Effects of hypergravity on the distributions of lung ventilation and perfusion in sitting humans assessed with a simple two-step maneuver

First-time imaging of effects of inspired oxygen concentration on regional lung volumes and breathing pattern during hypergravity

PURPOSE

Aeroatelectasis can develop in aircrew flying the latest generation high-performance aircraft. Causes alleged are relative hyperoxia, increased gravity in the head-to-foot direction (+Gz), and compression of legs and stomach by anti-G trousers (AGT). We aimed to assess, in real time, the effects of hyperoxia, +Gz accelerations and AGT inflation on changes in regional lung volumes and breathing pattern evaluated in an axial plane by electrical impedance tomography (EIT).

METHODS

The protocol mimicked a routine peacetime flight in combat aircraft. Eight subjects wearing AGT were studied in a human centrifuge during 1 h 15 min exposure of +1 to +3.5Gz. They performed this sequence three times, breathing AIR, 44.5 % O2 or 100 % O2. Continuous recording of functional EIT enabled uninterrupted assessment of regional lung volumes at the 5th intercostal level. Breathing pattern was also monitored.

RESULTS

EIT data showed that +3.5Gz, compared with any moment without hypergravity, caused an abrupt decrease in regional tidal volume (VT) and regional end-expiratory lung volume (EELV) measured in the EIT slice, independently of inspired oxygen concentration. Breathing AIR or 44.5 % O2, sub-regional EELV measured in the EIT slice decreased similarly in dorsal and ventral regions, but sub-regional VT measured in the EIT slice decreased significantly more dorsally than ventrally. Breathing 100 % O2, EELV and VT decreased similarly in both regions. Inspired tidal volume increased in hyperoxia, whereas breathing frequency increased in hypergravity and hyperoxia.

CONCLUSIONS

Our findings suggest that hypergravity and AGT inflation cause airway closure and air trapping in gravity-dependent lung regions, facilitating absorption atelectasis formation, in particular during hyperoxia.

Sustained mild hypergravity reduces spontaneous cardiac baroreflex sensitivity

Head-to-foot gravitational force >1G (+Gz hypergravity) augments venous pooling in the lower body and reduces central blood volume during exposure, compared with 1Gz. Central hypovolemia has been reported to reduce spontaneous cardiac baroreflex sensitivity. However, no investigations have examined spontaneous cardiac baroreflex sensitivity during exposure to sustained mild +Gz hypergravity. We therefore hypothesized that mild +Gz hypergravity would reduce spontaneous cardiac baroreflex sensitivity, compared with 1Gz. To test this hypothesis, we examined spontaneous cardiac baroreflex sensitivity in 16 healthy men during exposure to mild +Gz hypergravity using a short-arm centrifuge. Beat-to-beat arterial blood pressure (tonometry) and R-R interval (electrocardiography) were obtained during 1Gz and 1.5Gz exposures. Spontaneous cardiac baroreflex sensitivity was assessed by sequence slope and transfer function gain. Stroke volume was calculated from the arterial pressure waveform using a three-element model. All indices of spontaneous cardiac baroreflex sensitivity decreased significantly (up slope: 18.6±2.3→12.7±1.6ms/mmHg, P<0.001; down slope: 19.0±2.5→13.2±1.3ms/mmHg, P=0.002; transfer function gain in low frequency: 14.4±2.2→10.1±1.1ms/mmHg, P=0.004; transfer function gain in high frequency: 22.2±7.5→12.4±3.5ms/mmHg, P<0.001). Stroke volume decreased significantly (88±5→80±6ml, P=0.025). Moreover, although systolic arterial pressure variability increased, R-R interval variability did not increase. These results suggest that even mild +Gz hypergravity reduces spontaneous cardiac baroreflex sensitivity, increasing the risk of cardiovascular disturbance during the exposure.

