Systemic but not local rehydration restores dehydration-induced changes in pulmonary function in healthy adults

Water transport and local (airway) hydration are critical for the normal functioning of lungs and airways. Currently, there is uncertainty regarding the effects of systemic dehydration on pulmonary function. Our aims were 1 ) to clarify the impact of exercise- or ﬂ uid restriction-induced dehydration on pulmonary function in healthy adults; and 2 ) to establish whether systemic or local rehydration can reverse dehydration-induced alterations in pulmonary function. Ten healthy participants performed four experimental trials in a randomized order (2 h exercise in the heat twice and 28 h ﬂ uid restriction twice). Pulmonary function was assessed using spirometry and whole body plethysmography in the euhydrated, dehydrated, and rehydrated states. Oral ﬂ uid consumption was used for systemic rehydration and nebulized isotonic saline inhalation for local rehydration. Both exercise and ﬂ uid restriction induced mild dehydration (2.7± 0.7% and 2.5± 0.4% body mass loss, respectively; P < 0.001) and elevated plasma osmolality ( P < 0.001). Dehydration across all four trials was accompanied by a reduction in forced vital capacity (152 ±143mL, P < 0.01) and concomitant increases in residual volume (216 ± 177 mL, P < 0.01) and functional residual capacity (130± 144 mL, P < 0.01), with no statistical differences between modes of dehydration. These changes were normalized by ﬂ uid consumption but not nebulization. Our results suggest that, in healthy adults: 1 ) mild systemic dehydration induced by exercise or ﬂ uid restriction leads to pulmonary function impairment, primarily localized to small airways; and 2 ) systemic, but not local, rehydration reverses these potentially deleterious alterations. NEW & NOTEWORTHY This study demonstrates that, in healthy adults, mild systemic dehydration induced by exercise in the heat or a prolonged period of ﬂ uid restriction leads to negative alterations in pulmonary function, primarily localized to small airways. Oral rehydration, but not nebulized isotonic saline, is able to restore pulmonary function in dehydrated individuals. Our ﬁ ndings highlight the importance of maintaining an adequate systemic ﬂ uid balance to preserve pulmonary function.


INTRODUCTION
Systemic dehydration, defined as a deficit of total body water (1), commonly occurs in individuals who perform sustained physical activity in hot environments (2), as well as in patients and older adults (3). Even at mild levels [i.e., $2-3% body mass loss (1)], systemic dehydration can have unfavorable effects on multiple organ systems, including the cardiovascular, renal, and nervous systems (4) and can compromise physical and cognitive performance (1). Limited and contradictory data currently exist regarding the effects of systemic dehydration on the respiratory system.
Two previous studies showed deleterious alterations in expiratory flow or lung volumes in healthy adults (5) and in athletes with mild asthma (6) following mild systemic dehydration (up to 2.5% body mass loss). A third study (7) however showed improvements in selected measures of pulmonary function [including increases in forced expiratory volume in 1 s (FEV 1 ) and in the FEV 1 to forced vital capacity ratio (FEV 1 /FVC)] in healthy adults following moderate dehydration (4.5% body mass loss). The different population groups and severities of dehydration, as well as the various modes of dehydration [i.e., fluid restriction (5) vs. exercise (6) vs. diuretic drug (7)], have generated uncertainty regarding the impact of systemic dehydration on pulmonary function.
Fluid supply to the airways stems primarily from the bronchial circulation, which itself arises from the systemic circulation (8). Optimal lung fluid balance is a critical component of normal pulmonary functioning (9), with airway surface liquid dehydration implicated in several respiratory diseases, such as cystic fibrosis (10) and exercise-induced bronchoconstriction (11). Water flows across the airway epithelia in response to an osmotic gradient. Thus, when individuals become dehydrated and bronchial blood flow and/or composition changes, water movement to the airways is modified and airway hydration may become compromised. Alterations in airway surface liquid thickness ("depth"), composition, and/or rheology can promote peripheral or small airway instability and provoke premature airway closure (12), potentially increasing respiratory symptoms (in particular, breathlessness) and worsening respiratory reserve in susceptible individuals.
The aim of this study was to clarify the impact of mild systemic dehydration on pulmonary function in healthy adults. Since exercise in hot environment and/or insufficient fluid intake are two common causes of systemic dehydration, we compared the effects of 2 h of exercise in the heat with 28 h of fluid restriction on pulmonary volumes, capacities, and flows. In line with our findings in athletes with asthma (6), we hypothesized that pulmonary function would be compromised in dehydrated healthy adults, as evidenced by changes in spirometry and whole body plethysmography parameters. We did not anticipate any difference between dehydration modalities, as both exercise and fluid restriction cause the same type of dehydration, i.e., hyperosmotic hypovolemia (13). Additionally, we aimed to establish: 1) whether dehydration-induced changes in pulmonary function are reversible with immediate rehydration; and 2) whether local rehydration delivered directly at the site of the airways (via nebulized isotonic saline) is superior to systemic rehydration (via oral fluid intake) in restoring pulmonary function.

