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.
- estimates of alveolar NO
- analysis of exhaled NO
- human centrifuge
- lung diffusion
- respiratory system
exhaled nitric oxide (NO) is increasingly used to evaluate the inflammatory state of the lung. Endogenous NO in the lung results from the activity of three isoforms of NO synthase (NOS), one of them being triggered by inflammation. In future spaceflights, humans will risk being exposed to toxic dust on the moon and Mars, which may cause airway inflammation (12). Analysis of exhaled NO (FeNO) is a simple method to monitor airway inflammation not only in asthma (3, 8) but also after dust inhalation (e.g., Refs. 16, 27). In addition, estimates of alveolar NO (CalvNO) can be obtained from FeNO measurements (10, 20). In preparation for possible future monitoring of lung health during long-term space missions, we wanted to investigate the effects of weightlessness (microgravity) and gravity per se on FeNO.
Gravity influences the blood distribution within the lung (18), which in turn will induce changes in the rate of alveolar-to-lung capillary transfer [lung diffusing coefficient (Dl)] of blood-soluble gases such as NO and carbon monoxide (CO). This is because alveolar-to-blood transfer of gas is highly dependent of the magnitude of the contact area between the blood and the alveolar air. Thus Prisk et al. (19) and Verbanck et al. (32) have shown that, in zero gravity, Dl for carbon monoxide (DlCO) is enhanced by 11–27%. For very short-lasting microgravity, Vaida et al. (30) showed that both DlNO and the membrane component of DlCO were increased by >40%. Furthermore, Rohdin et al. (21) showed that DlCO is decreased by 21% at two times normal gravity (2 G) and by 34% at three times normal gravity (3 G) (21). We therefore exposed healthy subjects to both decreases and increases of gravity and studied their exhaled NO. We hypothesized that there would be reduced FeNO levels in microgravity due to increased alveolar uptake of NO, caused by an increased lung diffusing capacity (19, 30, 32). Mass and time constraints typical of space experimentation would not allow other than simple FeNO0.05 (exhalation at 50 ml/s) measurements in space. We therefore performed parallel experiments on ground in normal and increased gravity with more complex equipment. We reasoned that such parallel studies of the effects of gravity on FeNO and related parameters would help in the analysis of the space data. As a corollary of the effects of weightlessness, we hypothesized that not only FeNO but also estimates of CalvNO would be increased in increased gravity due to a reduced Dl (21) and a resulting slowed alveolar uptake of NO.
The experiments were approved by the European Space Agency (ESA) Medical Board (070-4, 092, 117-6, 133-3, and 147-7) and by the Regional Ethical Review Board in Stockholm (2005/507-31 and 2005/658-31).
This part of the study was performed onboard the International Space Station (ISS) during the period 2005–2008. Five healthy male ESA astronauts and Russian cosmonauts were studied. All were nonsmokers and had no history of asthma or other inflammatory airway diseases. Age, body mass, and height ranged from 34 to 52 yr, 68 to 78 kg, and 1.72 to 1.82 m, respectively.
Portable NO analyzers (NIOX MINO, Aerocrine, Solna, Sweden) (9, 29) were procured from Aerocrine and were certified for space use (Damec Research Aps, Odense, Denmark). Functionally equivalent instruments were used on ground for training and preflight controls. The NIOX MINO analyzer is equipped with an exchangeable NO sensor, and both the analyzer and the sensor have an electronically controlled lifetime with preprogrammed termination dates. The semi-disposable NO sensor is precalibrated at manufacturing to ensure correct measurements within the specified lifetime. Compliance control was repeated by independent experts before use. When the expiry date is reached, the hardware goes into a lock-up state, and no further measurements can be made.
Training and preflight measurements (1 G) were performed in Russia or in Germany. Postflight measurements (1 G) were made in Russia or in the US. Astronauts performed at least four control measurements at one to three occasions before the spaceflight and then approximately every 6 wk during their 23- to 28-wk-long stays onboard the ISS. Here, the gravitational force of the earth is counterbalanced by the centrifugal force resulting from the circular trajectory of ISS (weightlessness or microgravity). After return to earth, they performed daily measurements during the first week after landing.
