Air Travel Hypoxemia vs the Hypoxia Inhalation Test in Passengers With COPD*

Transcription

1 Original Research AIR TRAVEL Air Travel Hypoxemia vs the Hypoxia Inhalation Test in Passengers With COPD* Paul T. Kelly, MSc; Maureen P. Swanney, PhD; Leigh M. Seccombe, MSc; Chris Frampton, PhD; Matthew J. Peters, MD, PhD, FCCP; and Lutz Beckert, MD, FCCP Background: Limited data are available comparing air travel with the hypoxia inhalation test (HIT) in passengers with COPD. The aim of this study was to assess the predictive capability of the HIT to in-flight hypoxemia in passengers with COPD. Methods: Thirteen passengers (seven female passengers) with COPD (mean [ SD], FEV 1 /FVC ratio, 44 17%) volunteered for this study. Respiratory function tests were performed preflight. Pulse oximetry, cabin pressure, and dyspnea were recorded in flight. The HIT and a 6-min walk test were performed postflight. The in-flight oxygenation response was compared to the HIT results and respiratory function parameters. Results: All subjects flew without the use of oxygen, and no adverse events were recorded in-flight (mean cabin altitude, 2,165 m; altitude range, 1,892 to 2,365 m). Air travel caused significant desaturation (mean preflight oxygen saturation, 95 1%; mean in-flight oxygen saturation, 86 4%), which was worsened by activity (nadir pulse oximetric saturation [SpO 2 ], 78 6%). The HIT caused mean desaturation that was comparable to that of air travel (84 4%). The mean in-flight partial pressure of inspired oxygen (PIO 2 ) was higher than the HIT PIO 2 (113 3mmHg vs mm Hg, respectively; p < 0.001). The HIT SpO 2 showed the strongest correlation with in-flight SpO 2 (r 0.84; p < 0.001). Conclusion: Significant in-flight desaturation can be expected in passengers with COPD. The HIT results compared favorably with the air travel data, with differences explainable by PIO 2 and physical activity. The HIT is the best widely available laboratory test to predict in-flight hypoxemia. (CHEST 2008; 133: ) Keywords: air travel; COPD; hypobaric hypoxia; hypoxia inhalation test Abbreviations: Dlco diffusing capacity of the lung for carbon monoxide; ERS European Respiratory Society; Fio 2 fraction of inspired oxygen; HIT hypoxia inhalation test; Pio 2 partial pressure of inspired oxygen; 6MWT 6-min walk test; Spo 2 pulse oximetric saturation Air travel causes significant hypobaric hypoxia in passengers with COPD. 1,2 Predicting the hypoxic response and related risk to air travel can be complicated. The Aerospace Medical Association guidelines 3 suggest that a stable preflight Pao 2 of 70 mm Hg is generally adequate for safe travel. Other suggested assessments include the ability to walk 50 yards at a normal pace or climbing a flight of stairs without becoming severely dyspneic. 3 A more sophisticated test for predicting air travel hypoxia, which has been recommended by the Aerospace Medical Association 3 and the British Thoracic Society, 4 is the hypoxia inhalation test (HIT). The HIT involves breathing a hypoxic gas mix for 20 min with the aim of predicting hypoxemia at the maximum allowable cabin pressure altitude of 2,438 m (8,000 For related articles see pages 914 and 1002 For editorial comment see page 839 feet). 5 At 2,438 m, the barometric pressure is approximately 565 mm Hg, 6 resulting in a partial pressure of inspired oxygen (Pio 2 ) of 108 mm Hg. 7 To replicate the in-flight Pio 2 at sea-level pressure, the 920 Original Research

