.3 +/- 12, 37.9 +/- 7.5, and 38.3 +/- 8.9% (means +/- SE) less than the control dge, respectively. Simultaneously, mouth pressure (Pm) increased by 2.5 +/- 4, 3.0 +/- 0.4, and 4.6 +/- 0.6 cmH2O, respectively. When subjects were switched from a resistance to a positive end-expiratory pressure at comparable values of Pm, both dge and expiratory flow returned to control values, whereas the level of hyperinflation remained constant. Glottic width during inspiration (unloaded) did not change on any of the resistive loads. There was a slight inverse relationship between the ratio of expiratory to inspiratory glottic width and the ratio of expiratory to inspiratory duration. Our results show noncompensatory glottic narrowing when subjects breathe against an expiratory resistance and suggest that the glottic dimensions are influenced by the time course of lung emptying during expiration. We speculate that the glottic constriction is related to the increased activity of expiratory medullary neurons during loaded expiration and, by increasing the internal impedance of the respiratory system, may have a stabilizing function.     

Abdominal muscle activity during breathing with and without inspiratory and expiratory loads in healthy subjects

Central Nervous System modulates the motor activities of all trunk muscles to concurrently regulate the intra-abdominal and intra-thoracic pressures. The study aims to evaluate the effect of inspiratory and expiratory loads on abdominal muscle activity during breathing in healthy subjects. Twenty-three higher education students (21.09 ± 1.56 years; 8 males) breathed at a same rhythm (inspiration: two seconds; expiration: four seconds) without load and with 10% of the maximal inspiratory or expiratory pressures, in standing. Surface electromyography was performed to assess the activation intensity of rectus abdominis, external oblique and transversus abdominis/internal oblique muscles, during inspiration and expiration. During inspiration, transversus abdominis/internal oblique activation intensity was significantly lower with inspiratory load when compared to without load (p = 0.009) and expiratory load (p = 0.002). During expiration, the activation intensity of all abdominal muscles was significantly higher with expiratory load when compared to without load (p < 0.05). The activation intensity of external oblique (p = 0.036) and transversus abdominis/internal oblique (p = 0.022) was significantly higher with inspiratory load when compared to without load. Transversus abdominis/internal oblique activation intensity was significantly higher with expiratory load when compared to inspiratory load (p < 0.001).Transversus abdominis/internal oblique seems to be the most relevant muscle to modulate the intra-abdominal pressure for the breathing mechanics.     

Control of larynx during loaded breathing in normal subjects

We examined laryngeal resistance (Rla) in six normal subjects in control and four kinds of loaded breathing: hypercapnia, chest strapping, added external resistance, and inhaled methacholine. Rla was measured with a low-frequency sound methed (Sekizawa et al., J. Appl. Physiol. 55: 591-597, 1983). In control and the four kinds of loaded breathing, changes in Rla were tightly coupled with ventilation and Rla decreased during inspiration and increased during expiration. Hypercapnia and chest strapping significantly decreased Rla in both inspiration and expiration in all subjects. Added external resistance decreased inspiratory Rla in all subjects, but decreased expiratory Rla in three subjects, did not change it in two subjects, and increased it in one subject. Inhaled methacholine increased Rla in both inspiration and expiration in all subjects. The present study suggests that although laryngeal movement is tightly coupled with ventilation, laryngeal aperture may be determined by the complex competition of dilating and constricting mechanisms associated with the activity of the respiratory center and neural reflexes from the airway.     

Effects of expiratory loading on respiration in humans

We examined the effects of expiratory resistive loads of 10 and 18 cmH2O.l-1.s in healthy subjects on ventilation and occlusion pressure responses to CO2, respiratory muscle electromyogram, pattern of breathing, and thoracoabdominal movements. In addition, we compared ventilation and occlusion pressure responses to CO2 breathing elicited by breathing through an inspiratory resistive load of 10 cmH2O.l-1.s to those produced by an expiratory load of similar magnitude. Both inspiratory and expiratory loads decreased ventilatory responses to CO2 and increased the tidal volume achieved at any given level of ventilation. Depression of ventilatory responses to Co2 was greater with the larger than with the smaller expiratory load, but the decrease was in proportion to the difference in the severity of the loads. Occlusion pressure responses were increased significantly by the inspiratory resistive load but not by the smaller expiratory load. However, occlusion pressure responses to CO2 were significantly larger with the greater expiratory load than control. Increase in occlusion pressure observed could not be explained by changes in functional residual capacity or chemical drive. The larger expiratory load also produced significant increases in electrical activity measured during both inspiration and expiration. These results suggest that sufficiently severe impediments to breathing, even when they are exclusively expiratory, can enhance inspiratory muscle activity in conscious humans.     

Integration-differentiation and gating of carotid afferent traffic that shapes the respiratory pattern.

The phase-dependent plasticity of carotid chemoafferent signaling was studied with electrical stimulation of a carotid sinus nerve during either inspiration or expiration in anesthetized, glomectomized, vagotomized, paralyzed, and ventilated rats. Stroboscopic and interferometric analyses of the resulting phase-contrast disturbances of the respiratory rhythm revealed that carotid chemoafferent traffic was dynamically filtered centrally by a parallel bank of leaky integrators and differentiators, each being logically gated to the inspiratory or expiratory phase in a stop-and-go manner as follows: 1) carotid short-term potentiation of inspiratory drive was mediated by dual integrators that both shortened inspiration and augmented phrenic motor output cooperatively in long and short timescales; 2) carotid short-term depression of respiratory frequency was mediated by a (possibly pontine) integrator that lengthened expiration with a relatively long memory; and 3) carotid "chemoreflex" shortening of expiration was mediated by an occult fast integrator, which, together with carotid short-term depression, formed a differentiator. These effects were modulated anteriorly by integrators in the nucleus tractus solitarius that were "auto-gated" to, or recruited by, the carotid sinus nerve input. Such phase-selective and activity-dependent time-frequency filtering of carotid chemoafferent feedback in parallel neurological-neurodynamic central pathways may profoundly affect respiratory stability during hypoxia and sleep and could contribute to the dynamic optimization of the respiratory pattern and maintenance of homeostasis in health and in disease states.     