Regional lung ventilation in humans during hypergravity studied with quantitative SPECT

Recently we challenged the view that arterial desaturation during hypergravity is caused by redistribution of blood flow to dependent lung regions by demonstrating a paradoxical redistribution of blood flow towards non-dependent regions. We have now quantified regional ventilation in 10 healthy supine volunteers at normal and three times normal gravity (1G and 3G). Regional ventilation was measured with Technegas ((99m)Tc) and quantitative single photon emission computed tomography (SPECT). Hypergravity caused arterial desaturation, mean decrease 8%, p<0.05 vs. 1G. The ratio for mean ventilation per voxel for non-dependent and dependent lung regions was 0.81±0.12 during 1G and 1.63±0.35 during 3G (mean±SD), p<0.0001. Thus, regional ventilation was shifted from dependent to non-dependent regions. We suggest that arterial desaturation during hypergravity is caused by quantitatively different redistributions of blood flow and ventilation. To our knowledge, this is the first study presenting high-resolution measurements of regional ventilation in humans breathing normally during hypergravity.

Pulmonary artery pulsatility is the main cause of cardiogenic oscillations

The genesis of cardiogenic oscillations, i.e. the small waves in airway pressure (COS(paw)) and flow (COS(flow)) signals recorded at the airway opening is under debate. We hypothesized that these waves are originated from cyclic changes in pulmonary artery (PA) pressure and flow but not from the physical transmission of heartbeats onto the lungs. The aim of this study was to test this hypothesis. In 10 anesthetized pigs, COS were evaluated during expiratory breath-holds at baseline with intact chest and during open chest conditions at: (1) close contact between heart and lungs; (2) no heart-lungs contact by lifting the heart apex outside the thoracic cavity; (3) PA clamping at the main trunk during 10 s; and (4) during manual massage after cardiac arrest maintaining the heart apex outside the thorax, with and without PA clamping. Baseline COS(paw) and COS(flow) amplitude were 0.70 ± 0.08 cmH(2)O and 0.51 ± 0.06 L/min, respectively. Both COS amplitude decreased during open chest conditions in step 1 and 2 (p < 0.05). However, COS(paw) and COS(flow) amplitude did not depend on whether the heart was in contact or isolated from the surrounding lung parenchyma. COS(paw) and COS(flow) disappeared when pulmonary blood flow was stopped after clamping PA in all animals. Manual heart massages reproduced COS but they disappeared when PA was clamped during this maneuver. The transmission of PA pulsatilty across the lungs generates COS(paw) and COS(flow) measured at the airway opening. This information has potential applications for respiratory monitoring.

Effects of acceleration in the Gz axis on human cardiopulmonary responses to exercise

The aim of this paper was to develop a model from experimental data allowing a prediction of the cardiopulmonary responses to steady-state submaximal exercise in varying gravitational environments, with acceleration in the G(z) axis (a (g)) ranging from 0 to 3 g. To this aim, we combined data from three different experiments, carried out at Buffalo, at Stockholm and inside the Mir Station. Oxygen consumption, as expected, increased linearly with a (g). In contrast, heart rate increased non-linearly with a (g), whereas stroke volume decreased non-linearly: both were described by quadratic functions. Thus, the relationship between cardiac output and a (g) was described by a fourth power regression equation. Mean arterial pressure increased with a (g) non linearly, a relation that we interpolated again with a quadratic function. Thus, total peripheral resistance varied linearly with a (g). These data led to predict that maximal oxygen consumption would decrease drastically as a (g) is increased. Maximal oxygen consumption would become equal to resting oxygen consumption when a (g) is around 4.5 g, thus indicating the practical impossibility for humans to stay and work on the biggest Planets of the Solar System.

