We studied the respiratory output in five subjects exposed to parabolic flights [gravity vector 1, 1.8 and 0 gravity vector in the craniocaudal direction (Gz)] and when switching from sitting to supine (legs bent at the knees). Despite differences in total respiratory compliance (highest at 0 Gz and in supine and minimum at 1.8 Gz), no significant changes in elastic inspiratory work were observed in the various conditions, except when comparing 1.8 Gz with 1 Gz (subjects were in the seated position in all circumstances), although the elastic work had an inverse relationship with total respiratory compliance that was highest at 0 Gz and in supine posture and minimum at 1.8 Gz. Relative to 1 Gz, lung resistance (airways plus lung tissue) increased significantly by 52% in the supine but slightly decreased at 0 Gz. We calculated, for each condition, the tidal volume changes based on the energy available in the preceding phase and concluded that an increase in inspiratory muscle output occurs when respiratory load increases (e.g., going from 0 to 1.8 Gz), whereas a decrease occurs in the opposite case (e.g., from 1.8 to 0 Gz). Despite these immediate changes, ventilation increased, going to 1.8 and 0 Gz (up to ≈23%), reflecting an increase in mean inspiratory flow rate, tidal volume, and respiratory frequency, while ventilation decreased (approximately −14%), shifting to supine posture (transition time ∼15 s). These data suggest a remarkable feature in the mechanical arrangement of the respiratory system such that it can maintain the ventilatory output with small changes in inspiratory muscle work in face of considerable changes in configuration and mechanical properties.
- respiratory mechanics
- respiratory control
- pulmonary ventilation
our laboratory has previously shown that microgravity represents a condition of minimum distortion of the respiratory system and, furthermore, that increasing the gravity vector in the craniocaudal direction (Gz) during parabolic flights or changing its orientation in the dorsoventral direction (shifting from head-up to supine) causes marked changes in the configuration and in the elastic properties of the lung and chest wall, as well as in their mechanical coupling (5, 6). This paper reports the first computation of respiratory work when the distortion of the respiratory system is minimal (i.e., during microgravity) and provides a comparison with conditions in which the system is loaded by varying Gz. Because the metabolic demand remains unchanged on suddenly varying Gz, we wished to determine how the respiratory activity is controlled after abrupt loading or unloading. We approached the question by evaluating how a given energy output delivered by the inspiratory muscles is being used when the mechanical load imposed on the respiratory system is varied and evaluated how this impacts on the ventilatory output. Our analysis indicates that an immediate control is operated to maintain ventilation within narrow limits. In particular, an overshoot in ventilation is buffered on unloading, and, conversely, an increase in inspiratory energy output prevents excessive decrease in ventilation on loading.
All experiments were performed during three European Space Agency-Centre National Exploration Spatiale campaigns of parabolic flights in the period between October 1999 and April 2000. Each campaign consisted of 3 flight days. We used an Airbus A300 aircraft for each flight, which included 30 parabolas (90 parabolas per campaign) and lasted 2.5–3 h per flight. The steady horizontal flight corresponds to the vector 1 Gz and changes during the parabolic flight. During the aircraft pull-up, an acceleration of 1.8 Gz was reached. Subsequently, engine thrust was reduced, allowing the aircraft to enter a free-falling parabolic trajectory, which generated a 0-Gz phase, and, during pull-out, another 1.8-Gz phase was reached. All three phases lasted ∼20 s.
Respiratory variables [lung volume and esophageal pressure (Pes)] were obtained during steady horizontal flight and during short periods of microgravity and hypergravity in four male (age: 53 ± 2 yr; weight: 74 ± 3 kg; height: 174 ± 1 cm) and one female (age: 40 yr; weight: 61 kg; height: 167 cm) subjects. All subjects were healthy nonsmokers with no preexisting pulmonary disease. The same subjects were also studied during ground experiments in a sitting and supine posture with the use of the same equipment. The subjects had previous parabolic flight experience, were trained to perform the respiratory maneuvers, and were well accustomed to abrupt changes in Gz, which could occur several times during each flight. This same subject set had previously performed studies on how changes in Gz affect chest wall and lung mechanics (5, 6). Subjects gave their informed consent, and the protocol was approved by a review board.
Experimental equipment and system.