Flow and volume dependence of respiratory mechanical properties studied by forced oscillation

The influence of inspiratory and expiratory flow magnitude, lung volume, and lung volume history on respiratory system properties was studied by measuring transfer impedances (4-30 Hz) in seven normal subjects during various constant flow maneuvers. The measured impedances were analyzed with a six-coefficient model including airway resistance (Raw) and inertance (Iaw), tissue resistance (Rti), inertance (Iti), and compliance (Cti), and alveolar gas compressibility. Increasing respiratory flow from 0.1 to 0.4 1/s was found to increase inspiratory and expiratory Raw by 63% and 32%, respectively, and to decrease Iaw, but did not change tissue properties. Raw, Iti, and Cti were larger and Rti was lower during expiration than during inspiration. Decreasing lung volume from 70 to 30% of vital capacity increased Raw by 80%. Cti was larger at functional residual capacity than at the volume extremes. Preceding the measurement by a full expiration rather than by a full inspiration increased Iaw by 15%. The data suggest that the determinants of Raw and Iaw are not identical, that airway hysteresis is larger than lung hysteresis, and that respiratory muscle activity influences tissue properties.     

Inspiratory muscle activity during bird song

The apparently continuous flow of bird song is in reality punctuated by brief periods of silence during which there are short inspirations called minibreaths. To determine whether these minibreaths are accompanied, and thus perhaps caused, by activity in inspiratory muscles, electromyographic (EMG) activity was recorded in M. scalenus in zebra finches and in M. scalenus and Mm. levatores costarum in cowbirds, together with EMGs from the abdominal expiratory muscles, air sac pressure and tracheal airflow. EMG activity in Mm. scalenus and levatores costarum consistently preceded the onset of negative air sac pressure by ∼11 ms during both quiet respiration and singing in both species. The electrical activity of these two muscles was very similar. Compared with during quiet respiration, the amplitude of inspiratory muscle EMG during singing was increased between five- and 12-fold and its duration was decreased from >200 ms to on average 41 ms during minibreaths, again for both species, but inspiratory muscle activity did not overlap with that of the expiratory muscles. Thus, there was no indication that the inspiratory muscles acted either to shorten the duration of expiration or to reduce the expiratory effort as might occur if both expiratory and inspiratory muscles were simultaneously active. Inspiratory and expiratory muscle activities were highly stereotyped during song to the extent that together, they defined the temporal pattern of the songs and song types of individual birds. © 1998 John Wiley & Sons, Inc. J Neurobiol 36: 441–453, 1998     

Internal intercostal nerve discharges in the cat: influence of chemical stimuli

We studied the influence of central and peripheral chemoreceptor stimulation on the activities of the phrenic and internal intercostal (iic) nerves in decerebrate, vagotomized, and paralyzed cats with bilateral pneumothoraces. Whole iic nerves of the rostral thorax (T2-T5) usually discharged during neural inspiration, whereas those of the caudal thorax (T7-T11) were primarily active during neural expiration. Filaments of rostral iic nerves that terminated in iic muscles generally discharged during expiration, suggesting that inspiratory activity recorded in whole iic nerves may have innervated other structures, possibly parasternal muscles. All nerves were phasically active at hyperoxic normocapnia and increased their activities systematically with hypercapnia. Isocapnic hypoxia or intra-arterial NaCN injection consistently increased phrenic and inspiratory iic nerve activities. In contrast, expiratory iic nerve discharges were either decreased (10 cats) or increased (7 cats) by hypoxia. Furthermore, expiratory responses to NaCN were highly variable and could not be predicted from the corresponding response to hypoxia. The results show that central and peripheral chemoreceptor stimulation can affect inspiratory and expiratory motoneuron activities differentially. The variable effects of hypoxia on expiratory iic nerve activity may reflect a relatively weak influence of carotid body afferents on expiratory bulbospinal neurons. However, the possibility that the magnitude of expiratory motoneuron activity is influenced by the intensity of the preceding centrally generated inspiratory discharge is also discussed.     

Effects of negative upper airway pressure on pattern of breathing in sleeping infants

Negative upper airway (UAW) pressure inhibits diaphragm inspiratory activity in animals, but there is no direct evidence of this reflex in humans. Also, little is known regarding reflex latency or effects of varying time of stimulation during the breathing cycle. We studied effects of UAW negative pressure on inspiratory airflow and respiratory timing in seven tracheostomized infants during quiet sleep with a face mask and syringe used to produce UAW suction without changing lower airway pressure. Suction trials lasted 2-3 s. During UAW suction, mean and peak inspiratory airflow as well as tidal volume was markedly reduced (16-68%) regardless of whether stimulation occurred in inspiration or expiration. Reflex latency was 42 +/- 3 ms. When suction was applied during inspiration or late expiration, the inspiration and the following expiration were shortened. In contrast, suction applied during midexpiration prolonged expiration and tended to prolong inspiration. The changes in flow, tidal volume, and timing indicate a marked inhibitory effect of UAW suction on thoracic inspiratory muscles. Such a reflex mechanism may function in preventing pharyngeal collapse by inspiratory suction pressure.