Microgravity decreases and hypergravity increases exhaled nitric oxide

Inhalation of toxic dust during planetary space missions may cause airway inflammation, which can be monitored with exhaled nitric oxide (NO). Gravity will differ from earth, and we hypothesized that gravity changes would influence exhaled NO by altering lung diffusing capacity and alveolar uptake of NO. Five subjects were studied during microgravity aboard the International Space Station, and 10 subjects were studied during hypergravity in a human centrifuge. Exhaled NO concentrations were measured during flows of 50 (all gravity conditions), 100, 200, and 500 ml/s (hypergravity). During microgravity, exhaled NO fell from a ground control value of 12.3 ± 4.7 parts/billion (mean ± SD) to 6.6 ± 4.4 parts/billion ( P = 0.016). In the centrifuge experiments and at the same flow, exhaled NO values were 16.0 ± 4.3, 19.5 ± 5.1, and 18.6 ± 4.7 parts/billion at one, two, and three times normal gravity, where exhaled NO in hypergravity was significantly elevated compared with normal gravity ( P ≤ 0.011 for all flows). Estimated alveolar NO was 2.3 ± 1.1 parts/billion in normal gravity and increased significantly to 3.9 ± 1.4 and 3.8 ± 0.8 parts/billion at two and three times normal gravity ( P < 0.002). The findings of decreased exhaled NO in microgravity and increased exhaled and estimated alveolar NO values in hypergravity suggest that gravity-induced changes in alveolar-to-lung capillary gas transfer modify exhaled NO.

Transient influence of end-tidal carbon dioxide tension on the postural restraint in cerebral perfusion

In the upright position, cerebral blood flow is reduced, maybe because arterial carbon dioxide partial pressure (PaCO2) decreases. We evaluated the time-dependent influence of a reduction in PaCO2, as indicated by the end-tidal Pco2 tension (PetCO2), on cerebral perfusion during head-up tilt. Mean arterial pressure, cardiac output, middle cerebral artery mean flow velocity (MCA Vmean), and dynamic cerebral autoregulation at supine rest and 70° head-up tilt were determined during free breathing and with PetCO2 clamped to the supine level. The postural changes in central hemodynamic variables were equivalent, and the cerebrovascular autoregulatory capacity was not significantly affected by tilt or by clamping PetCO2. In the first minute of tilt, the decline in MCA Vmean (10 ± 4 vs. 3 ± 4 cm/s; mean ± SE; P < 0.05) and PetCO2 (6.8 ± 4.3 vs. 1.7 ± 1.6 Torr; P < 0.05) was larger during spontaneous breathing than during isocapnic tilt. However, after 2 min in the head-up position, the reduction in MCA Vmean was similar (7 ± 5 vs. 6 ± 3 cm/s), although the spontaneous decline in PetCO2 was maintained ( P < 0.05 vs. isocapnic tilt). These results suggest that the potential contribution of PaCO2 to the postural reduction in MCA Vmean is transient, leaving the mechanisms for the sustained restrain in MCA Vmean to be identified.

Residual heterogeneity of intra- and interregional pulmonary perfusion in short-term microgravity

We hypothesized that the perfusion heterogeneity in the human, upright lung is determined by nongravitational more than gravitational factors. Twelve and six subjects were studied during two series of parabolic flights. We used cardiogenic oscillations of O2/SF6 as an indirect estimate of intraregional perfusion heterogeneity ( series 1) and phase IV amplitude (P4) as a indirect estimate of interregional perfusion heterogeneity ( series 2). A rebreathing-breath holding-expiration maneuver was performed. In flight, breath holding and expiration were performed either in microgravity (0 G) or in hypergravity. Controls were performed at normal gravity (1 G). In series 1, expiration was performed at 0 G. Cardiogenic oscillations of O2/SF6 were 19% lower when breath holding was performed at 0 G than when breath holding was performed at 1 G [means (SD): 1.7 (0.3) and 2.3 (0.6)% units] ( P = 0.044). When breath holding was performed at 1.8 G, values did not differ from 1-G control [2.6 (0.8)% units, P = 0.15], but they were 17% larger at 1.8 G than at 1 G. In series 2, expiration was performed at 1.7 G. P4 changed with gravity ( P < 0.001). When breath holding was performed at 0 G, P4 values were 45 (46)% of control. When breath holding was performed at 1.7 G, P4 values were 183 (101)% of control. We conclude that more than one-half of indexes of perfusion heterogeneity at 1 G are caused by nongravitational mechanisms.