The subjects performed the FeNO maneuver sitting and in duplicates. Since ingested food and beverages rich in nitrite and nitrate have shown to affect exhaled NO (33), the subjects had to refrain from food and beverages rich in nitrite and nitrate 24 h before the tests. They rinsed their mouths with water before each test. The subjects performed the following respiratory maneuver: 1) exhalation to functional residual capacity (FRC), 2) inhalation of NO-free air from the NIOX MINO to close to total lung capacity (TLC), and 3) controlled exhalation into the NIOX MINO during 10 s at a flow rate of 50 ml/s. The expiratory pressure was kept between 1.0 and 2.0 kPa to keep the soft palate closed and hence avoid nasal contamination of the exhalate. The measurements fully complied with the joint ERS/ATS recommendations (1).
Data acquisition and analysis.
During the last 3 s of the 10-s-long expiration of the FeNO maneuver, the last 3-s portion of the exhaled air is stored in a buffering unit from where the sample is led to the NO sensor during the following 100 s of analysis. The NO concentration result is then presented on the LCD screen of the analyzer. The readings are stored on individual memory cards for later assessment. The NIOX MINO analyzer has a lowest detection limit of 2.3 parts/billion (9).
Ten healthy subjects (seven men and three women) were studied. All were nonsmokers and had no history of asthma or other inflammatory airway diseases. Age, body mass, and height ranged from 23 to 42 yr, 53 to 87 kg, and 1.64 to 1.91 m, respectively.
The subjects were studied while sitting in a human centrifuge with a radius of 7.25 m. The backrest of the seat was at a 28° angle to the direction of the gravitational vector. Standard monitoring included audiovisual communication between the gondola of the centrifuge and the test supervisor, assessment of brain perfusion from peripheral vision, an ECG, and a pulse oximeter coupled to an anesthesia monitoring system (type AS/3, Datex, Helsinki, Finland).
Subjects breathed through a NIOX mouthpiece breathing filter (type 02-1201, Aerocrine) that was connected to a pneumatically operated rotational valve (14) (see Fig. 1). In one position of the rotary valve, the subjects' airways were connected to the cabin air, and in the other to a non-rebreathing valve. The inspiratory port of the non-rebreathing valve was connected to a charcoal filter providing a NO-free inspirate. The exhalation port of the non-rebreathing valve was connected to a 1.5-m-long, 30-mm inner diameter plastic tube for condensation of exhaled humidity, a second low-resistance air filter (type PF 305, Pall, East Hills, NY), a heated pneumotachograph (3500 series Hans Rudolph, Kansas City, MO; range up to 580 ml/s), and finally an array of four orifices with different resistances connected in series. Venting openings between the series of four orifices were controlled by solenoid valves (type 0323 F, Bürkert, Ingelfingen, Germany) and could be closed or opened in different combinations so that the expired flow at a preset expired pressure was 50, 100, 200, or 500 ml/s. The pneumotachograph was coupled to a pressure transducer (type CD12, Validyne, Northridge, CA), mounted so that its membrane was perpendicular to the direction of rotation, to minimize the effects of the normal and centrifugal G vectors on the differential pressure recordings. One side port in the mouthpiece was connected to a pressure transducer (type DPT 6003, PVB, Kircheseeon, Germany). The mouthpiece pressure signal was amplified in the anesthesia monitor and was displayed on a LCD screen in front of the subject, together with reference lines for zero pressure and +1.5 kPa. From a second side port, there was an inlet to a 10-m-long capillary tube that forwarded sample gas to a chemiluminescence NO analyzer (type 280, Sievers, Boulder, CO) located at the center of the centrifuge. Through a third side port, a sample was sent to an infrared CO2 analyzer (type Ir 3107, Servomex, Crowborough, UK). Gas concentration readings were shifted in time corresponding to the sampling delays to become synchronized with flow and pressure readings. The pneumotachometer was calibrated before and after each experiment using a 3-liter syringe (Hans Rudolph, Kansas City, MO), mouthpiece pressure was calibrated with a water manometer, and exhaled NO and CO2 were calibrated with gases of known composition (AGA Specialgaser, Lidingö, Sweden).