2 HIT uses a fraction of inspired oxygen (Fio 2 ) of 0.15 in a nitrogen balance. In-flight oxygen therapy is recommended if the HIT Pao 2 falls to 50 to 55 mm Hg. 3,4 The integrity of the HIT has been previously assessed. 8,9 Using a hypobaric chamber, Dillard et al 8 and Naughton et al 9 showed that hypobaric hypoxia at 2,438 m can be replicated by the HIT (Fio 2, 0.151) in patients with COPD. This implies that the HIT can predict air-travel hypobaric hypoxia. However, both the HIT and chamber studies are performed in a stable environment, and may not fully represent the physical stresses and environmental variability of air travel, particularly over longer flight segments. The HIT is considered to be the best available test for patient evaluation before air travel, 10 yet several authors 11,12 have questioned the HIT outcome recommendations. An 80% HIT failure rate (Pao 2, 55 mm Hg) has been reported in patients with COPD and interstitial lung disease, with 80% of subjects having flown asymptomatically in the last 5 years. 11 In a cohort of healthy passengers, we have reported 12 a normal HIT Pao 2 range of 51 to 68 mm Hg, with three subjects falling to 55 mm Hg. Participants whose Pao 2 fell to 55 mm Hg flew without the occurrence of an adverse event and had acceptable in-flight oxygen saturations. The HIT is designed to replicate hypoxemia at the lowest allowable cabin pressure (approximately 565 mm Hg). However, previous studies have reported cabin pressures significantly 565 mm Hg, indicating that the HIT may be more stressful than flight. Furthermore, a HIT and in-flight Pao 2 of 55 mm Hg may be a typical and tolerable response for passengers with chronic lung disease. There have been a limited number of studies investigating the actual response to air travel in passengers with COPD. Significant hypoxemia has been reported during unpressurized flights (1,650 *From the Respiratory Physiology Laboratory (Mr. Kelly and Drs. Swanney and Beckert), Christchurch Hospital, Christchurch, New Zealand; the Department of Thoracic Medicine (Ms. Seccombe and Dr. Peters), Concord Hospital, NSW, Australia; and the University of Otago (Dr. Frampton), Christchurch School of Medicine, Christchurch, New Zealand. This research was supported by a grant from the Christchurch School of Medicine, University of Otago. This research was conducted independent of any organizations that may in any way gain or lose from the publication of these results. All of the authors have stated they have no conflicts of interest and are agreeable to have these results published. Manuscript received June 20, 2007; revision accepted September 27, Reproduction of this article is prohibited without written permission from the American College of Chest Physicians ( org/misc/reprints.shtml). Correspondence to: Paul T. Kelly, MSc, Christchurch Hospital, Pulmonary Function Laboratory, Riccarton Ave, Christchurch 8001, New Zealand; paul.kelly@cdhb.govt.nz DOI: /chest and 2,250 m) 1 and pressurized flights (1,829 m pressure equivalent) 2 in patients with COPD. Data comparing the HIT results using a recommended Fio 2 of 0.15 and air travel in passengers with COPD are lacking. Furthermore, the current literature does not include continuous oximetry data on flights with various aircraft models and destinations. The aim of this study was to assess the capability of the HIT to predict in-flight hypoxemia in passengers with COPD. Materials and Methods Thirteen subjects (seven female subjects) with documented COPD volunteered for this study. Subjects who were embarking on air travel on their own accord were recruited from the respiratory clinic at Christchurch Hospital. The Upper South B Regional Ethics Committee approved the study, and written informed consent was obtained for all subjects. The study included the following three phases: (1) preflight respiratory function testing; (2) in-flight physiologic measures; and (3) postflight HIT and 6-min walk test (6MWT). Phases 1 and 3 were undertaken at the Christchurch Hospital Respiratory Laboratory (40 m above sea level). All participants were clinically stable throughout the three study phases. The HIT was performed postflight due to the ethical implications of a possible failed test that was not otherwise medically requested. Preflight Tests Respiratory function tests were performed within 1 month of travel. Spirometry, lung volumes, and diffusing capacity of the lung for carbon monoxide (Dlco) were assessed using a body plethsymograph (V max 6200 