Subjects were seated in a body plethysmograph made of wood (empty volume of 360 liters), which was equipped with a pneumothachograph and transducers to measure pressure changes in the box and at the mouth (Pm). Panting maneuvers were performed by using a mouthpiece provided with an electromagnetic shutter. We initially performed parabolas with the transducers alone to evaluate the response of the transducer signals to changes in acceleration. The transducers were orientated along the aircraft's transverse axis to minimize the effect of changes in aircraft accelerations on both transducers. Lung volumes were measured by integrating the flow signal. Pes was derived from a pressure transducer mounted on a Gaeltec CTO-2 catheter (2-mm external diameter). Transducer sensitivity and linear pressure ranges were 5 μV·V−1·mmHg−1 and ±300 mmHg, respectively. The subjects were trained to advance the catheter through their nose until the location of the esophageal recording site was the one determined during preliminary on-ground experiments (on, average, ∼15 cm below the jugular notch, which roughly corresponds to the apex of the lung). The location of the esophageal transducer was chosen to minimize cardiac artifact and stabilize the pressure signal.
The pneumotachograph response was linear for flow rates compatible with the respiratory maneuvers performed with a maximum error of 5% at high-flow rates (∼3 l/s). All signals were sampled with an analog-to-digital converter (Digimétrie; 50 Hz/channel). Online analysis was performed to quantify the lung volumes from the pneumotachograph flow and plethysmograph pressure signals. The current lung volume, pressure variables, and Gz were monitored on a video screen during the experiment.
Before take off, calibration of the plethysmograph was performed by using a 2-liter syringe. A syringe volume control was custom made for each subject to be used during the flight. Calibration for the body box and Pm transducers was carried out by using a water manometer. Cabin pressure tended to decrease during the ascending phase and to increase during the descending phase; hence, the plethysmograph would overestimate lung volume during the ascending phase and underestimate lung volume during the descending phase. Cabin pressure was manually checked and corrected during the parabola for mismatches in pressure. From the 30 parabolas that were checked, the overall change in lung volume during the 0-Gz phase due to mismatch in pressure correction averaged 0.029 ± 0.27 liters, or 0.6% of vital capacity (VC) (a nonsignificant underestimate).
The Pes transducers were calibrated by using a calibration chamber, which could set the pressure by using water manometers. The sensitivity of the transducers and zero drift at atmospheric pressure were recorded. The sensitivity of the transducer was independent of temperature, whereas the zero drift was slightly dependent on temperature. The zero value corresponding to body temperature was obtained by withdrawing the probe at the end of each experimental session. These zero values were then used to correct Pes previously recorded.
Protocols for in-flight experiments.
Subjects were seated inside the plethysmograph while breathing through the mouthpiece. During 0-Gz exposure, the subjects would float up due to the changing trajectory of the aircraft. To counteract the effect, the subjects were secured to their seat with straps at their thighs and feet. Loose bands around the arms kept the arms positioned parallel to the chest. The time frame for data acquisition during respiratory maneuvers started in the last minute of level flight (1 Gz): pull up (1.8 Gz, 20 s), injection (0 Gz, 20 s), pull out (1.8 Gz, 20 s), and level flight again (1 Gz). The subjects were asked to breathe quietly throughout the various phases of the parabolic flight. The subjects were also asked to perform a VC and panting maneuvers to measure total gas volume (TGV) at the onset of the 1-Gz phase, before the parabola, and after returning to 1 Gz after the parabola. The total number of parabolas necessary to gather a complete set of data for each subject varied from 15 to 20 times.
Protocol for ground experiments.
Ground experiments were performed, in the seated and supine position, on the same subjects, adopting the same protocol and equipments as during the in-flight experiments. The change in posture was obtained by leaning the plethysmograph backward; this implied that legs remained as in the sitting posture. This change of position was completed in ∼15 s. This time is slightly longer than that corresponding to the in-flight changing Gz, which was on the order of ∼5 s.
TGVs were computed from Boyle's law: where Pc is the in-flight cabin pressure and ΔV/ΔPm is the ratio of change (Δ) in thoracic volume (V) to the change in alveolar pressure (Pm) during panting maneuvers. The ratio was inferred from the slope of the linear regression between volume and Pm. The drift of the volume measurement was initially subtracted from the volume signal of the panting maneuvers so that regression coefficients were 0.99, suggesting an accurate TGV measurement.
