Influence of airway diameter and cell confluence on epithelial cell injury in an in vitro model of airway reopening.

Recent advances in the ventilation of patients with acute respiratory distress syndrome (ARDS), including ventilation at low lung volumes, have resulted in a decreased mortality rate. However, even low-lung volume ventilation may exacerbate lung injury due to the cyclic opening and closing of fluid-occluded airways. Specifically, the hydrodynamic stresses generated during airway reopening may result in epithelial cell (EpC) injury. We utilized an in vitro cell culture model of airway reopening to investigate the effect of reopening velocity, airway diameter, cell confluence, and cyclic closure/reopening on cellular injury. Reopening dynamics were simulated by propagating a constant-velocity air bubble in an adjustable-height parallel-plate flow chamber. This chamber was occluded with different types of fluids and contained either a confluent or a subconfluent monolayer of EpC. Fluorescence microscopy was used to quantify morphological properties and percentage of dead cells under different experimental conditions. Decreasing channel height and reopening velocity resulted in a larger percentage of dead cells due to an increase in the spatial pressure gradient applied to the EpC. These results indicate that distal regions of the lung are more prone to injury and that rapid inflation may be cytoprotective. Repeated reopening events and subconfluent conditions resulted in significant cellular detachment. In addition, we observed a larger percentage of dead cells under subconfluent conditions. Analysis of this data suggests that in addition to the magnitude of the hydrodynamic stresses generated during reopening, EpC morphological, biomechanical, and microstructural properties may also be important determinants of cell injury.     

Ventilation heterogeneity in excised lobes: effect of tidal volume.

Although several factors are known to influence nonuniformity of ventilation, including lung mechanical properties (regional structure and compliance), external factors (chest wall, pleural pressure, heart), and ventilatory parameters (tidal and preinspiratory volume, flow rate), their relative contributions are poorly understood. We studied five excised, unperfused, canine right-middle lobes under varied levels of tidal volume (VT), thus eliminating many factors affecting heterogeneity. Multiple-breath washouts of N(2) were analyzed for anatomic dead space volume (VD(anat)), nonuniformity of N(2) washout, and nonuniformity between joined acinar regions vs. that occurring between larger joined regions. Approximately 80% of ventilation heterogeneity was found among joined acinar regions at resting levels of VT, but increasing VT reduced intra-acinar heterogeneity to about 25% of that found at resting levels. Increasing VT had essentially no effect on VD(anat) and heterogeneity among larger joined regions. The results indicate that the magnitude of VT is a major influence on the dominant intra-acinar component of ventilation heterogeneity and that VT effects on VD(anat) are likely due to perfusion and/or influences normally external to the lobar structure.     

Ventilatory accommodation of oxygen demand and respiratory water loss in kangaroos from mesic and arid environments, the eastern grey kangaroo (Macropus giganteus) and the red kangaroo (Macropus rufus)

We studied ventilation in kangaroos from mesic and arid environments, the eastern grey kangaroo (Macropus giganteus) and the red kangaroo (Macropus rufus), respectively, within the range of ambient temperatures (T(a)) from -5 degrees to 45 degrees C. At thermoneutral temperatures (Ta=25 degrees C), there were no differences between the species in respiratory frequency, tidal volume, total ventilation, or oxygen extraction. The ventilatory patterns of the kangaroos were markedly different from those predicted from the allometric equation derived for placentals. The kangaroos had low respiratory frequencies and higher tidal volumes, even when adjustment was made for their lower basal metabolism. At Ta>25 degrees C, ventilation was increased in the kangaroos to facilitate respiratory water loss, with percent oxygen extraction being markedly lowered. Ventilation was via the nares; the mouth was closed. Differences in ventilation between the two species occurred at higher temperatures, and at 45 degrees C were associated with differences in respiratory evaporative heat loss, with that of M. giganteus being higher. Panting in kangaroos occurred as a graded increase in respiratory frequency, during which tidal volume was lowered. When panting, the desert red kangaroo had larger tidal volumes and lower respiratory frequencies at equivalent T(a) than the eastern grey kangaroo, which generally inhabits mesic forests. The inference made from this pattern is that the red kangaroo has the potential to increase respiratory evaporative heat loss to a greater level.     

Mechanical constraints on exercise hyperpnea in endurance athletes.