Effects of gravity and blood volume shifts on cardiogenic oscillations in respired gas

During the cardiac cycle, cardiogenic oscillations of expired gas (x) concentrations (COS([x])) are generated. At the same time, there are heart-synchronous cardiogenic oscillations of airway flow (COS(flow)), where inflow occurs during systole. We hypothesized that both phenomena, although primarily generated by the heartbeat, would react differently to the cephalad blood shift caused by inflation of an anti-gravity (anti-G) suit and to changes in gravity. Twelve seated subjects performed a rebreathing-breath-holding-expiration maneuver with a gas mixture containing O2 and He at normal (1 G) and moderately increased gravity (2 G); an anti-G suit was inflated to 85 mmHg in each condition. When the anti-G suit was inflated, COS(flow) amplitude increased (P = 0.0028) at 1 G to 186% of the control value without inflation (1-G control) and at 2 G to 203% of the control value without inflation (2-G control). In contrast, the amplitude of COS of the concentration of the blood-soluble gas O2 (COS([O2/He])), an index of the differences in pulmonary perfusion between lung units, declined to 75% of the 1-G control value and to 74% of the 2-G control value (P = 0.0030). There were no significant changes in COS(flow) or COS([O2/He]) amplitudes with gravity. We conclude that the heart-synchronous mechanical agitation of the lungs, as expressed by COS(flow), is highly dependent on peripheral-to-central blood shifts. In contrast, COS([blood-soluble gas]) appears relatively independent of this mechanical agitation and seems to be determined mainly by differences in intrapulmonary perfusion.

Cardiovascular effects of anti-G suit inflation at 1 and 2 G

We sought to determine to which pressure a full-coverage anti-G suit needs to be inflated in order to obtain the same stroke volume during a brief exposure to twice the normal gravity (2 G) as that at 1 G without anti-G suit inflation. Nine sitting subjects were studied at normal (1 G) and during 20 s of exposure to 2 G. They wore anti-G suits, which were inflated at both G-levels to the following target pressures: 0, 70, 140 and 210 mmHg. Stroke volume was computed from cardiac output, which was measured by rebreathing. Heart rate and mean arterial pressure at heart level were recorded. Inflation to 70 mmHg compensated for the decrease in stroke volume and cardiac output caused by hypergravity. Mean arterial pressure at heart level was comparable at 1 G and at 2 G and increased gradually and similarly with inflation (P<0.001) at both gravity levels. Thus, anti-G suits act by increasing both preload and afterload but the two effects counteract each other in terms of cardiac output, so that cardiac output at 2 G is maintained at its 1 G level. This effect is reached already at 70 mmHg of inflation. Greater inflation pressure further increases mean arterial pressure at heart level and compensates for the increased difference in hydrostatic pressure between heart and head in moderate hypergravity.

Distributions of lung ventilation and perfusion in prone and supine humans exposed to hypergravity

When normal subjects are exposed to hypergravity [5 times normal gravity (5 G)] there is an impaired arterial oxygenation that is less severe in the prone compared with supine posture. We hypothesized that under these conditions the heterogeneities of ventilation and/or perfusion distributions would be less prominent when subjects were prone compared with supine. Expirograms from a combined rebreathing-single breath washout maneuver (Rohdin M, Sundblad P, and Linnarsson D. J Appl Physiol 96: 1470–1477, 2004) were analyzed for vital capacity (VC), phase III slope, and phase IV amplitude, to analyze heterogeneities in ventilation (Ar) and perfusion [CO2-to-Ar ratio (CO2/Ar)] distribution, respectively. During hypergravity, VC decreased more in the supine than in the prone position (ANOVA, P = 0.02). Phase III slope was more positive for Ar ( P = 0.003) and more negative for CO2/Ar ( P = 0.007) in the supine compared with prone posture at 5 G, in agreement with the notion of a more severe hypergravity-induced ventilation-perfusion mismatch in supine posture. Phase IV amplitude became lower in the supine than in the prone posture for both Ar ( P = 0.02) and CO2/Ar ( P = 0.004) during hypergravity as a result of the more reduced VC in the supine posture. We speculate that results of VC and phase IV amplitude are due to the differences in heart-lung interaction and diaphragm position between postures: a stable position of the heart and diaphragm in prone hypergravity, in contrast to supine in which the weight of the heart and a cephalad shift of the diaphragm compress lung tissue.