Subjects performed the experiments at 1, 2, and 3 G. Thirty minutes before the first centrifuge run, subjects took a tablet to prevent motion sickness (Promethazine 25 mg). Once seated in the centrifuge, the vital capacity (VC) was determined (22). Once VC measurements were complete, the following respiratory maneuver was performed in triplicate for each combination of the four gravity conditions (1 G pre, 2 G, 3 G, and 1 G post) and for the four expired flows: exhalation to residual volume, activation of the rotary valve, inhalation of NO-free air to total lung capacity, and then controlled full exhalation while keeping airway pressure at +1.5 kPa by means of visual feedback. A typical recording from one session (the four exhalation profiles in triplicate during one gravity condition) is shown in Fig. 2.
In every case, the subjects initially exhaled half of their VC at a rate of 500 ml/s. When the time integral of the expired flow signal had reached 50% of the VC, the test leader then activated the solenoids so that the expired flow rate for +1.5 kPa airway pressure became either 50, 100, 200, or 500 ml/s for the remainder of the exhalation. This procedure allowed an initial rapid elimination of the dead-space gas in order not to prolong the maneuver so that subjects experienced “air hunger” while in hypergravity. At 2 and 3 G, the VC values were assumed to be reduced to 92 and 87%, respectively, of the 1-G values (22). During a typical session, the subject first sat for 1 min at the target G level and then repeated the above maneuver 12 times; four maneuvers with different expired flows in random order were performed with a 1-min interval in between. Thereafter, the subject rested for 3 min, followed by two more sets of four maneuvers. During the 2- and 3-G sessions, subjects rested at 1.4 G between the sets. The choice of 1.4 G rather than 1 G between the sets of four maneuvers at 2 and 3 G was made in an effort to avoid the vestibular stimulation caused by accelerating and breaking the centrifuge repeatedly. The subjects rested at 1 G for ∼30 min between repeated sessions. The order of the 2- and 3-G sessions was randomized. A typical recording from one maneuver is shown in Fig. 3. Subjects were instructed to abstain from food and beverages rich in nitrite and nitrate 24 h before the tests. Before each test session, they rinsed their mouth with water.
Data acquisition and analysis.
All physiological signals were stored continuously, together with G data from an accelerometer, on a digital data acquisition system (Biopac, Goleta, CA) at a rate of 200 Hz. Offline evaluation included the extraction of calibrated NO and CO2 concentration readings as time averages during the middle third of the expiration at 50, 100, 200, or 500 ml/s, i.e., during the middle third of the second part of the exhaled VC for flows of 50, 100, and 200 ml/s and during the middle third of the whole exhalation for flows of 500 ml/s. An individual mean value was computed for each combination of G condition and flow rate. Flow and pressure recordings were checked for deviations from target levels. For each subject and condition, plots were generated for FeNO as a function of 1/expired flow. As described by Pietropaoli et al. (17), estimates of CalvNO were derived from the 1/flow = 0 intercept (theoretical infinite flow). The rate of NO given off from the conductive airways to the expired gas (J′awNO) was assessed from the slope of FeNO vs. 1/flow. Heart rate (HR) was obtained during 20-s periods starting 30 s before each FeNO maneuver.
Microgravity and Hypergravity
Probability plots indicated that both microgravity and centrifuge data were normally distributed. Analysis of variance (Statistica 7.1, Statsoft, Tulsa, OK) with a repeated-measures design was used to test for differences between G levels at each flow. In most cases, and when the sphericity assumption held, Tukey's honestly significant difference post hoc test was performed. In other cases, paired t-tests with Bonferroni correction were used. Results were considered statistically significant at P < 0.05, and all tests were two sided. Data are given as means ± SD if not stated otherwise.
Figure 4 shows group mean data for preflight, in-flight, and postflight measurements. Data from all five subjects were obtained for the first 14 wk in space. For one subject, the analyzer malfunctioned after 14 wk, and the analyzer was subsequently exchanged when the next subject arrived at ISS. For another subject, the stay on ISS lasted only 23 wk.