We determined how close highly trained athletes [n = 8; maximal oxygen consumption (VO2max) = 73 +/- 1 ml.kg-1.min-1] came to their mechanical limits for generating expiratory airflow and inspiratory pleural pressure during maximal short-term exercise. Mechanical limits to expiratory flow were assessed at rest by measuring, over a range of lung volumes, the pleural pressures beyond which no further increases in flow rate are observed (Pmaxe). The capacity to generate inspiratory pressure (Pcapi) was also measured at rest over a range of lung volumes and flow rates. During progressive exercise, tidal pleural pressure-volume loops were measured and plotted relative to Pmaxe and Pcapi at the measured end-expiratory lung volume. During maximal exercise, expiratory flow limitation was reached over 27-76% of tidal volume, peak tidal inspiratory pressure reached an average of 89% of Pcapi, and end-inspiratory lung volume averaged 86% of total lung capacity. Mechanical limits to ventilation (VE) were generally reached coincident with the achievement of VO2max; the greater the ventilatory response, the greater was the degree of mechanical limitation. Mean arterial blood gases measured during maximal exercise showed a moderate hyperventilation (arterial PCO2 = 35.8 Torr, alveolar PO2 = 110 Torr), a widened alveolar-to-arterial gas pressure difference (32 Torr), and variable degrees of hypoxemia (arterial PO2 = 78 Torr, range 65-83 Torr). Increasing the stimulus to breathe during maximal exercise by inducing either hypercapnia (end-tidal PCO2 = 65 Torr) or hypoxemia (saturation = 75%) failed to increase VE, inspiratory pressure, or expiratory pressure. We conclude that during maximal exercise, highly trained individuals often reach the mechanical limits of the lung and respiratory muscle for producing alveolar ventilation. This level of ventilation is achieved at a considerable metabolic cost but with a mechanically optimal pattern of breathing and respiratory muscle recruitment and without sacrifice of a significant alveolar hyperventilation.     

Effect of transrespiratory pressure on PETCO2-PaCO2 and ventilatory reflexes in humans

Inspiratory muscle activity increases when lung volume is increased by continuous positive-pressure breathing in conscious human subjects (Green et al., Respir. Physiol. 35: 283-300, 1978). Because end-tidal CO2 pressure (PETCO2) does not change, these increases have not been attributed to chemoreflexes. However, continuous positive-pressure breathing at 20 cmH2O influences the end-tidal to arterial CO2 pressure differences (Folkow and Pappenheimer, J. Appl. Physiol. 8: 102-110, 1955). We have compared PETCO2 with arterial CO2 pressure (PaCO2). We have compared PETCO2 with arterial CO2 pressure (PaCO2) in healthy human subjects exposed to continuous positive airway pressure (10 cmH2O) or continuous negative pressure around the torso (-15 cmH2O) sufficient to increase mean lung volume by about 650 ml. The difference between PETCO2 and PaCO2 was not decreased, and we conclude that PETCO2 is a valid measure of chemical drive to ventilation in such circumstances. We observed substantial increases in respiratory muscle electromyograms during pressure breathing as seen previously and conclude this response must originate by proprioception. On average, the compensation of tidal volume thus afforded was complete, but the wide variability of individual responses suggests that there was a large cerebral cortical component in the responses seen here.     

Airway pressures in an asymmetrically branched airway model of the dog respiratory system

A computer model of the mechanical properties of the dog respiratory system based on the asymmetrically branching airway model of Horsfield et al. (11) is described. The peripheral ends of this airway model were terminated by a lumped-parameter impedance representing gas compression in the alveoli, and lung and chest wall tissue properties were derived from measurements made in this laboratory. Using this model we predicted the respiratory system impedance and the distribution of pressures along the airways in the dog lung. Predicted total respiratory system impedances for frequencies between 4 and 64 Hz at three lung volumes were found to compare quite closely to measured impedances in dogs. Serial pressure distributions were found to be frequency-dependent and to result in higher pressures in the lung periphery than at the airway opening at some frequencies. The implications of this finding for high-frequency ventilation are discussed.     