Autobox; Viasys; Yorba Linda, CA). Reference equations by Hankinson et al 15 and the European Respiratory Society (ERS) 16 were used for the predicted lung function variables. Procedures adhered to the 2005 American Thoracic Society/ERS pulmonary function guidelines In-Flight Measurements In-flight pulse oximetric saturation (Spo 2 ) and pulse rate were measured continuously using a portable pulse oximeter (3100 WristOx; Nonin Medical; Minneapolis, MN). Parameters were recorded 5 min before the flight and 5 min after the flight as the preflight/postflight parameters, respectively. The mean Spo 2 during the cruise phase of the flight was taken as the in-flight Spo 2.Spo 2 nadir was considered to be the lowest Spo 2 value during the cruise phase of the flight. Subjects were advised to carry out in-flight activities normally and to ignore oximeter readings. Subjects recorded in-flight activities and symptoms of breathlessness using the Borg dyspnea scale. 20 Cabin pressure was measured with a wrist altimeter (RA109; Oregon Scientific; Neu-Isenburg, Germany) that was calibrated to the departure port barometric pressure. Altimeter recordings began before entering the aircraft cabin and were automatically stored every 60 seconds. The altimeters and pulse oximeters had been checked previously for accuracy. 12 Continuous in-flight measures (Spo 2, pulse rate, and cabin altitude) were averaged for each phase of the flight. Postflight Tests The HIT and 6MWT were performed approximately 2 weeks after the flight. Pre-HIT, arterial blood was withdrawn from the CHEST / 133 / 4/ APRIL,

3 Table 1 Demographics and Respiratory Function Parameters* Variables Mean SD Range % Predicted Age, yr Height, m Mass, kg FEV 1,L FVC, L FEV 1 /FVC ratio, % TLC, L RV, L Dlco, ml/mm Hg/min *Values are given as the mean SD. TLC total lung capacity; RV residual volume. radial artery with a 25-gauge needle attached to a heparanized syringe and immediately analyzed with a blood gas analyzer (ABL 700; Radiometer; Copenhagen, Denmark). The HIT was performed using a technique described by Gong et al. 21 Subjects inhaled a hypoxic gas mixture of 15% oxygen in a nitrogen balance (BOC Gases; Auckland, New Zealand) via a reservoir bag and a two-way nonrebreathing valve (Hans Rudolph Inc; Kansas City, MO) for 20 min. Subjects breathed through the HIT circuit via a mouthpiece, with a peg occluding the nose. Spo 2 and pulse rate were measured continuously for 20 min or until the Spo 2 was 80%, at which time a second arterial blood sample was taken. After 20 min of recovery, subjects completed a 6MWT. 22 Subjects were instructed to cover as much distance over the 6 min as possible. Statistical Analysis Results are expressed as the mean SD, unless otherwise stated. Paired t tests were used to evaluate the differences between the HIT and in-flight parameters; p values of 0.05 were considered to be significant. A Pearson correlation was used to assess the relationships among baseline respiratory function, HIT, and the in-flight response. Results Table 1 shows the group demographics and respiratory function parameters. The FEV 1 /FVC ratio and the FEV 1 were below the lower limit of normal for all participants. The cabin environment is summarized in Table 2. In-flight data were recorded on six commercial carriers and four types of aircraft. Flight times ranged from 1 to 11 h, and included both domestic and international sectors. There was variability in the cruise altitude cabin pressure (Table 2). The mean time from takeoff to cruise altitude was 20 2 min, and the mean time from cruise altitude to landing was 20 4 min. In-Flight Measurements Results from the in-flight measures and the HIT are presented in Table 3. The mean preflight pulse rate did not differ from the mean in-flight pulse rate. The mean peak in-flight pulse rate ( beats/ min) was significantly higher (p 0.001) than the mean preflight pulse rate (84 14 beats/min). The mean preflight Spo 2 was 95 1%, and it decreased