Positive end-expiratory pressure (PEEP) is commonly used in critical care medicine to improve gas exchange. Altitude sickness is associated with exaggerated reduction in arterial oxygenation. We assessed the effect of PEEP and pursed lips breathing (PLB) on arterial and tissue oxygenation under normobaric and hypobaric hypoxic conditions.
Sixteen healthy volunteers were exposed to acute normobaric hypoxia (Laboratory study, FiO2=0.12). The protocol consisted in 3-min phases with PEEPs of 0, 5 or 10 cmH2O, PLB or similar ventilation than with PEEP-10, interspaced with 3-min phases of free breathing. Arterial (pulse oximetry) and quadriceps (near-infrared spectroscopy) oxygenation, ventilation, cardiac function, esophageal and gastric pressures and subjects’ subjective perceptions were recorded continuously. In addition, the effect of PEEP on arterial oxygenation was tested at 4,350 m of altitude in 9 volunteers breathing for 20 min with PEEP-10 (Field study).
During the laboratory study, PEEP-10 increased arterial and quadriceps oxygenation (arterial oxygen saturation +5.6±5.0% and quadriceps oxyhemoglobin +58±73 µmol.cm compared to free breathing; p<0.05). Conversely, PLB did not increase oxygenation. Oxygenation improvement with PEEP-10 was accompanied by an increase in expiratory esophageal and gastric pressures (esophageal pressure swing +5.4±3.2 cmH2O, p<0.05) but no change in minute ventilation, breathing pattern, end-tidal CO2 or cardiac function (all p>0.05) compared to PEEP-0. During the field study, PEEP-10 increased arterial oxygen saturation by +6.7±6.0% after the 3rd minute with PEEP-10 without further significant increase until the 20th minute with PEEP-10. Subjects did not report any significant discomfort with PEEP.
These data indicate that 10-cmH2O PEEP significantly improves arterial and muscle oxygenation under both normobaric and hypobaric hypoxic conditions in healthy subjects. PEEP-10 could be an attractive non-pharmacological tool to limit blood oxygen desaturation and possibly symptoms at altitude.
Citation: Nespoulet H, Rupp T, Bachasson D, Tamisier R, Wuyam B, Lévy P, et al. (2013) Positive Expiratory Pressure Improves Oxygenation in Healthy Subjects Exposed to Hypoxia. PLoS ONE 8(12): e85219. https://doi.org/10.1371/journal.pone.0085219
Editor: James West, Vanderbilt University Medical Center, United States of America
Received: September 9, 2013; Accepted: November 25, 2013; Published: December 23, 2013
Copyright: © 2013 Nespoulet et al. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Funding: The present study was funded by the scientific council of AGIRàdom (http://www.agiradom.com/spip.php?article35). The funder had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.
Competing interests: The authors have declared that no competing interests exist.
Pursed-lips breathing (PLB) is commonly used in order to reduce dyspnea and to improve gas exchange in patients with chronic obstructive pulmonary disease [1,2]. Positive end-expiratory pressure (PEEP) is also used in critical care medicine to improve pulmonary gas exchange and compliance [3,4]. At least part of the effect of PEEP and PLB on gas exchange is thought to result from an increase in alveolar pressure, leading to improved lung diffusion capacity and blood oxygenation [5,6]. At high-altitude or in long-time flight, the decrease in barometric pressure induces a significant hypoxemia in mountaineers and travelers. As a consequence of hypoxemia, acute mountain sickness (AMS) and pulmonary or cerebral edema may occur and put one’s life at risk . It is well known that the severity of these pathologies is closely correlated with the degree of arterial oxygen desaturation (SpO2). Burtscher et al.  showed that, for a given altitude (>2,500 m) or equivalent normobaric hypoxic level, a difference of about 4.9% of SpO2 is a key factor that distinguishes people who develop symptoms of altitude intolerance and those remaining clinically healthy. Peripheral O2 chemosensitivity appears to be a critical factor involved in this SpO2 difference [9,10]. On field, acetazolamide and theophylline are commonly used in order to prevent or treat AMS, because of their ability to increase hypoxic ventilatory response and improve SpO2. In addition to increased ventilation, improved pulmonary diffusion is another mechanism associated with altitude acclimatization and enhanced blood oxygenation . Therefore, any intervention that might improve pulmonary diffusion could potentially increase blood oxygenation and altitude tolerance.
Using positive expiratory pressure as a method to improve arterial oxygenation at altitude has been suggested by some authors [11–15], but its efficiency and the potential mechanisms involved remain to be clarified. In particular, no study has been able to verify whether the improvement in oxygen saturation while increasing expiratory pressure is associated with changes in minute ventilation or breathing pattern . Increased tidal volume  and minute ventilation (VE)  as previously reported with PEEP may induce per se some improvement in blood oxygenation . Various levels of PEEP have been used (5 cmH2O [13,14], 7 cmH2O , 10 cmH2O ) but no comparison has been done between different PEEP levels to determine the optimal pressure. The individual tolerance to the device also remains to be investigated. In addition, PEEP is known to significantly modify cardiac function  but heart rate (HR) only has been measured during PEEP in hypoxia and inconsistent changes have been reported [11,13,15]. Hence, it remains to assess whether PEEP during hypoxic exposure induces significant changes in left ventricular stroke volume (SV) and cardiac output (Qc).