Redistribution of blood flow and lung volume between lungs in lateral decubitus postures during unilateral atelectasis and PEEP

The effect of left lung atelectasis on the regional distribution of blood flow (Q), ventilation (V(A)) and gas exchange on the right lung ventilated with 100% O2 was studied in anesthetized dogs in the lateral decubitus posture. Q and V(A) were measured in 1.7 ml lung volume pieces using injected and aerosolized fluorescent microspheres, respectively. Hypoxic pulmonary vasoconstriction (HPV) in the atelectatic lung shifted flow to the ventilated lung. The increased flow in the ventilated lung ensured adequate gas exchange, compensating for the hypoxemia due to shunt contributed by the atelectatic lung. Left lung atelectasis caused a compensatory increase in the ventilated lung FRC that was smaller in the right (RLD) than left (LLD) lateral posture, the effect of lung compression by the atelectatic lung and mediastinal contents in the RLD posture. The O2 deficit measured by (A-a)DO2 increased with left lung atelectasis and was exacerbated in the LLD posture by 10 cm H2O PEEP, a result of increased shunt caused by a shift in Q from the ventilated to the atelectatic lung. The PEEP-induced O2 deficit was eliminated with inversion to the RLD posture.     

Pressure gradient, not exposure duration, determines the extent of epithelial cell damage in a model of pulmonary airway reopening.

The reduction of tidal volume during mechanical ventilation has been shown to reduce mortality of patients with acute respiratory distress syndrome, but epithelial cell injury can still result from mechanical stresses imposed by the opening of occluded airways. To study these stresses, a fluid-filled parallel-plate flow chamber lined with epithelial cells was used as an idealized model of an occluded airway. Airway reopening was modeled by the progression of a semi-infinite bubble of air through the length of the channel, which cleared the fluid. In our laboratory's prior study, the magnitude of the pressure gradient near the bubble tip was directly correlated to the epithelial cell layer damage (Bilek AM, Dee KC, and Gaver DP III. J Appl Physiol 94: 770-783, 2003). However, in that study, it was not possible to discriminate the stress magnitude from the stimulus duration because the bubble propagation velocity varied between experiments. In the present study, the stress magnitude is modified by varying the viscosity of the occlusion fluid while fixing the reopening velocity across experiments. This approach causes the stimulus duration to be inversely related to the magnitude of the pressure gradient. Nevertheless, cell damage remains directly correlated with the pressure gradient, not the duration of stress exposure. The present study thus provides additional evidence that the magnitude of the pressure gradient induces cellular damage in this model of airway reopening. We explore the mechanism for acute damage and also demonstrate that repeated reopening and closure is shown to damage the epithelial cell layer, even under conditions that would not lead to extensive damage from a single reopening event.     

Phase locking of the respiratory rhythm in cats to a mechanical ventilator

Mechanical ventilation of paralyzed, pentobarbital-anesthetized adult cats was performed while recording phrenic nerve activity. The periodic changes in lung volume owing to mechanical ventilation affected the rhythm of central respiratory activity, resulting in a variety of regular and irregular patterns of coupling between respiratory system output, monitored by phrenic activity, and the mechanical ventilator. Phase-locked patterns, in which phrenic burst onset occurred at specific and repetitive phase(s) of the mechanical ventilator, with ratios of ventilator frequency: phrenic burst frequency of 1:2, 1:1, 3:2, 2:1, and 3:1 were observed. Regular and irregular patterns occurred over specific ranges of frequency and volume of the mechanical ventilator. A careful study was made of the 1:1 phase locking as the frequency and inflation volume of the mechanical ventilator were changed. The inspiratory time (TI) was defined as the interval between the time when phrenic activity began to rise and the onset of its rapid decline, and the expiratory time (TE) as the time between inspirations. In the 1:1 phase-locking region, as the frequency of the ventilator was increased both TI and TE decreased, and the phase of phrenic onset in the ventilator cycle changed. During ventilation with frequencies higher than the intrinsic phrenic frequency (initial burst frequency of phrenic activity with the ventilator turned off) inspiratory activity was prematurely terminated by lung inflation (Hering-Breuer inspiratory inhibitory reflex). During ventilation with frequencies lower than the intrinsic phrenic frequency, the onset of phrenic activity was delayed (TE was prolonged) by lung inflation (Hering-Breuer expiratory promoting reflex).     

Mechanics of edematous lungs.