Lung volume recorded throughout the time frame also displayed a drift because of increasing temperature inside the plethysmograph. The lung volumes were obtained after correcting for volume drift between two successive TGV values, assuming a linear drift with time. Pes data were corrected for the zero drift on withdrawal of the catheter at the end of the session. A “moving average filter” that uses a moving window of 30 samples was employed to reduce high-frequency noise in the pressure records.
Lung resistance (Rl) (airway plus lung tissue resistance) was calculated according to Mead and Whittenberger (18).
The number of respiratory cycles for each in-flight phase was four to five. The total number of respiratory cycles considered for 0 Gz ranged from 60 to 100.
Each respiratory cycle analyzed was normalized to 100 time interval units. Readings of tidal volume (Vt) and Pes were done at each interval unit.
Values, if not otherwise specified, are presented as means ± SE. A nonparametric Wilcoxon’s test determined statistical significance of changes between different conditions. Significance was taken as P < 0.05.
Figure 1 shows a schema of the volume-pressure relationships of the chest wall and the lung for volumes ranging between 10 and 80% of the VC. The lung curve is presented by plotting volumes as a function of pleural pressure. This figure defines the various components of respiratory work during quiet breathing for a Vt beginning at resting volume of the respiratory system [functional residual capacity (FRC)], indicated by point A. The area between points ACEFA corresponds to the work required to overcome the elastic energy of the lung during inspiration. The area AEFA corresponds to the elastic energy released by the chest wall as it approaches its mechanical resting point, indicated by point E, during inspiration. On considering the inspiratory work (Wi), its net elastic component is, therefore, given by the difference between areas ACEFA and AEFA. Areas ABCA and ACDA correspond to the inspiratory and expiratory resistive work, respectively. The total Wi (elastic plus resistive) is given by the area ABCEA. Figure 2 presents the average volume-pressure relationships (obtained as described in Ref. 5) of the lungs and chest wall and the average volume pressure loops at 1, 1.8, and 0 Gz, and in the supine posture for all the subjects. During quiet breathing, the loops begin at the intersection between the lung and chest wall volume-pressure curves, namely FRC, which corresponds to the mechanical resting point of the respiratory system. For these same subjects, previous studies (5, 6) showed that a change in gravity and posture induced a change in the mechanical properties of both lung and chest wall and, consequently, a displacement of FRC.
As shown in Table 1, total Wi (elastic plus resistive) changes slightly when altering the gravity or posture of the subject, with a statistically significant increase (∼33%) only when passing from 1 to 1.8 Gz. Significant changes in resistive Wi were found only when passing from 1 to 0 Gz. Inspiratory resistive work accounted for a share of the total Wi, ranging from a minimum of 17% at 1 Gz to a maximum of 32% at 0 Gz. Table 1 also shows that there were no significant changes observed in resistive expiratory work.
Changes were expected in the various components of the elastic energy as changing gravity and posture were shown to influence the volume-pressure curves of the lung and chest wall, as well as lung volume at FRC. FRC decreased at 0 Gz and in the supine posture, relative to 1 and 1.8 Gz (5). As shown in Fig. 3A, at 0 Gz, the work necessary to overcome the elastic energy of the lung (area ACEF) was greatly and significantly reduced due to the decrease in lung recoil pressure (rightward displacement of the lung volume-pressure curve). In addition, the elastic energy released by the chest wall (area AEF) was significantly reduced due to the rightward shift of the volume-pressure curve. When changing from sitting to supine, despite a leftward displacement of the lung volume-pressure curve, the energy necessary to overcome the pulmonary elastic component was decreased, due to a marked decrease in FRC [∼1,200 ml (5)]. Furthermore, in the supine posture, the elastic energy released by the chest wall was significantly reduced, as shown by the rightward shift of the volume-pressure curve. Thus, as a first approximation, the condition at 0 Gz is comparable to supine, although the mechanisms are in part similar (rightward displacement of chest wall PV curve) and in part different (much larger decrease in FRC in supine). On average, in the supine posture, the net elastic Wi remained essentially equal to that at 1 Gz. Finally, at 1.8 Gz, no consistent changes in elastic properties of lung and chest wall were observed nor in FRC, and, accordingly, the elastic components of Wi were essentially unchanged relative to 1 Gz. Thus, again to a first approximation, 1 Gz is equivalent to 1.8 Gz.
We explored a possible relationship between total elastic Wi and total respiratory compliance (Ct). The Ct was calculated as Ct = (Cw + Cl)/(Cw·Cl), where Cw and Cl are the chest wall and lung compliance values, respectively, determined in a previous study on the same set of data (5, 6). Values for Ct are reported in Table 1. We show in Fig. 3B that total elastic Wi decreases with increasing Ct.