PLB, which does not require any equipment, can also increase tidal volume, minute ventilation and oxygen saturation in healthy subjects or in patients [2,18,19]. A recent field case study assessed the effect of PLB in a subject presenting severe AMS at 4,330 m . A rapid and critical increase in SpO2 has been observed but no other parameter was measured. Hence, whether PLB may lead to similar effects than PEEP on blood oxygenation during hypoxic exposure remains to be investigated.
To clarify the effects of PEEP and PLB in healthy subjects during hypoxic exposure, we tested the following hypotheses: i) PEEP improves SpO2 and muscle oxygenation in hypoxia irrespective of any change in ventilation; ii) PEEP level (0 versus 5 versus 10 cmH2O) has a significant impact on oxygenation improvement; iii) PLB is as efficient as PEEP in improving oxygenation under hypoxic conditions; iv) PEEP improves SpO2 under both normobaric and hypobaric hypoxic conditions. In addition, we aimed to evaluate the effect of PEEP and PLB on pleural pressure, cardiac function and subjects’ subjective perception.
Materials and Methods
Sixteen healthy male volunteers (age: 38.1 ± 11.5 yrs, BMI: 22.1 ± 1.8 Kg·m-2) participated in the laboratory study and 9 additional healthy male volunteers (age: 31.2 ± 11.9 yrs, BMI: 23.4 ± 2.0 Kg·m-2) participated in the field study. All participants gave their informed written consent, and the study was approved by the local ethics committee (CPP Sud-Est V, Grenoble). All subjects were natives of low altitude and none stayed at altitude above 3,000 m during the two months before the study. All were free of cardiovascular, pulmonary or neurological diseases. They all had normal lung function (forced vital capacity: 110 ± 11% predicted; forced expiratory volume in 1 s: 109 ± 12% predicted) and normal lung diffusion capacity (alveolar capillary membrane conductance: 97 ± 16% predicted; pulmonary capillary blood volvolume: 83 ± 15% predicted; alveolar volvolume: 109 ± 24% predicted) . The laboratory study was conducted in the Grenoble University Hospital (210 m of altitude; barometric pressure, Pb = 740 mmHg) while the field study was conducted at the Observatoire Vallot near the Mont-Blanc, France (4,350 m; Pb = 448 mmHg).
During the laboratory study, subjects were seated comfortably in a semi-supine position and were connected with a mouthpiece and a three-way valve to an ergospirometric device measuring gas exchange and breathing pattern (SensorMedics, Yorba-Linda, Ca, USA). The inspiratory side of the valve was connected to an Altitrainer® (SMTEC, Nyon, Switzerland) delivering a gas mixture with an inspiratory oxygen fraction (FiO2) of 0.12 (normobaric hypoxia, equivalent to 4,300 m of altitude; inspiratory oxygen pressure, PiO2 = 83 mmHg). The expiratory side of the valve was adapted to fit with a mechanical resistance (Ambu PEEP®, 0-10 cmH2O, Ballerup, Danmark). SpO2 was continuously measured at the finger with a Nonin 4100 oxymeter (Nonin Medical, Plymouth, MN). After 20 min of quiet breathing in hypoxic conditions in order to stabilize arterial desaturation, subjects had to breathe for successive phases of 3 min with specific conditions of ventilation, 3 min of free breathing (FB, no PEEP device) separating each condition. Five conditions were tested: a PEEP of 0 cmH2O (PEEP-0), 5 cmH2O (PEEP-5), 10 cmH2O (PEEP-10), breathing at the ventilatory level spontaneously reached with PEEP-10 (the target ventilation was continuously coached by the experimenter to the subject) but with no PEEP device (T10) and PLB. The PEEP-0 condition was used as a sham condition since subjects were blinded for the PEEP level. For the PLB condition, subjects were trained to perform this technique before the experimental session to avoid any learning effect during the protocol. They were asked to breathe using a short inhalation followed by a long exhalation against a resistance induced by pursed lips that would induce no major breathing discomfort. The image of “breathing through a straw” was used. Within 30 min, all five conditions (PEEP-0, PEEP-5, PEEP-10, T10 and PLB) were performed in a random order (except that T10 was always performed after PEEP-10). This 30-min sequence was repeated twice, the whole experimental session lasting for 80 min. Ventilation, esophageal and gastric pressures, cardiac function, quadriceps oxygenation and subjects’ subjective perceptions were continuously measured during the laboratory study (see below).