Using the parenchymal marker technique, we measured pressure (P)-volume (P-V) curves of regions with volumes of approximately 1 cm3 in the dependent caudal lobes of oleic acid-injured dog lungs, during a very slow inflation from P = 0 to P = 30 cmH2O. The regional P-V curves are strongly sigmoidal. Regional volume, as a fraction of volume at total lung capacity, remains constant at 0.4-0.5 for airway P values from 0 to approximately 20 cmH2O and then increases rapidly, but continuously, to 1 at P = approximately 25 cmH2O. A model of parenchymal mechanics was modified to include the effects of elevated surface tension and fluid in the alveolar spaces. P-V curves calculated from the model are similar to the measured P-V curves. At lower lung volumes, P increases rapidly with lung volume as the air-fluid interface penetrates the mouth of the alveolus. At a value of P = approximately 20 cmH2O, the air-fluid interface is inside the alveolus and the lung is compliant, like an air-filled lung with constant surface tension. We conclude that the properties of the P-V curve of edematous lungs, particularly the knee in the P-V curve, are the result of the mechanics of parenchyma with constant surface tension and partially fluid-filled alveoli, not the result of abrupt opening of airways or atelectatic parenchyma.     

A two-component simulation model to teach respiratory mechanics

Interactive learning has been proven instrumental for the understanding of complex systems where the interaction of interdependent components is hard to envision. Due to the mechanical properties and mutual coupling of the lung and thorax, respiratory mechanics represent such a complex system, yet their understanding is essential for the diagnosis, prognosis, and treatment of various respiratory disorders. Here, we present a new mechanical model that allows for the simulation of respiratory pressure and volume changes in different ventilation modes. A bellow reflecting the "lung" is positioned within the inverted glass cylinder of a bell spirometer, which is sealed by a water lock and reflects the "thorax." A counterweight attached to springs representing the elastic properties of the chest wall lifts the glass cylinder, thus creating negative "pleural" pressure inside the cylinder and inflating the bellow. Lung volume changes as well as pleural and intrapulmonary pressures are monitored during simulations of spontaneous ventilation, forced expiration, and mechanical ventilation, allowing for construction of respiratory pressure-volume curves. The mechanical model allows for simulation of respiratory pressure changes during different ventilation modes. Individual relaxation curves constructed for the lung and thorax reflect the basic physiological characteristics of the respiratory system. In self-assessment, 232 medical students passing the physiology laboratory course rated that interactive teaching at the simulation model increased their understanding of respiratory mechanics by 70% despite extensive prior didactic teaching. Hence, the newly developed simulation model fosters students' comprehension of complex mechanical interactions and may advance the understanding of respiratory physiology.     

An assessment of dead space in pulmonary ventilation of the toad Bufo schneideri

The respiratory cycles of Rana and Bufo has been disputed in relation to flow patterns and to the respiratory dead-space of the buccal volume. A small tidal volume combined with a much larger buccal space motivated the "jet steam" model that predicts a coherent expired flow within the dorsal part of the buccal space. Some other studies indicate an extensive mixing of lung gas within the buccal volume. In Bufo schneideri, we measured arterial, end-tidal and intrapulmonary PCO(2) to evaluate dead-space by the Bohr equation. Dead-space was also estimated as: V(D)=(total ventilation-effective ventilation)/f(R), where total ventilation and f(R) were measured by pneumotachography, while effective ventilation was derived from the alveolar ventilation equation. These approaches were consistent with a dead space of 30-40% of tidal volume, which indicates a specific pathway for the expired lung gas.     

Effect of tidal volume and PEEP in ethchlorvynol-induced asymmetric lung injury.

We examined the effects of positive end-expiratory pressure (PEEP) and tidal volume on the distribution of ventilation and perfusion in a canine model of asymmetric lung injury. Unilateral right lung edema was established in 10 animals by use of a selective infusion of ethchlorvynol. Five animals were tested in the supine position (horizontal asymmetry) and five in the right decubitus position (vertical asymmetry). Raising PEEP from 5 to 12 cmH2O improved oxygenation despite a redistribution of blood flow toward the damage lung and a consistent decrease in total respiratory system compliance. This improvement paralleled a redistribution of tidal ventilation to the injured lung. This was effected primarily by a fall in the compliance of the noninjured lung due to hyperinflation. The effects of higher tidal volume were additive to those of PEEP. We propose that the major effect of PEEP in inhomogeneous lung injury is to restore tidal ventilation to a population of alveoli recruitable only at high airway pressures.     