Rl values are reported in Table 1 and show that Rl increases significantly in the supine posture relative to all other conditions.
Pattern of breathing.
Table 2 summarizes data on the timing and pattern of breathing and the corresponding changes in pulmonary ventilation as given by where V̇e is ventilation, Ti is inspiratory time, Tt is total time, and Ti/Tt is commonly referred to as the duty cycle. V̇e varied in the conditions studied as a combination of changing both Vt/Ti and duty cycle (Fig. 4A). Considering V̇e = f·Vt, where f is the respiratory frequency, one can appreciate that changes in ventilation are accomplished by parallel changes in Vt and f (Fig. 4B). Therefore, data from Fig. 4 indicate that resting pulmonary ventilation is affected by changing gravity and posture.
Because, under steady-state conditions during quiet breathing, the inspiratory muscles deliver a certain amount of work in a cyclic fashion at each inspiration, one could ask how a given neuromuscular output available in a given condition is being used when the mechanical load imposed to the respiratory system is varied. Considering Vt as a dependent controlled variable, we computed how Vt should be modified when passing from one condition to the next, depending on the changes in Rl and Ct (e.g., going from 1 to 1.8 Gz) and assuming a constant value for the total Wi.
Wi can be expressed as function of Rl, Ct, and volume V, considering the classic equation of motion for breathing (assuming that the effect of inertial forces during quiet breathing was negligible): (1) were P is transrespiratory pressure, V is the volume displacement from FRC, R is resistance, and V̇ is the flow of the airway opening. Ct and Rl are assumed to be constant during quiet breathing. From the differential expression for work dW = P dV, substituting the expression for P from Eq. 1, supposing that the lung volume during quiet breathing can be approximated by a sine wave (8), and integrating over an inspiration, the Wi can be expressed as: (2) One can now calculate how Vt should change when passing from one condition to the next as a function of changes in Rl and Ct (e.g., going from 1 to 1.8 Gz), assuming a constant value for Wi. In fact, the predicted Vt in the new condition can be obtained solving Eq. 2 for Vt and using for Rl and Ct the values relative to the new condition. Accordingly, the expression of Vt is: (3) In this equation, Wi is the value computed from Eq. 2 using Rl and Ct relative to the reference condition.
Figure 5 shows the results of the analysis. When switching from 1 to 1.8 Gz, the energy output of the inspiratory muscles available at 1 Gz would cause Vt to decrease mostly due to the decrease in Ct, yet the actual Vt at 1.8 Gz was slightly larger than at 1 Gz, suggesting an increase in energy output. When switching from 1.8 to 0 Gz, the respiratory output available at 1.8 Gz would cause a large increase in Vt, but, because the actual increase was much less, this suggests that a decrease in energy output has occurred. When returning from 0 to 1.8 Gz, a marked decrease in Vt would be expected, but the actual decrease is much less (no difference was found in pattern of breathing between the rising and descending phase at 1.8 Gz), and, therefore, an increase in energy output should have occurred. Returning from 1.8 to 1 Gz would deliver energy to cause an increase in Vt, but, because the actual Vt returns toward the control value, a decrease in total output should have occurred. Moving from sitting to supine would be expected to cause a decrease in Vt for the same energy expenditure, mostly due to the decrease in Rl; however, the observed decrease in Vt was larger than expected.
The pattern of breathing is very regular in the resting condition, reflecting a stereotypical model for a given metabolic demand. The action of inspiratory muscles allows the lung to overcome resistances and elastic recoil while the chest wall frees elastic energy as volume increases. During quiet breathing, the elastic energy stored by the lung on inspiration is dissipated, allowing expiration to proceed passively (Fig. 2). Previous studies showed that an abrupt change in modulus and direction of the gravity vector modified the mechanical resting point of the respiratory system (FRC) (5) due to a change in the lung-chest wall recoil balance. However, in all of our subjects (the same on whom respiratory mechanics was previously studied), the end-expiratory volume corresponded in each condition to FRC: accordingly, although we did not record inspiratory and expiratory EMG muscle activity, this suggests that, in all subjects, expiration was largely based on elastic energy release, such as occurs when breathing quietly.
Changes in resistive and elastic work.