The field study was performed following the laboratory study to assess the effect of PEEP during hypobaric hypoxic exposure (after 2.9 ± 1.5 days at 4,350 m following helicopter ascent, i.e. after the initial 2 days of acclimatization; PiO2 = 84 mmHg) on SpO2. Subjects were seated comfortably and breathed with a mouthpiece through a three-way valve that was connected on the expiratory side to the same device as during the laboratory study inducing a PEEP of 10 cmH2O. After 3 minutes of FB, subjects breathed for 20 min with PEEP-10 while SpO2 was continuously measured at the finger (Nonin Medical).
Esophageal and gastric pressures.
In order to estimate thoracic (pleural) and abdominal pressures and the work of breathing, esophageal (Pes) and gastric (Pga) pressures were measured by conventional balloon catheters (Milic-Emili et al. 1964), connected to pressure transducers (model DP45-30, Validyne, Northridge, CA). The pressure analog signal was digitized (MacLab, ADInstruments, Castle Hill, Australia) and recorded on a computer (Chart Software version 5.0, ADInstruments).
SV, HR and Qc were measured using a non-invasive impedance cardiography device (Physioflow®, Manatec Biomedical, Paris, France). Physioflow is an impedance technique based on the principle that variations in the impedance to a high-frequency (75 kHz) low-magnitude (1.8 mA) alternating current across the thorax during cardiac ejection result in a waveform from which SV can be calculated. Initially, SV index is calculated at rest by evaluating 24 consecutive heart beats (autocalibration procedure), using measurements of the largest impedance difference during systole, as well as the largest rate of variation of the impedance signal (contractility index), the thoracic fluid inversion time, HR, and pulse pressure (i.e. the difference between systolic and diastolic arterial pressure) . Cardiac output is then calculated by multiplying the SV index with body surface area and HR .
After cleaning the skin, two pairs of electrodes (FS50, Skintact, Innsbruck, Austria) were positioned at the left base of the neck and along the xiphoid for transmitting and receiving electrical currents. Two electrodes were also placed on the chest (V1/V6 position) to obtain the ECG signal. The autocalibration procedure was started after a period of at least 5 min, in which patients were immobile. SV, HR and Qc values were stored beat-to-beat.
Pulmonary function tests.
Lung function and lung diffusion capacity for carbon monoxide and nitric oxide (TLCO/NO) measurements were performed in a BodyBox 5500 (Medi-Soft, Dinant, Belgium). For TLCO and TLNO, procedures and normal values were those reported by Aguilaniu et al. , from which transfer coefficient, membrane conductance, capillary volume and alveolar volume were derived.
Skeletal muscle oxygenation.
Muscle oxygenation was measured continuously by near-infrared spectroscopy (NIRS)  with a four-wavelength (775, 810, 850, 905 nm) high temporal resolution NIRS device (NIRO-300, Hamamatsu Photonics, Hamamatsu City, Japan). NIRS probes were attached to the skin on the lower third of the belly of the right vastus lateralis (range of 15–20 cm above the proximal border of the patella) and in parallel with the long axis of the muscle. The distance between the transmitting and receiving optodes was fixed at 4 cm by a probe holder secured to the skin using double-sided tape and covered with a black sweatband maintained with an elastic muff net to shield the optodes from ambient light.
Data were collected with a sampling frequency of 2 Hz. For each phase, relative concentration changes (∆μmol·cm) of oxy-(∆[O2Hb]), deoxy-(∆[HHb]) and total (∆[THb] = [O2Hb] + [HHb]) hemoglobin were measured from the previous FB condition. We used a multidistance spatially resolved tissue oximeter (NIRO-300) that is able to quantify tissue oxy-hemoglobin saturation directly as a tissue oxygenation index (TOI) reflecting the dynamic balance between O2 supply and O2 consumption in the investigated muscle volume .
Subjects indicated their subjective perception of well-being and breathing effort on two visual analog scales (VAS) of 100-mm long. At the beginning and at the end, there were labeled with ‘perfect well-being’ on the left end and ‘unbearable condition’ on the right end or ‘no breathing effort’ on the left end and ‘extreme breathing effort’ on the right end. Subjects were carefully instructed and asked to score their subjective perception of well-being from ‘as before starting the study’ (i.e. ‘perfect well-being’) to ‘it cannot be even worse” (i.e. ‘unbearable condition’) and ‘how difficult is it to breathe from ‘no difficulty’ (i.e. ‘no breathing discomfort’) to ‘extremely difficult’ (i.e. ‘extreme breathing discomfort’). The subject’s marks on the 100-mm VAS were converted in scores between 0 and 10. Both VAS were filled during the last 30 s of each condition during the whole experiment.
For the laboratory study, all measurements represent the mean value of the last 30 s of each condition. Because no significant difference was observed for any breathing condition between the first and the second 30-min period, data presented for each breathing condition (PEEP-0, PEEP-5, PEEP-10, T10 and PLB) are the averaged value from the two repetitions of each 3-min phases. Because no significant difference was observed between all FB conditions, FB values are the average of all FB conditions. The comparison of parameters between the different breathing conditions (FB, PEEP-0, PEEP-5, PEEP-10, T10 and PLB) was achieved using one-way analysis of variance (ANOVA) with repeated measurements. When significant main effects were found, Fischer’s p-tests were used for post hoc analysis.