Preflight FeNO was 12.3 ± 4.7 parts/billion (mean ± SD). There was no clear trend for changes of FeNO over time during the stays on ISS. Thus all in-flight data for each subject were pooled, and in-flight FeNO averaged 6.6 ± 4.4 parts/billion. This was significantly lower than the preflight value (P = 0.016). Similarly pooled data for the first week postflight was 9.7 ± 2.8 parts/billion, which did not differ significantly from either preflight (P = 0.30) or in-flight (P = 0.28). Figure 4 also shows group mean values of measurements obtained every 6 wk during spaceflight. There was no clear time trend of FeNO after landing.
All 10 subjects completed the experiments. Data on levels of NO and CO2 from the four conditions (1 G pre, 2 G, 3 G, 1 G post) and for exhalation rates of 50, 100, 200, and 500 ml/s (0.02, 0.01, 0.005, and 0.002 s/ml) are presented in Figs. 5 and 6, respectively. Exhaled NO at the flow of 50 ml/s were 16.0 ± 4.3, 19.5 ± 5.1, and 18.6 ± 4.7 parts/billion at 1, 2, and 3 G, respectively. There was a significant effect on FeNO of the main factor G (P < 0.001 for all flows). FeNO values for a given flow were higher at 2 and 3 G than at 1 G pre but did not differ between 1 G post and 1 G pre (Fig. 5). As also shown in this figure, there was a linear relationship between FeNO and the inverse of expired flow in the group mean data. Group mean values of parameters for the regression analysis of FeNO on 1/flow values are given in Table 1. CalvNO was significantly higher (∼65%) at both 2 and 3 G compared with 1 G pre. Estimated CalvNO was 2.3 ± 1.1 parts/billion in 1 G and increased significantly to 3.9 ± 1.4 and 3.8 ± 0.8 parts/billion at 2 and 3 G (P < 0.002). J′awNO tended to be elevated at 2 G compared with 1 G pre.
As with FeNO, expired partial pressure of CO2 (PeCO2) changed significantly with the main factor G (P < 0.001) but only for the three lowest flows. When this change was analyzed (Fig. 6), it was found that PeCO2 was significantly lower at 3 G compared with 1 G pre at all exhalation flows except 500 ml/s (Fig. 6C). PeCO2 at 2 G did not differ from 1 G pre. PeCO2 tended to be higher at 1 G post than 1 G pre at 50 ml/s. HR increased from 63 ± 11 beats/min at 1 G before hypergravity to 82 ± 18 and 99 ± 19 beats/min at 2 and 3 G, respectively. Post hypergravity HR was 65 ± 10 beats/min.
Our principal observation was that FeNO is gravity dependent; its values were lowered in microgravity and elevated in seated humans during hypergravity (Fig. 7). According to current models of NO transport, NO originates from conductive airways and alveoli, each source having a different impact on FeNO. Thus a change in FeNO may come from an alteration of the overall balance between production and blood recapture in the alveolar compartment as well as by a change in the airway production. Indeed, in the airways, the influence of uptake to the blood is likely negligible under the present experimental conditions (10, 20). In our hypergravity experiments, the method proposed by Pietropaoli et al. (17) with multi-flow measurements allowed discrimination between the two effects.
Alveolar NO in Hypergravity
Assuming no change in alveolar production in hypergravity, observed CalvNO changes allow estimating a loss of DlNO of 41% between 1 and 2 G and a quasi-steady state between 2 and 3 G (+2.6%). In a separate study (21), DlCO was seen to decrease by 21% between 1 and 2 G and by 16.6% between 2 and 3 G. Recently, Glenet et al. (7) proposed an interpretation of the DlNO-to-DlCO ratio. They showed that this ratio is proportional to the DmCO-to-Vc ratio, where DmCO is the membrane component of DlCO and Vc is the capillary volume. Consequently, pure recruitment or de-recruitment of capillaries will change the contact surface with blood and, thus, will affect DmCO and Vc the same way, leaving DlNO/DlCO unchanged. Any change of DlNO/DlCO will, thus, reflect a change in alveolo-capillary membrane thickness (affecting DmCO) and/or a change in the thickness component of Vc (i.e., a change in the perfusion of already recruited capillaries). Assuming no change in the alveolo-capillary membrane thickness in hypergravity, a 41% decrease in DlNO suggests a de-recruitment of capillaries from 1 to 2 G and the decrease (−26%) in DlNO-to-DlCO ratio suggests an increase in the thickness component of Vc, i.e., an over-filling in the still recruited capillaries. This is in line with the increase in tissue volume found by Rohdin et al. (21) and is compatible with an increase of zone I (not perfused) at the apex of the lung with a redirection of capillary blood flow to the dependent zones. From 2 to 3 G, an increase (+23%) in DlNO-to-DlCO ratio would indicate some perfusion decrease, in contradiction with the further increase in tissue volume found in Ref. 21. Regardless, whatever the level of perfusion, it little affects DlNO (NO is quasi perfusion-independent), and the quasi-unchanged DlNO strongly suggests no further capillary de-recruitment between 2 and 3 G. However, the above reasoning assumes constancy of alveolar NO production and alveolo-capillary thickness, which has not necessarily been the case.