with the reduction in cabin pressure until cruise altitude was achieved. Figure 1 shows an example of the Spo 2 /cabin pressure profile. The mean in-flight Spo 2 was 86 4% (p vs preflight Spo 2 ). The in-flight Spo 2 was reduced further during activity, resulting in a mean nadir Spo 2 of 78 6% (p vs in-flight Spo 2 ). The nadir Spo 2 occurred over a wide time range (24 to 198 min after takeoff). Table 4 outlines the individual responses to the HIT and flight. The in-flight Spo 2 nadir coincided with walking down the aisle (n 4), visiting the lavatory (n 5), and being seated (n 4). The postflight Spo 2 returned to preflight values. Figure 2 describes the grouped in-flight Spo 2 overlaid with the HIT data. In-Flight Clinical Observations The Borg dyspnea scale score increased from 1 ( very slight preflight score) to 2 ( slight in-flight score). The highest dyspnea score in flight was 4 ( somewhat severe ). Five subjects reported no signs of breathlessness. Flight dyspnea scores were repli- Table 2 Flight Summary* Aircraft Model Flight Time, min Cruise Altitude Time, min Cruise Cabin Pressure, mm Hg Cruise Cabin Altitude, m A320 (n 5) , B737 (n 4) , B767 (n 2) ,229 B777 (n 2) ,892 All flights , Range ,892 2,365 *Values are given as the mean SD, unless otherwise indicated. A320 Airbus A320; B737 Boeing 737 (both 300 and 800 series); B767 Boeing ; B777 Boeing 777. Cabin pressure data were not available for one B767 flight and one B777 flight. Population was n 11 for these summary data. 922 Original Research

4 Table 3 Responses to the HIT and Air Travel* Laboratory Tests Flight Variables Resting Room Air HIT End Test Preflight In-Flight Cruise In-Flight Nadir Pb, mm Hg Fio Approximately 0.21 Approximately 0.21 Pio 2,mmHg ph Paco 2,mmHg Pao 2,mmHg Spo 2,% Pulse rate, beats/min *Values are given as the mean SD. Pb barometric pressure.; PIO (Pb-47). 7 PIO 2 not corrected for relative humidity. Two data points are missing; n 11 for this parameter. Fio 2 is assumed to be 0.21 in flight. p (resting room air laboratory vs HIT laboratory ). p (pre-flight vs in-flight cruise altitude). p (in-flight nadir vs in-flight cruise altitude). cated in the HIT (1, pre-hit; 2, at end-hit). No adverse events were reported in flight. Five participants reported dyspnea while disembarking (two participants required wheelchairs). Postflight Tests Significant desaturation (p 0.001) occurred during the HIT compared to the baseline measures (Table 3). There was a significant decrease in Spo 2 from 0 to 5 min (p 0.001), 5 to 10 min (p 0.01), and 10 to 15 min (p 0.01). There was no significant change in Spo 2 during the last epoch of the HIT (15 min to HIT end). Two participants had blood sampled at 19 min after Spo 2 fell to 80%. There was no change in Paco 2 or pulse rate during the HIT (Table 3). Figure 3 illustrates the desaturation characteristics of the HIT. The mean time to reach the Spo 2 nadir was 17 4 min. Figure 1. Actual cabin altitude and in-flight Spo 2 on a study subject. HIT Spo 2 range (15 to 20 min) presented on the y 2 -axis. Laboratory Testing and In-Flight Relationships The relationships between laboratory tests and in-flight hypoxemia are shown in Table 5. There was a strong relationship between the post-hit Spo 2 and the mean in-flight Spo 2 (r 0.84). However, the difference in these parameters was significant (p 0.05). Both FEV 1 and Dlco yielded significant relationships with in-flight Spo 2. Resting Pao 2 and 6MWT parameters did not correlate with in-flight Spo 2. There was no relationship between in-flight Borg scores and in-flight Spo 2. Similarly, in-flight Borg scores did not correlate with HIT Spo 2. Discussion In the present study, the in-flight oxygen response was compared to the HIT in 13 passengers with COPD. Air travel caused significant desaturation, which was worsened by activity. The HIT Spo 2 was comparable with the mean flight Spo 2. Despite significant in-flight oxygen desaturation, there were no adverse events reported. The