For the field study, changes in SpO2 over time (averaged for 3 min of FB and for each minute with PEEP-10) were assessed with a one-way analysis of variance (ANOVA) with repeated measurements and Fischer’s p-tests for post hoc analysis.
All statistical calculations were performed on standard statistics software (Statview 5.0, SAS Institute, Cary, North Carolina). Significance was set at p < 0.05. All descriptive statistics presented as mean values ± SD.
Changes in SpO2.
Figure 1 shows SpO2 during the different experimental conditions. PEEP-10 induced a significant increase in SpO2 compared to all conditions except PEEP-5. PEEP-5 also induced a significant increase in SpO2 compared to FB and PEEP-0. Compared to FB, the average increase in SpO2 was 5.6 ± 5.0% with PEEP-10 and 3.3 ± 4.6% with PEEP-5. Neither T10 nor PLB induced significant improvement in SpO2 compared to FB and PEEP-0.
FB, free breathing with no PEEP device; PEEP-0/5/10, Positive end expiratory pressure at 0, 5 and 10 cmH2O; T10, same ventilatory level as measured in PEEP-10 condition but without PEEP device; PLB, pursed lips breathing. * significant difference between two conditions (* p < 0.05, **, p < 0.01, *** p < 0.001).
See Figure 1 for abbreviations. * significant difference between two conditions (*p < 0.05, **p < 0.01).
|HHb (Δµmol·cm)||-13.6 (42.5)||-18.2 (51)||-43.3 (82.6)||-4.6 (68.7)||-14.6 (65.6)|
|THb (Δµmol·cm)||-3.3 (23.2)||-22.6 (12.6)||-42.6 (30.5)||1.9 (20.5)||-4.1 (39.3)|
|TOI (Δ%)||1.1 (3.3)||1.0 (2.5)||2.7 (4.2)||0.3 (3.8)||0.6 (3.5)|
|TOI (%)||67.4 (6.0)||68.1 (4.9)||69.1 (5.7)||67.9 (6.5)||66.8 (5.0)|
Minute ventilation was larger in FB compared to PEEP-0 only (Figure 3). No other difference was observed between the experimental conditions. Table 2 shows end-tidal carbon dioxide partial pressure (PETCO2) and breathing pattern in all experimental conditions. PETCO2 was significantly larger in FB compared to all the other experimental conditions. Tidal volume was increased significantly with PEEP-10, T10 and PLB compared to FB while PLB only induced significantly larger tidal volume compared to PEEP-0. Breathing frequency was significantly larger in FB compared to all other experimental conditions while PLB only induced significantly lower values compared to PEEP-0. Expiratory time was smaller in FB compared to all other experimental conditions and it was also higher with PLB compared to all other conditions. Inspiratory time did not differ between conditions.
See Figure 1 for abbreviations. * significant difference between two conditions (p < 0.05).
|PETCO2 (mmHg)||37.1 (3.0)||35.7 (3.4)*||34.7 (3.9)*||34.6 (4.3)*||35.0 (4.5)*||36.4 (4.7)*|
|Vt (l)||0.83 (0.34)||0.93 (0.34)||0.94 (0.37)||1.07 (0.37)*||1.00 (0.51)*||1.26 (0.64)*$|
|fR (/min)||15.3 (4.9)||11.4 (3.8)*||11.5 (3.9)*||10.4 (4.0)*||11.7 (4.3)*||8.0 (4.8)*$|
|Ti (s)||2.1 (1.0)||2.5 (1.0)||2.2 (1.1)||2.4 (1.2)||2.8 (1.3)||2.5 (1.1)|
|Te (s)||2.5 (1.3)||3.5 (1.4)*||3.5 (1.6)*||4.0 (1.6)*||3.5 (1.5)*||8.1 (4.66)*$|
|Pes int expi (cmH2O·s·min-1)||5.7 (6.1)||10.4 (9.2)||14.9 (7.6)||22.8 (26.0)*£||3.8 (8.0)||45.1 (40.0)*$£|
|Pga int expi (cmH2O·s·min-1)||24.1 (14.7)||33.9 (14.1)||44.7 (23.8)||60.5 (23.1)*||28.4 (12.6)£||110.4 (96.9)*$£|
|VAS ‘well being’||1.5 (1.9)||1.7 (2.0)||1.5 (1.5)||1.4 (1.6)||1.8 (1.8)||1.6 (2.0)|
|VAS ‘breathing effort’||0.8 (1.2)||1.2 (1.4)||2.0 (1.6)||2.3 (1.8)||1.3 (1.3)||1.4 (1.1)|
Esophageal and gastric pressures.
Integrated expiratory Pes and Pga and amplitudes of expiratory Pes and Pga swings during the experimental conditions are showed in Table 2 and Figure 4, respectively. The integrals of expiratory Pes and Pga were significantly larger in PEEP-10 and PLB only compared to FB. PEEP-10 induced larger expiratory Pes swing compared to all the other conditions. PLB, PEEP-5 and PEEP-0 also induced larger expiratory Pes swing compared to FB. PEEP-10 and PLB induced larger expiratory Pga swing compared to all the other conditions.