FRC is lower in supine than at 0 Gz, and this can partly explain the increase in Rl (Fig. 6), as a hyperbolic relationship has been described between Rl and lung volume (7). One may interpret this finding, considering that the deformation of the lung due to gravity should be reduced in weightlessness, and, therefore, all lung regions are more uniformly expanded (17, 19). In this condition, the contribution of the heterogeneity of time constants throughout the lung to total Rl should be reduced (20), resulting in a lower Rl around breathing frequencies.
On average, the percentage of contribution of the resistive work does not exceed ∼30% of the total Wi. Therefore, changes in the resistance due to changes in Gz affect the total respiratory work only marginally.
Because the elastic properties of lung and chest wall are affected by changing Gz, the lung and chest wall components to the elastic work are also significantly changed (Fig. 3A).
Changes in ventilatory pattern and control of breathing.
The data from Fig. 5 allow discussion of the complex matching between the work of breathing and control of the neuromuscular respiratory output in response to loading and unloading of the respiratory muscles. These data were obtained by applying the model described by Eqs. 2 and 3, based on three important assumptions: 1) the hypothesis of sinusoidal Vt; 2) the equation of motion (Eq. 1) accurately describes the dynamics of the respiratory system; and 3) the contribution of chest wall resistance is negligible. To validate our model, we compared the measured Vt to the computed Vt obtained by solving Eq. 3 using the mechanical properties and the breathing pattern for the same condition. Wi and compliance values were measured on the volume-pressure curves, the Rl was measured by using the Mead and Whittenberger method, and the f was measured on volume traces. The validity of the model was satisfactory, as the predicted Vt values differed by no more than 9% from the measured Vt in each condition.
We found that increasing the load results in a larger inspiratory output; however, the resultant Vt may either increase (switching from 1 to 1.8 Gz) or decrease (switching from 0 to 1.8 Gz), reflecting a greater decrease in the respiratory compliance in the latter case. Unloading always induces a decrease in inspiratory output: the resulting Vt decreased from 2 to 1 Gz but increased moving from 2 to 0 Gz, reflecting a larger increase in respiratory compliance. A decrease in respiratory neural drive was also found on immersion in water up to the xiphoid process, a situation mimicking the exposure to microgravity because it counteracts the weight of the abdomen (23). These immediate responses occurred through combined and coordinated modifications in Vt and in inspiratory flow rate, with minor changes in duty cycle and f (Fig. 4).
Because the metabolic demand is unchanged on quickly varying Gz and posture, the question arises as to how the breathing pattern is affected when the operational features of the respiratory muscles are modified. Despite changes in elastic features and configuration, the respiratory system is allowed to return to its resting mechanical point on expiration; therefore, this suggests that the control of breathing pattern mainly acts on inspiratory muscles, namely external intercostals and diaphragm. Although these two groups of muscles work together to operate the respiratory pump, they are affected differently by force-length properties due to changes in configuration, in particular considering the abdominal and the rib cage contribution to total lung volume. Furthermore, they are also subject to a different reflex control from proprioceptors as external intercostals are rich in spindles, whereas the diaphragm is rich in Golgi tendon organs (24).
We did not find significant differences between the tidal breaths within the 20 s of a given Gz exposure. This suggests that the response in the breathing pattern is accomplished within a short time (less than one breath) of changing the mechanical properties of the respiratory system. This prompt respiratory response is compatible with the short time constants of the control mechanisms.
Previous studies considered how the respiratory pattern is affected by changing the respiratory load through a decrease (breathing He-O2 mixture) or an increase in airway resistance (returning to air breathing) (15). The immediate response to loading (returning from He to room air breathing) consisted of an increase in Vt and ventilation, which is in line with our findings when moving from 1 to 1.8 Gz. The immediate response to unloading (switching from room air to He breathing) was again an increase in ventilation due to an increase in frequency but a decrease in Vt. We found that, when unloading from 1.8 to 1 G2 (with a 23% increase in compliance), both ventilation and Vt decreased. Conversely, when unloading from 1.8 to 0 Gz (with an increase in compliance as large as 71%), we found that both ventilation and Vt increased.
The use of He-O2 mixture allows changes only in pulmonary resistance to ∼20% (9). In our case, loading and unloading were mainly due to changes in respiratory compliance, with negligible effects on Rl. Therefore, during the He experiments, the change of the load applied to inspiratory muscles is mainly proportional to the inspiratory flow. Conversely, when changing gravity, the load to inspiratory muscles is mainly proportional to the lung volume. Given these differences, it is conceivable to hypothesize that there are also differences in afferent input and reflex control between the two conditions.