HIT Spo 2 showed the strongest correlation with the mean in-flight Spo 2. The HIT is a practical laboratory assessment for predicting air-travel hypoxemia in patients with pulmonary disease. Yet, it has not previously been compared to actual pressurized air travel. Our primary objective was to evaluate the relationship between the HIT and air travel in passengers with COPD. The HIT Spo 2 was marginally lower than mean in-flight Spo 2. A likely cause of the Spo 2 disparity was the difference in Pio 2 between the HIT and the flight. The HIT Pio 2 was calculated to be 107 mm Hg, which should approximate the Pio 2 CHEST / 133 / 4/ APRIL,

5 Table 4 Individual Responses to the HIT and Air Travel* Laboratory Tests Air Travel (Cruise Phase) Gender FEV 1,% predicted Dlco, % predicted Pre-HIT PaO 2, mm Hg HIT PaO 2, mm Hg HIT SpO 2,% Mean Spo 2,% Nadir Spo 2,% Activity at Nadir Spo 2 Disembarking Aircraft F Walking Dyspnea M Walking F Sitting Wheelchair M Sitting Wheelchair F Walking F Toilet M Toilet Dyspnea M Toilet M Walking Dyspnea F Toilet F Toilet F Sitting M Sitting Mean SD *Participants are listed in ascending order of HIT Pao 2.M male; F female. (approximately 108 mm Hg) during air travel at the lowest allowable cabin pressure of 565 mm Hg. We calculated the average in-flight Pio 2 at 113 mm Hg, more favorable than the HIT, resulting in higher oxygen saturations. Cabin pressures of 565 mm Hg (equivalent of 2,438 m) have been regularly reported, 2,12 14 indicating that the hypobaric stress of flight may be less than that predicted by the HIT. This was the most likely cause for the disparity of the HIT Spo 2 vs the mean in-flight Spo 2. One weakness of the HIT is the lack of an activity stimulus during the procedure. Air travel involves secondary stresses to the hypobaria (eg, moving around the cabin or visits to the lavatory). Seccombe et al 11 showed that the addition of a 50-m walk task during the HIT significantly worsened hypoxemia in patients with respiratory disease. This has also been demonstrated in studies using hypobaric chambers 9,23,24 and in flight in passengers with COPD. 2 We found that the preflight Spo 2 of 95% decreased to 86% in flight, and further decreased to 78% during in-flight activity, which is significantly lower than the HIT Spo 2. For a more accurate estimation of in-flight oxygenation, the HIT should include a light exercise component. Aeromedical guidelines 3 advocate a preflight Pao 2 of 70 mm Hg for safe travel. Numerous authors have questioned the utility of the 70 mm Hg threshold and suggest this does not always ensure an adequate in-flight Pao 2. Oxygen supplementation therapy is advised if the Pao 2 is expected to fall to below the threshold of 50 to 55 mm Hg. 3,4 In the present study, the pre-hit Pao 2 for 11 of 13 subjects was 70 mm Hg (Fig 4); therefore, it was considered safe for those subjects to fly without oxygen therapy. After HIT, Pao 2 had decreased to Figure 2. Grouped in-flight Spo 2 data vs HIT data. Figure 3. Mean ( SD) grouped HIT Spo 2 desaturation curve. 924 Original Research

6 Table 5 Relationship Between In-Flight Hypoxemia and Laboratory Tests* Variables Resting FEV 1 % Pao 2 HIT Pao 2 HIT Spo 2 Predicted Dlco % Predicted 6MWT Distance Mean flight Spo 2 r 0.36/NS r 0.77/p 0.01 r 0.84/p r 0.65/p 0.05 r 0.79/p 0.01 r 0.35/NS Nadir flight Spo 2 r 0.28/NS r 0.61/p 0.05 r 0.62/p 0.05 r 0.49/NS r 0.53/NS r 0.29/NS *NS not significant. 