Measurements of cardiac variables are shown in Table 3. All PEEP levels and T10 induced a significant decrease in HR compared to FB, without significant difference between PEEP, PLB and T10 conditions. SV was similar in all experimental conditions. Cardiac output was significantly larger in FB compared to all other conditions that did not differ between each other.
|HR (bpm)||68.9 (8.5)||65.9 (9.4)*||65.2 (10.0)*||64.8 (10.3)*||65.1 (9.9)*||66.9 (9.3)|
|Stroke volume (ml)||70.8 (20.0)||69.2 (17.5)||69.4 (16.0)||71.7 (14.8)||69.9 (15.9)||68.0 (16.2)|
|QC (l·min-1)||4.88 (1.48)||4.50 (1.06)*||4.44 (0.92)*||4.56 (0.79)*||4.49 (0.95)*||4.51 (1.08)*|
During the whole experimental session, no significant change in subjects’ perception of ‘breathing effort’ and ‘well-being’ was obtained (Table 2).
The increases in O2Hb and SpO2 from FB to PEEP-10 were significantly correlated (r = 0.71, p = 0.02). No other significant correlation was found between individual changes in oxygenation (SpO2 or O2Hb) and changes in ventilation, Pes, cardiac function and subjects’ perception for any condition (all r < 0.35 and p > 0.05).
At high altitude, PEEP-10 induced a significant increase in SpO2 compared to FB (Figure 5). SpO2 increased by 6.7 ± 6.0% from FB to the 3rd minute with PEEP-10, with no further significant improvement until the 20th minute with PEEP-10.
The present findings demonstrate that using a 10-cmH2O PEEP in healthy subjects under normobaric (FiO2 = 0.12) and hypobaric (4,350 m) hypoxia induced a significant increase in SpO2 (+5.6% and +6.7% on average after 3 min PEEP-10 compared to FB, respectively). Measurements in the laboratory show that PEEP-10 also increased quadriceps oxygenation. This improved oxygenation was accompanied by a significant increase in expiratory Pes but with no changes in minute ventilation, breathing pattern or PETCO2 compared to PEEP-0. No impairment in cardiac function was observed with increasing levels of PEEP and subjects did not report significant discomfort with positive expiratory pressure. These data suggest that breathing with a 10-cmH2O PEEP is an efficient non-pharmacological intervention to significantly improve arterial and muscle oxygenation under hypoxic conditions. This happens without substantial or deleterious changes in ventilation, cardiac function and subjects’ subjective perception of comfort. Conversely, PLB did not appear to provide similar positive effects on oxygenation.
Our results showed that PEEP-10 was the most efficient experimental condition to improve both SpO2 and muscle oxygenation (O2Hb) in healthy subjects exposed to hypoxia (Figures 1 and 2). Arterial and tissue oxygenation improvements were correlated. Even if quadriceps TOI was not significantly increased, a tendency for larger values with PEEP-10 was observed (p = 0.1) and further suggests an improvement in tissue oxygenation. While in the laboratory study the effects of PEEP breathing were measured for 3 min periods only, the field study showed that the SpO2 increase with PEEP-10 plateaued after 3 to 5 min (no further significant increase in SpO2 was observed after 3 min PEEP-10) and was maintained for up to 20 min (Figure 5). Importantly, the present results also show that PEEP-10 increases arterial oxygenation similarly during both normobaric and hypobaric hypoxic conditions despite their potential differences as recently emphasized .
The laboratory study showed that the oxygenation changes with PEEP-10 were observed without any increase in minute ventilation compared to PEEP-0, while PEEP conditions induced slightly lower minute ventilation compared to FB. Schoene et al.  have previously reported similar (in subjects with high-altitude pulmonary edema) or larger (in healthy subjects) minute ventilation with PEEP-10 compared to PEEP-0 at 4,400 m. Changes in minute ventilation with PEEP may depend on psychological factors (in the present study subjects were blinded for the PEEP level), subjects’ health status  or some PEEP device effects (as suggested by the significant difference between PEEP-0 and FB). Although no significant difference in breathing pattern was observed between PEEP conditions, PEEP-10 tended to induce deeper and slower breathing compared to PEEP-0. Such a breathing pattern may improve gas exchange and therefore potentially contribute to higher SpO2. However, the T10 condition that reproduced the same breathing pattern as PEEP-10 did not induce any improvement in blood or muscle oxygenation compared to PEEP-0 or FB, demonstrating that slight changes in breathing pattern associated with PEEP-10 are not the reasons for improved oxygenation. Hence, the improved oxygenation with PEEP-10 is not the consequence of changes in minute ventilation or breathing pattern as previously suggested .