Conductive Airway NO in Hypergravity
The present values of J′awNO at 1 G are in line with corresponding data in the literature (6, 11, 17). There was a trend for J′awNO to be increased at 2 G compared with 1 G (P = 0.054). Quantitatively, this increase could account for half of the increase of FeNO (at 50 ml/s) from 1 to 2 G. There are a number of factors, which theoretically can influence J′awNO in hypergravity. Tissue stretch may induce increased NO synthesis (4, 5), and during hypergravity lung tissue is certainly both stretched in some parts and compressed in others (22). Both stretch and compression may cause reductions of peripheral airway caliber. A recent theoretical study (31) showed that such reductions may influence the net NO production as measured at the mouth (and thus FeNO). Particularly, a reduction in the cross-sectional area of acinar airways may cause an increase of the observed net NO production by impairing the peripheral effect of molecular diffusion.
Effects of Stress in Hypergravity
The increased HR levels in hypergravity indicated that subjects were under physiological and probably also psychological stress in hypergravity with arterial hypotension in the head and neck (13) and with associated increases of plasma cathacolamine levels (24, 25). Adrenaline infusions have been shown to increase exhaled NO in a rabbit model (2). Therefore, had stress been a major factor for the FeNO increase, higher values should have been found at 3 G than at 2 G (25), and that was not the case. In contrast, Persoons et al. (15) showed that the NO production from alveolar macrophages was suppressed by stress in a rat model. If also true for humans, such a mechanism may have accounted for the lack of increase of exhaled NO from 2 to 3 G.
Exhaled CO2 in Hypergravity
CO2 has been shown to inhibit NO formation (26); therefore, such an interaction should also be considered in the present study. When 1 G pre and 2 G are compared, CO2 levels were practically identical, strongly suggesting that CO2/NO interaction is not a major mechanism for the increased FeNO at 2 G. The lowering of CO2 at 3 G compared with 1 and 2 G would rather result in further elevation of FeNO compared with 2 G, and since that was not the case, it is not likely that CO2/NO interactions play an important role in the present hypergravity experiments.
The CO2 values observed in the present study can be regarded as a function of different degrees of relative hyperventilation. Thus the FeNO maneuver with a relatively rapid deep inhalation and then a relatively rapid full exhalation at 500 ml/s, naturally results in an instantaneous lowering of alveolar and expired CO2 compared with normal tidal breathing. Since expired flow is decreased to 200, 100, and or 50 ml/s, during the second half of the VC expired CO2 has time to rise toward a eucapnic level (Fig. 6). This was true for all conditions except 3 G, where sustained hyperventilation was underway before the FeNO maneuvers started, leading to generally lower CO2 values during the maneuvers. The much lower CO2 values at 3 G are in agreement with a previously described G-induced hyperventilation at 3 G (23).