55 mm Hg in 11 subjects, indicating that those passengers should have flown using supplementary oxygen, which contradicts the recommendations of the current guidelines. This is consistent with previous studies, 2,11,23 confirming that a baseline Pao 2 of 70 mm Hg does not ensure a flight Pao 2 of 55 mm Hg. Despite the ambiguity of these oxygen thresholds, all participants in the present study flew without using supplementary oxygen and without experiencing an adverse event. This highlights the inconsistencies in both the preflight Pao 2 threshold and the HIT Pao 2 threshold pertaining to the prescription of in-flight supplementary oxygen. Larger prospective studies are required to provide clarity regarding absolute oxygen thresholds and the requirement for in-flight oxygen. In the present study, we observed severe hypoxemia during the HIT and air travel in passengers with COPD. Other studies using the HIT, 1,8,9,11 hypobaric chambers, 8,9,23 unpressurized air travel, 1 and pressurized air travel 2 in patients with COPD have reported similar levels of hypoxemia. No previous studies have documented the occurrence of any serious events due to acute hypoxic exposure. Transient hypoxemia occurs regularly in this cohort, and the hypobaric hypoxia experienced in flight may be tolerated without adverse event when their condition is stable. The extent to which overall risk is Figure 4. The HIT and in-flight oxygenation response, in ascending order from lowest to highest end-hit Pao 2. Obstructive severity was based on the American Thoracic Society/ERS 2005 classification guidelines. 26 v.severe very severe obstruction; m.severe moderately severe obstruction. related to the severity or duration of hypoxemia is currently unknown. 25 Our results confirm that passengers with COPD will experience significant desaturation, which is considered unacceptable by current guidelines; however, they may be able to tolerate this level of hypoxemia with minimal health risk. Postflight stress in passengers with COPD has not been reported in the literature. Five subjects in the present study reported significant dyspnea disembarking the aircraft, with two subjects requiring wheelchair assistance. Data for respiratory parameters (ie, FEV 1 percent predicted and Dlco percent predicted), HIT Pao 2, and flight Spo 2 on average were worse in these subjects when compared to the grouped data. Although the mechanisms of this posthypoxic stress are unknown, passengers with severe mechanical and lung diffusion impairments are more likely to require assistance to disembark the aircraft. Investigators have had variable success in predicting the response to hypobaria from laboratory measures in individuals with lung disease. Reported correlates include Dlco, 2,24 aerobic capacity, 2 and resting Pao 2. 2,8,21,24 In the present study, despite some cabin pressure and flight time variability, we found a strong relationship between the mean inflight Spo 2 and the HIT Pao 2 and Spo 2, confirming the usefulness of the HIT for predicting in-flight oxygenation. Percent predicted Dlco showed a strong relationship with mean in-flight Spo 2, highlighting the role of gas transfer and hypoxia in patients with COPD. We found no relationship between the 6MWT parameters and in-flight Spo 2, suggesting that the notion using walk tests to assess fitness to fly is not valid. In summary, we recorded the oxygenation response to air travel in a group of 13 passengers with COPD. Air travel caused significant desaturation, which was worsened by in-flight physical activity. Despite significant in-flight hypoxia, there were no adverse events reported during flight. The HIT Spo 2 had a strong relationship with the mean in-flight Spo 2, confirming its predictive capability with hypobaric hypoxia. Although many of our participants would have qualified for the use of in-flight oxygen therapy, none of them felt as though the use of in-flight oxygen was necessary. In our opinion, if a CHEST / 133 / 4/ APRIL,