Pes, an index of pleural and intrathoracic pressures, was significantly higher with PEEP-10 compared to other PEEP conditions or FB. This Pes increase suggests that alveolar pressure was enhanced with PEEP-10 and this could have played a major role regarding the increased SpO2. In case of an early pulmonary sub-edema, as suggested by Agostoni et al. , PEEP-10 may resorb some of the extra-vascular fluid accumulation and as a consequence improve oxygen diffusion. We recently reported significant extravascular fluid accumulation in healthy subjects during the first days of high altitude exposure similar to the present field study . While some reduction in extravascular fluid accumulation due to PEEP-10 may explain the improvement in arterial oxygenation in our field study, the presence of extravascular fluid accumulation during a short duration hypoxic exposure as in the laboratory study (< 2 h) remains however hypothetical. Other mechanisms such as increased operational lung volume and improved ventilation/perfusion matching (especially by increasing ventilation in lung regions with low ventilation/perfusion ratio) may underlie the improved gas exchange and arterial oxygenation observed in the present study.
An increase in intrathoracic and abdominal (as shown by Pga measurement in the present study) pressure is also known to decrease the venous return and therefore to have potential deleterious effects on cardiac function . However, in the present study, the absence of change in SV estimated by impedance cardiography suggests that applying a 10-cmH2O PEEP under resting conditions had no major consequences on venous return and left cardiac function. HR and consequently Qc were slightly but significantly reduced during all PEEP conditions compared to FB. The underlying mechanisms of this device-effect remain to clarify but may be, at least in part, related to the slight differences in ventilation observed between PEEP conditions and FB. Expiratory phases were longer in PEEP conditions compared to FB and it is known that respiratory-circulatory interactions such as respiratory sinus arrhythmia by which the ECG R-R interval is shortened during inspiration and prolonged during expiration (i.e., reduced HR), are part of the efficiency of pulmonary gas exchanges 
No significant discomfort was reported by the subjects when using PEEP as indicated by the levels of ‘breathing effort’ and ‘well-being’ scored all over the protocol. These results indicate that using PEEP even at 10 cmH2O in healthy subjects does not induce any deleterious perceptions and is well tolerated both under simulated or real altitude conditions.
PEEP-5 also induced a significant increase in SpO2 compared to both FB and PEEP-0 conditions, but no change in quadriceps oxygenation was induced. The relevance of this improvement in arterial oxygenation is questionable since the average increase in SpO2 induced by PEEP-5 (3.3%) was clearly below the 4.9% SpO2 difference previously reported to distinguish subjects with or without symptoms at altitude . Pes measurements indicate that PEEP-5 induced significantly lower increase in intrathoracic pressure and possibly in alveolar pressure compared to PEEP-10. Conversely, PEEP-5 had similar effects than PEEP-10 on ventilation and cardiac variables. Hence, we suggest that PEEP-5 does not sufficiently increase intrathoracic pressure to promote relevant improvement in both SpO2 and O2Hb as observed with PEEP-10.
In our laboratory study assessing for the first time the effect of PLB in a group of healthy subjects under normobaric hypoxic condition, PLB did not induce any significant change in SpO2 and quadriceps oxygenation. These results are in contrast to the case study by Tannheimer et al.  reporting a large SpO2 increase in a subject with high altitude pulmonary edema at 4,330 m after 30 min of PLB. An important difference between Tannheimer’s study and ours is that we investigated healthy subjects during a short-duration hypoxic exposure with 3-min phases whereas PLB was investigated after a 4-day rapid ascent and during 30 min in Tannheimer’s study. Also, SpO2 during FB was higher in our subjects (78 ± 6% with FiO2 = 0.12) compared to Tannheimer’s case report (62%) where the potential for an improvement in SpO2 was larger. Moreover, expiratory pressure and minute ventilation were not measured in this previous study, making comparisons with the present work difficult. The specific changes induced by PEEP-10 compared to PLB regarding breathing pattern (expiratory duration twice longer with PLB) and Pes (greater integrated expiratory Pes but lower expiratory Pes swing with PLB) may explain the absence of oxygenation improvement with PLB.
Several limitations of the present study should be underlined. First, in the laboratory study each experimental condition was tested for 3-min periods only. The field study showed however that most of the SpO2 change with PEEP-10 observed over 20 min occurs within this time frame. In addition, physiological variables measured during the laboratory study reached steady states during all 3-min phases and repeating all conditions in a randomized order two times allowed us to verify the absence of any incidental effect on the main outcomes. The laboratory study was performed during a short hypoxic exposure duration (i.e. <2 h) while the field study was performed at least 2 days after arrival at 4,350 m of altitude. This has two important consequences. First, laboratory and field hypoxic conditions were associated with different states of physiological adaptation to hypoxia. The present results suggest however that PEEP has similar effect on arterial oxygenation despite differences in prior hypoxic exposure duration in the laboratory and field studies, which is in contrast with previous results suggesting that the effect of continuous positive airway pressure on arterial oxygen saturation may depend on the time spent at high altitude . Second, it is well known that AMS normally appears within 8 to 48 h after arrival at high altitude. Therefore, since the subjects in the laboratory had no AMS symptoms while those at altitude had no more significant AMS symptoms after the initial 2 days of acclimatization (Lake Louise score < 3 ), the present study cannot evaluate whether PEEP may improve AMS symptoms and acclimatization to high altitude. Further studies are needed to clarify the influence of previous hypoxic exposure duration on the effect of PEEP and to evaluate its ability to reduce AMS symptoms. At last, it should be acknowledged that impedance cardiography and NIRS are indirect measurements of cardiac output and tissue oxygenation, respectively. The measurement of cardiac output with the Physioflow® system has been validated against the direct Fick method . Although NIRS measurement displays important limitations such as the relatively superficial portion of the tissue assessed and the potential influence of superficial layers (skin, subcutaneous fat, etc) , it provides in the present study a reliable estimate of relative changes in muscle hemoglobin oxygenation and oxygenation index between free breathing and PEEP breathing conditions.