The overshoot in Vt on unloading was also observed on removal of external resistances in both animals (15, 16) and humans (1). This effect can be explained by extending our hypothesis to say that a given inspiratory output necessary to overcome either elastance (mostly our case) or resistance (1, 15, 16) may result in some overshoot of ventilation when loading is suddenly removed, despite some immediate control.
A volitional component in the respiratory response could be invoked, as our subjects were aware of the incoming condition due to a repetitive experimental protocol. However, we may comment that they were starting inspiration always at FRC under the various conditions, despite considerable changes in configuration of the respiratory system, suggesting that they were breathing “quietly”; accordingly, these considerations would rule out a volitional component.
Comparison to previous in-flight data and to supine and water immersion.
A previous study on parabolic flights showed no difference in Vt between 1 and 1.8 Gz (21) and a small, although insignificant, increase in Vt at 0 Gz compared with control (11). Interestingly, during sustained microgravity, Vt was found to be significantly decreased relative to preflight standing control but also, although not significantly, relative to supine (12). On comparing 0-Gz acute exposure to supine, the former leads to hyperventilation relative to control, whereas the latter leads to hypoventilation (Fig. 4 and Table 2), yet the two conditions share various similarities. Switching from 1 to 0 Gz or from standing to supine position causes 1) a cranial displacement of the diaphragm, modifying both lung volumes and chest wall configuration (21, 25), 2) a decrease of inspiratory muscle activity (10, 13), 3) a blood shift from lower body to thorax (14, 22), and 4) a decrease in FRC.
The decrease in ventilation when shifting from sitting to supine confirms what has been previously found (2, 4) and could be replicated by water immersion up to the xyphoid (2). One may postulate that the increase in Vt observed at 0 Gz during parabolic flights might relate to the specific conditions occurring during the flight. To explore this possibility, one could estimate what the effect of 0 Gz exposure would be, were it possible to shift directly from 1 Gz without going through the 1.8-Gz phase. In this hypothetical case, one may calculate (Eq. 3), based on the available energy output at 1 Gz, that Vt would increase to ∼1.066 liters at 0 Gz, considering only the changes in the mechanical properties. Assuming now a decrease in the energy output equal to that occurring when going from 1.8 to 0 Gz, the calculated Vt at 0 Gz would be reduced to 0.939 liter. Finally, considering the mechanical properties occurring in the supine position, one would further reduce Vt to 0.710 liter, a value similar to that measured in the supine posture (0.718 liter). This is in agreement with data on ventilatory response in sustained microgravity (12), although this comparison should be taken with reservation, as the metabolic level may not be the same.
Why the conditions of sustained microgravity and supine posture lead to reduction in ventilation remains to be explained. We wish to recall the hypothesis put forward by Anthonisen et al. (2) back in 1965. They reasoned that the relative hyperventilation in erect posture is based on gravity-dependent changes in brain perfusion: moving from supine to erect would cause a decrease in brain blood flow and a local brain tissue increase in CO2 partial pressure for the same metabolic demand. This would trigger a reflex hyperventilation that lowers alveolar CO2 pressure to increase brain-blood oxygenation (2).
It appears difficult to reconcile in a model the complex interaction between passive muscle properties (force-length), afferents of opposite sign (excitatory from spindles and inhibitory from tendon organs), differences in receptor stimulation threshold, and, possibly, volitional component in response to sudden changes in the mechanical properties of the respiratory system. It appears, however, that a compensatory reflex would readjust the overall muscle output so as to match the need of force generation to changing loads. This concept integrates the idea of “operational length compensation” (3), which proposes a readjustment of neural drive when the force-length characteristics have been modified.
It appears interesting to note a remarkable feature of the mechanical arrangement of the respiratory system. In fact, within the Vt range, the changes in lung and chest wall elastic work are similar on changing the gravity vector, but, because the two structures operate in opposite directions, the resultant change in Wi and in ventilation appears relatively buffered.
P. Vaida was the recipient of grants 793/1999/CNES/7660 and 793/2000/CNES/8147; G. Miserocchi was the recipient of a research grant from Agenzia Spaziale Italiana.
The authors are grateful to Nora Tgavalekos for revising the manuscript.
The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
- Copyright © 2004 the American Physiological Society