Influence of lung volume and alveolar pressure on reverse pulmonary venous blood flow.

We have reported that left atrial blood refluxes through the pulmonary veins to gas-exchanging tissue after pulmonary artery ligation. This reverse pulmonary venous flow (Qrpv) was observed only when lung volume was changed by ventilation. This was believed to drive Qrpv by alternately distending and compressing the alveolar and extra-alveolar vessels. Because lung and pulmonary vascular compliances change with lung volume, we studied the effect of positive end-expiratory pressure (PEEP) on the magnitude of Qrpv during constant-volume ventilation. In prone anesthetized goats (n = 8), using the right lung to maintain normal blood gases, we ligated the pulmonary and bronchial arterial inflow to the left lung and ventilated each lung separately. A solution of SF6, an inert gas, was infused into the left atrium. SF6 clearance from the left lung was determined by the Fick principle at 0, 5, 10, and 15 and again at 0 cmH2O PEEP and was used to measure Qrpv. Left atrial pressure remained nearly constant at 20 cmH2O because the increasing levels of PEEP were applied to the left lung only. Qrpv was three- to fourfold greater at 10 and 15 than at 0 cmH2O PEEP. At these higher levels of PEEP, there were greater excursions in alveolar pressure for the same ventilatory volume. We believe that larger excursions in transpulmonary pressure during tidal ventilation at higher levels of PEEP, which compressed alveolar vessels, resulted in the reflux of greater volumes of left atrial blood, through relatively noncompliant extra-alveolar veins into alveolar corner vessels, and more compliant extra-alveolar arteries.     

Effects on body position and cholinergic blockade on mechanics of collateral ventilation

We studied the effects of position and cholinergic blockade on the mechanics of collateral ventilation in anesthetized paralyzed dogs. Resistance to collateral flow (Rcoll) is higher when an obstructed segment is dependent than when it is nondependent. Decreases of Rcoll in response to the local infusion of low oxygen mixtures are greater in dependent regions. We conclude that 1) changes in position affect Rcoll directly through local changes in lung volume related to the gradient of pleural pressure; 2) responses of collateral channels to local concentrations of CO2 and O2 are determined by ventilation perfusion relationships, which vary at different heights in the lung; and 3) resting cholinergic tone in the anesthetized dog varies at different heights in the lung.     

Pulmonary circulation in experimental pulmonary edema during diverse artificial respiration regimes

In acute experiments on cats with closed chest by ultrasonic method the authors studied the blood flow in low-lobar pulmonary artery and the vein, the blood pressure in pulmonary artery, lung vessels resistance in experimental pulmonary edema caused by intravenous infusion of mixture fatty acids at artificial ventilation of increased frequencies or volumes, at was shown, that artificial ventilation of increased frequencies in pulmonary edema reduces the pressure increase in pulmonary artery, lung vessels resistance and increases the blood flow in pulmonary artery and vein. Artificial ventilation of increased volumes produces more intense pressure increase in pulmonary artery and lung vessels resistance than in initial ventilation but the blood flow was slightly changed. The authors assume that artificial ventilation of increased frequencies or volumes in pulmonary edema due to pulmonary circulation change reduces the pulmonary edema intensity at the beginning.     

Abdominal distension alters regional pleural pressures and chest wall mechanics in pigs in vivo

Abdominal distension (AD) occurs in pregnancy and is also commonly seen in patients with ascites from various causes. Because the abdomen forms part of the "chest wall," the purpose of this study was to clarify the effects of AD on ventilatory mechanics. Airway pressure, four (vertical) regional pleural pressures, and abdominal pressure were measured in five anesthetized, paralyzed, and ventilated upright pigs. The effects of AD on the lung and chest wall were studied by inflating a liquid-filled balloon placed in the abdominal cavity. Respiratory system, chest wall, and lung pressure-volume (PV) relationships were measured on deflation from total lung capacity to residual volume, as well as in the tidal breathing range, before and 15 min after abdominal pressure was raised. Increasing abdominal pressure from 3 to 15 cmH2O decreased total lung capacity and functional residual capacity by approximately 40% and shifted the respiratory system and chest wall PV curves downward and to the right. Much smaller downward shifts in lung deflation curves were seen, with no change in the transdiaphragmatic PV relationship. All regional pleural pressures increased (became less negative) and, in the dependent region, approached 0 cmH2O at functional residual capacity. Tidal compliances of the respiratory system, chest wall, and lung were decreased 43, 42, and 48%, respectively. AD markedly alters respiratory system mechanics primarily by "stiffening" the diaphragm/abdomen part of the chest wall and secondarily by restricting lung expansion, thus shifting the lung PV curve as seen after chest strapping. The less negative pleural pressures in the dependent lung regions suggest that nonuniformities of ventilation could also be accentuated and gas exchange impaired by AD.     