To conclude, our study shows for the first time that PEEP-10 significantly improves both SpO2 and O2Hb in healthy subjects during hypoxic exposure, without significant changes in ventilation, impairments of cardiac function and subjects’ subjective perception of discomfort. The main physiological mechanism promoting improved oxygenation may be the increased intrathoracic and consequently alveolar pressures induced by PEEP-10, enhancing alveolo-capillary oxygen diffusion. The rapid ~6% average increase in SpO2 induced by PEEP-10 may be relevant to improve AMS symptoms  and could be an attractive non-pharmacological tool to prevent or treat high-altitude illnesses in combination with other interventions.
Conceived and designed the experiments: HN TR DB RT BW PL SV. Performed the experiments: HN TR DB BW SV. Analyzed the data: HN TR DB SV. Contributed reagents/materials/analysis tools: HN TR DB SV. Wrote the manuscript: HN TR DB RT BW PL SV. Discussed the data: HN TR DB RT BW PL SV.
- 1. Nield MA, Soo Hoo GW, Roper JM, Santiago S (2007) Efficacy of pursed-lips breathing: a breathing pattern retraining strategy for dyspnea reduction. J Cardiopulm Rehabil Prev 27: 237-244. doi:https://doi.org/10.1097/01.HCR.0000281770.82652.cb. PubMed: 17667021.
- 2. Mueller RE, Petty TL, Filley GF (1970) Ventilation and arterial blood gas changes induced by pursed lips breathing. J Appl Physiol 28: 784-789. PubMed: 5419502.
- 3. Dongelmans DA, Hemmes SN, Kudoga AC, Veelo DP, Binnekade JM et al. (2012) Positive end-expiratory pressure following coronary artery bypass grafting. Minerva Anestesiol 78: 790-800. PubMed: 22475803.
- 4. Reissmann HK, Ranieri VM, Goldberg P, Gottfried SB (2000) Continuous positive airway pressure facilitates spontaneous breathing in weaning chronic obstructive pulmonary disease patients by improving breathing pattern and gas exchange. Intensive Care Med 26: 1764-1772. doi:https://doi.org/10.1007/s001340000725. PubMed: 11271083.
- 5. Pinto Da Costa N, Di Marco F, Lyazidi A, Carteaux G, Sarni M et al. (2011) Effect of pressure support on end-expiratory lung volume and lung diffusion for carbon monoxide. Crit Care Med 39: 2283-2289. doi:https://doi.org/10.1097/CCM.0b013e3182227fce. PubMed: 21666442.
- 6. Di Marco F, Devaquet J, Lyazidi A, Galia F, da Costa NP et al. (2010) Positive end-expiratory pressure-induced functional recruitment in patients with acute respiratory distress syndrome. Crit Care Med 38: 127-132. doi:https://doi.org/10.1097/CCM.0b013e3181b4a7e7. PubMed: 19730254.
- 7. Hackett PH, Roach RC (2001) High-altitude illness. N Engl J Med 345: 107-114. doi:https://doi.org/10.1056/NEJM200107123450206. PubMed: 11450659.
- 8. Burtscher M, Flatz M, Faulhaber M (2004) Prediction of susceptibility to acute mountain sickness by SaO2 values during short-term exposure to hypoxia. High Alt Med Biol 5: 335-340. doi:https://doi.org/10.1089/ham.2004.5.335. PubMed: 15453999.
- 9. Richalet JP, Larmignat P, Poitrine E, Letournel M, Canouï-Poitrine F (2012) Physiological risk factors for severe high-altitude illness: a prospective cohort study. Am J Respir Crit Care Med 185: 192-198. doi:https://doi.org/10.1164/rccm.201108-1396OC. PubMed: 22071330.
- 10. Nespoulet H, Wuyam B, Tamisier R, Saunier C, Monneret D et al. (2012) Altitude illness is related to low hypoxic chemoresponse and low oxygenation during sleep. Eur Respir J 40: 673-680. doi:https://doi.org/10.1183/09031936.00073111. PubMed: 22523356.
- 11. Agostoni P, Caldara G, Bussotti M, Revera M, Valentini M et al. (2010) Continuous positive airway pressure increases haemoglobin O2 saturation after acute but not prolonged altitude exposure. Eur Heart J 31: 457-463. doi:https://doi.org/10.1093/eurheartj/ehp472. PubMed: 19903683.