Lung Deformation and Ventilation/Perfusion Distribution in Hypergravity
In addition to the effect on the blood distribution in the lung, gravity and even more so hypergravity will lead to deformation of the lung tissue so that apical lung units in the seated human are stretched, basal units are compressed, and vascular hydrostatic gradients within the pulmonary circulation are increased (18). These alteration lead to regional heterogeneities of ventilation and perfusion. Recent studies (28, 31) showed that large heterogeneities of ventilation have only little effect on FeNO. At the same time, there is a gravity-induced caudal displacement of blood in the systemic circulation resulting in decreases of cardiac preload, stroke volume, and cardiac output (21, 23). Because the perfusion is more affected by gravity than ventilation, this will lead to an impaired matching between ventilation and perfusion in the lungs, manifesting itself as an increased alveolar-to-arterial oxygen partial pressure difference, arterial desaturation (23), and reduced DlCO (21). Lung expansion and emptying during a single-breath maneuver becomes increasingly sequential at 2 and 3 G, so that basal well perfused lung units are subject to airway closure toward the end of expiration (22). Thus the air at a given moment of an expiration may not be representative for the average alveolar gas composition. The impact of these phenomena for FeNO would be an increased relative contribution of nonperfused apex zones, even without further capillary de-recruitment between 2 and 3 G. Since the closing volume continuously increases from 1 to 3 G, it would tend to similarly increase the alveolar part of FeNO with gravity. Such an increase, in fact, is not observed, at least not between 2 and 3 G. One reason can be the counter-pressure applied at the mouth during the FeNO maneuver to avoid nasal NO contamination. This pressure would tend to partially prevent airway closure otherwise seen during VC exhalations in hypergravity (22). It cannot be excluded, however, that gravity-induced sequential emptying and airway closure could have contributed to elevate FeNO in hypergravity.
Exhaled NO in Microgravity
Each of the five subjects had lower FeNO values during the first measurement inflight than before the flight, and decreases ranged 21–76% from the preflight value. In this specific experimental setup, only FeNO was measured; therefore, alveolar and airway contributions may not be estimated separately. However, the decrease averaged more than 5 parts/billion, which is a numerically larger value than what is a normal value for CalvNO. Thus a reduction of CalvNO could have contributed to, but could not explain the full extent of, the FeNO reduction in microgravity. Again, assuming unchanged alveolar NO production in sustained microgravity, the previously documented improvement of the membrane component of DlCO by 28% in microgravity (19) could only have resulted in a proportional reduction of CalvNO.
With only FeNO0.05 data at hand, we can just speculate about additional explanations for the fall in FeNO. Tissue relaxation may lead to peripheral airway lumen increase improving peripheral molecular diffusion and decreasing the net NO production as measured at the mouth. In addition, decreased mechanical stress of lung tissue might have resulted in less induction of NO formation in the lung tissue (4). Also, FeNO in microgravity might represent a much better estimate of the NO concentrations in the lungs because a more representative sample for all lung units is obtained during exhalation. These potential effects of tissue relaxation as well as the DlCO increase would be “mirror effects” compared with hypergravity. Finally, one should consider the possibility of systematic differences between the analyzer units used in space and on ground. However, the analyzer units and the sensors on ISS were exchanged at the required intervals and they were compliance controlled before launch to ISS, which renders such an explanation less likely.
Conclusions and Perspectives
We have found that exhaled NO is lower in microgravity and higher in hypergravity than in normal gravity. Exhaled NO changes were in the direction that would be expected from previously determined gravity-induced alterations in lung diffusing capacity. Thus these data are consistent with the hypothesis that alveolar NO uptake is increased in microgravity and decreased in hypergravity. However, changes in lung diffusing capacity can explain neither the full extent of the lowered exhaled NO in microgravity nor the leveling off of exhaled NO from 2 to 3 G, so further studies are required to find additional mechanisms.
Funding was provided by Swedish National Space Board, European Space Agency, the Magnus Bergwall Foundation, Fraenckel's Fund for Medical Research, Swedish Heart-Lung Foundation, and Swedish Science Council (07919).
L. E. Gustafsson is an inventor of patents on exhaled NO. L. E. Gustafsson is a minority shareholder (<0.5% of total shares) in Aerocrine, which markets instruments for exhaled NO measurements for monitoring of airway inflammation. T. Hemmingsson is a former employee of Aerocrine.
The technical assistance from Thomas Andersen and Björn Johannesson is gratefully acknowledged.
- Copyright © 2009 the American Physiological Society