7 HIT Pao 2 is 55 mm Hg, it should not automatically indicate the need for in-flight oxygen therapy. Rather, it should prompt an informed discussion with a clinician about the benefits and risks of flying with or without supplemental oxygen. Further research is required to determine the degree and duration of hypobaric hypoxia in relation to its relative risk of adverse event. This would help to establish appropriate evidence-based air-travel guidelines for passengers with respiratory disease. In our opinion, the HIT is the best widely available laboratory test for predicting the occurrence of in-flight hypoxia. ACKNOWLEDGMENT: The authors thank the Christchurch School of Medicine, University of Otago, and the enthusiastic subjects who participated in the study. References 1 Schwartz JS, Bencowitz HZ, Moser KM. Air travel hypoxemia with chronic obstructive pulmonary disease. Ann Intern Med 1984; 100: Akerø A, Christensen CC, Edvardsen A, et al. Hypoxaemia in chronic obstructive pulmonary disease patients during a commercial flight. Eur Respir J 2005; 25: Aerospace Medical Association Medical Guidelines Task Force. Medical guidelines for air travel. 2nd ed. Aviat Space Environ Med 2003; 7:A1 A19 4 British Thoracic Society Standards of Care Committee. Managing passengers with respiratory disease planning air travel: British Thoracic Society recommendations. Thorax 2002; 57: Code of Federal Regulations. Title 14, part Washington, DC: US Government Printing Office, West JB. Prediction of barometric pressures at high altitudes with the use of model atmospheres. J Appl Physiol 1996; 81: West JB, Shoene RB, Milledge JS. High altitude medicine and physiology. 4th ed. London, UK: Hodder Arnold, 2007; 23 8 Dillard TA, Moores LK, Bilello KL, et al. The preflight evaluation: a comparison of the hypoxia inhalation test with hypobaric exposure. Chest 1995; 107: Naughton MT, Rochford PD, Pretto JJ, et al. Is normobaric simulation of hypobaric hypoxia accurate in chronic airflow limitation? Am J Respir Crit Care Med 1995; 152: Dillard TA, Bansal AK. Commentary: pulse oximetry during air travel. Aviat Space Environ Med 2007; 78: Seccombe LM, Kelly PT, Wong CW, et al. Effect of simulated commercial flight on oxygenation in patients with interstitial lung disease and chronic obstructive pulmonary disease. Thorax 2004; 59: Kelly PT, Swanney MP, Frampton C, et al. Normobaric hypoxia inhalation test vs. response to airline flight in healthy passengers. Aviat Space Environ Med 2006; 77: Kelly PT, Seccombe LM, Rogers PG, et al. Directly measured cabin pressure conditions during Boeing commercial aircraft flights. Respirology 2007; 12: Cottrell JJ. Altitude exposures during aircraft flight: flying higher. Chest 1988; 93: Hankinson JL, Odencrantz JR, Fedan KB. Spirometric reference values from a sample of the general US population. Am J Respir Crit Care Med 1999; 159: Cotes JE, Chinn DJ, Quanjer PH, et al. Standardization of the measurement of transfer factor (diffusing capacity): Report Working Party Standardization of Lung Function Tests, European Community for Steel and Coal; Official Statement of the European Respiratory Society. Eur Respir J 1993; 6(suppl): Miller MR, Hankinson J, Brusasco V, et al. Standardisation of spirometry. Eur Respir J 2005; 26: Wanger J, Clausen JL, Coates A, et al. Standardisation of the measurement of lung volumes. Eur Respir J 2005; 26: MacIntyre N, Crapo RO, Viegi G, et al. Standardisation of the single-breath determination of carbon monoxide uptake in the lung. Eur Respir J 2005; 26: Borg GA. Psychophysical bases of perceived exertion. Med Sci Sports Exerc 1982; 14: Gong H Jr, Tashkin DP, Lee EY, et al. Hypoxia-altitude simulation test: evaluation of patient with chronic airway obstruction. Am Rev Respir Dis 1984; 130: American Thoracic Society. Guidelines for the six-minute walk test: American Thoracic Society statement. Am J Respir Crit Care Med 2002; 166: Christensen CC, Ryg M, Refvem OK, et al. Development of severe hypoxaemia in chronic obstructive pulmonary disease patients at 2,438 m (8,000 ft) altitude. Eur Respir J 2000; 15: Christensen CC, Ryg MS, Refvem OK, et al. Effect of hypobaric hypoxia on blood gases in patients with restrictive lung disease. Eur Respir J 2002; 20: Seccombe LM, Peters MJ. Oxygen supplementation for chronic obstructive pulmonary disease patients during air travel. Curr Opin Pulm Med 2006; 12: Pellegrino R, Viegi G, Brusasco V, et al. Interpretive strategies for lung function tests. Eur Respir J 2005; 26: Original Research