- 12. Larson EB (1985) Positive airway pressure for high-altitude pulmonary oedema. Lancet 1: 371-373. PubMed: 2857423.
- 13. Launay JC, Nespoulos O, Guinet-Lebreton A, Besnard Y, Savourey G (2004) Prevention of acute mountain sickness by low positive end-expiratory pressure in field conditions. Scand J Work Environ Health 30: 322-326. doi:https://doi.org/10.5271/sjweh.801. PubMed: 15458016.
- 14. Savourey G, Caterini R, Launay JC, Guinet A, Besnard Y et al. (1998) Positive end expiratory pressure as a method for preventing acute mountain sickness. Eur J Appl Physiol Occup Physiol 77: 32-36. PubMed: 9459518.
- 15. Schoene RB, Roach RC, Hackett PH, Harrison G, Mills WJ Jr (1985) High altitude pulmonary edema and exercise at 4,400 meters on Mount McKinley. Effect of expiratory positive airway pressure. Chest 87: 330-333. doi:https://doi.org/10.1378/chest.87.3.330. PubMed: 3882350.
- 16. Savourey G, Besnard Y, Launay JC, Guinet A, Hanniquet AM et al. (2001) Positive end expiratory pressure (PEEP) slightly modifies ventilatory response during incremental exercise. Aviat Space Environ Med 72: 21-24. PubMed: 11194989.
- 17. Luecke T, Pelosi P (2005) Clinical review: Positive end-expiratory pressure and cardiac output. Crit Care 9: 607-621. doi:https://doi.org/10.1186/cc3795. PubMed: 16356246.
- 18. Spahija J, de Marchie M, Grassino A (2005) Effects of imposed pursed-lips breathing on respiratory mechanics and dyspnea at rest and during exercise in COPD. Chest 128: 640-650. doi:https://doi.org/10.1378/chest.128.2.640. PubMed: 16100149.
- 19. Ugalde V, Breslin EH, Walsh SA, Bonekat HW, Abresch RT et al. (2000) Pursed lips breathing improves ventilation in myotonic muscular dystrophy. Arch Phys Med Rehabil 81: 472-478. doi:https://doi.org/10.1053/mr.2000.3790. PubMed: 10768538.
- 20. Tannheimer M, Tannheimer S, Thomas A, Engelhardt M, Schmidt R (2009) Auto-PEEP in the therapy of AMS in one person at 4,330 m. Sleep Breath 13: 195-199. doi:https://doi.org/10.1007/s11325-008-0237-z. PubMed: 19052788.
- 21. Aguilaniu B, Maitre J, Glénet S, Gegout-Petit A, Guénard H (2008) European reference equations for CO and NO lung transfer. Eur Respir J 31: 1091-1097. doi:https://doi.org/10.1183/09031936.00063207. PubMed: 18216061.
- 22. Charloux A, Lonsdorfer-Wolf E, Richard R, Lampert E, Oswald-Mammosser M et al. (2000) A new impedance cardiograph device for the non-invasive evaluation of cardiac output at rest and during exercise: comparison with the "direct" Fick method. Eur J Appl Physiol 82: 313-320. doi:https://doi.org/10.1007/s004210000226. PubMed: 10958374.
- 23. Ferrari M, Muthalib M, Quaresima V (2011) The use of near-infrared spectroscopy in understanding skeletal muscle physiology: recent developments. Philos Trans A Math Phys Eng Sci 369: 4577-4590. doi:https://doi.org/10.1098/rsta.2011.0230. PubMed: 22006907.
- 24. Millet GP, Faiss R, Pialoux V (2012) Point: Hypobaric hypoxia induces different physiological responses from normobaric hypoxia. J Appl Physiol (1985) 112: 1783-1784. doi:https://doi.org/10.1152/japplphysiol.00067.2012. PubMed: 22267386.
- 25. Bouzat P, Walther G, Rupp T, Doucende G, Payen JF et al. (2013) Time course of asymptomatic interstitial pulmonary oedema at high altitude. Respir Physiol Neurobiol 186: 16-21. doi:https://doi.org/10.1016/j.resp.2012.12.009. PubMed: 23279868.
- 26. Yasuma F, Hayano J (2004) Respiratory sinus arrhythmia: why does the heartbeat synchronize with respiratory rhythm? Chest 125: 683-690. doi:https://doi.org/10.1378/chest.125.2.683. PubMed: 14769752.
- 27. Roach RC, Bartsch P, Hackett PH, Oelz O (1993) The lake louise acute mountain sickness scoring system. In: Sutton JR, Coates J, S. HC, editors. Hypoxia and Molecular Medicine: Queen City Printers. pp. 272-274.
- 28. Ferrari M, Mottola L, Quaresima V (2004) Principles, techniques, and limitations of near infrared spectroscopy. Can J Appl Physiol 29: 463-487. doi:https://doi.org/10.1139/h04-031. PubMed: 15328595.