Effect of positive-pressure ventilatory frequency on regional pleural pressure

Regional lung ventilation is modulated by the spatiotemporal distribution of alveolar distending forces. During positive-pressure ventilation, regional transmission of airway pressure (Paw) to the pleural surface may vary with ventilatory frequency (f), thus changing interregional airflow distribution. Pendelluft phenomena may result owing to selective regional hyperventilation or phase differences in alveolar distension. To define the effects of f on regional alveolar distension during positive-pressure ventilation, we compared regional pleural pressure (Ppl) swings from expiration to inspiration (delta Ppl) and end-expiratory Ppl over the f range 0-150 min-1 in anesthetized, paralyzed, close-chested dogs with normal lungs. We inserted six pleural balloon catheters to analyze Ppl distribution along three orthogonal axes of the right hemithorax. Increases in regional Ppl were synchronously coupled with inspiratory increases in Paw regardless of f. However, at a constant tidal volume and percent inspiratory time, end-expiratory Paw and Ppl increased in all regions once a f threshold was reached (P less than 0.01). Supradiaphragmatic delta Ppl were less than in other regions (P less than 0.05), but thoracoabdominal binding abolished this difference by decreasing thoracoabdominal compliance. We conclude that the distribution of forces determining dynamic regional alveolar distension are temporally synchronous but spatially asymmetric during positive-pressure ventilation at f less than or equal to 150/min.     

Dynamic lung compliance in newborn and adult cats

Static (Cstat) and dynamic (Cdyn) lung compliance and lung stress relaxation were examined in isolated lungs of newborn kittens and adult cats. Cstat was determined by increasing volume in increments and recording the corresponding change in pressure; Cdyn was calculated as the ratio of the changes in volume to transpulmonary pressure between points of zero flow at ventilation frequencies between 10 and 110 cycles/min. Lung volume history, end-inflation volume, and end-deflation pressure were maintained constant. At the lowest frequency of ventilation, Cdyn was less than Cstat, the difference being greater in newborns. Between 20 and 100 cycles/min, Cdyn of the newborn lung remained constant, whereas Cdyn of the adult lung decreased after 60 cycles/min. At all frequencies, the rate of stress relaxation, measured as the decay in transpulmonary pressure during maintained inflation, was greater in newborns than in adults. The frequency response of Cdyn in kittens, together with the relatively greater rate of stress relaxation, suggests that viscoelasticity contributes more to the dynamic stiffening of the lung in newborns than in adults. A theoretical treatment of the data based on a linear model of viscoelasticity supports this conclusion.     

Gas transport during high-frequency ventilation

During high-frequency small-volume ventilation (HFV), the transport rate of gas from the mouth to a lung region is a function of two conductances (conductance is the transfer rate of a gas divided by its partial pressure difference): regional longitudinal gas conductance along the airways (Grlongi) and gas conductance between lung regions (Ginter). Grlongi per unit regional lung (gas) volume [Grlongi/(Vr beta g)] was determined during HFV in 11 anesthetized paralyzed dogs lying supine. The distribution of Grlongi/(Vr beta g) was nearly uniform during HFV when stroke volumes were less than approximately two-thirds of the Fowler dead-space volume. By contrast, the distribution of Grlongi/(Vr beta g) was nonuniform when the stroke volume exceeded approximately two-thirds of the Fowler dead-space volume and the oscillation frequency was 5 Hz. Gas conductance along the airways per unit lung gas volume [average Glongi/(V beta g)], for the entire lung, increased with stroke volume at all frequencies, but for a given product of oscillation frequency and stroke volume, the average Glongi/(V beta g) was greater when stroke volume was large and oscillation frequency was low. The average Glongi/(V beta g) increased with frequency up to a maximal value; the frequency at which the maximum occurred depended on the kinematic viscosity of the inspired gas mixture.