METHODS Participants
Ten healthy, nonsmoking individuals (2 females; age: 29 ± 8 yr; body mass: 62.8 ± 8.5 kg; and stature: 173 ± 10 cm), with no history of respiratory illness (including asthma and exercise-induced bronchoconstriction) completed this study. Pulmonary function was checked for normality via spirometry, with FEV 1 >80% predicted and FEV 1 /FVC >70% used as inclusion criteria and Global Lung Function Initiative 2012 equations used as reference (14). Participants provided written informed consent and the study was approved by the Brunel University London Research Ethics Committee (Reference No. 6639-TISS-Jul/2017-7774-2).

Experimental Overview
A repeated-measures, randomized crossover design was utilized, with all participants completing four experimental trials on separate days. The trials comprised: 1) fluid restriction plus systemic rehydration; 2) fluid restriction plus local rehydration; 3) exercise plus systemic rehydration; and 4) exercise plus local rehydration.
Alcohol, caffeine, and strenuous exercise were prohibited in the 24 h before testing. Before the first trial, participants completed a 24 h food diary, which they subsequently replicated before each trial.
Participants arrived at the laboratory at 0900 h (±1 h) in a euhydrated state. Basic anthropometrics, hydration status, and pulmonary function were assessed. Resting minute ventilation (V _ E ) was then recorded to allow for estimation of airway surface liquid loss over the 28 h fluid restriction trials. Following baseline measurements, participants underwent one of two dehydration trials: 28 h fluid restriction or exercise-induced dehydration. Upon completion of the dehydration trials, hydration status and pulmonary function were reassessed; this was followed by a period of rehydration with oral fluid or isotonic saline nebulization. Spirometry was performed 15 and 35 min within the rehydration periods. Hydration status and pulmonary function were recorded 60 min after commencing rehydration.

Hydration Status
Capillary blood samples were collected in triplicate from the participant's fingertip. The samples were analyzed for hemoglobin (Hb) concentration (HemoCue Ltd., Dronfield, Derbyshire, UK) and hematocrit (Hct). Hematocrit tubes were centrifuged at 12-14,000 rpm for 3 min (Hematospin 1300 Centrifuge, Hawksley & Sons Ltd., West Sussex, UK) and assessed via microscopy. Hemoglobin and Hct were then used for the calculation of plasma volume (15). Plasma osmolality (P osm ) was analyzed via freezing point depression osmometry (Advanced 3320 Micro-Osmometer, Vitech Scientific Ltd., West Sussex, UK). Urine osmolality (U osm ) was measured using a portable refractometer (Pocket Pal-Osmo, Atago Vitech Scientific, Scotland, UK), and the threshold for euhydration was set at <700 mosmol/kgH 2 O (2). Nude body mass was recorded to the nearest 0.1 kg.

Pulmonary Function
Pulmonary function was measured via spirometry then whole body plethysmography. All tests were performed in accordance with American Thoracic Society/The European Respiratory Society guidelines (16,17). At 15 and 35 min of rehydration, forced maneuvers were performed in duplicate only (as long as FEV 1 and FVC were reproducible). Specific airway resistance (sRaw) was measured using the interrupter (i.e., airway occlusion) technique (18).

Dehydration Protocols
Fluid restriction.
As done previously (19), participants were prohibited from consuming any fluid and were restricted to consuming foods with low water content (<30%) from a list of acceptable/prohibited foods (4) for 28 h. Participants kept a diary of all food consumed, which they replicated during the second fluid restriction trial. Participants were fitted with an activity monitor (ActivPal, PAL Technologies Ltd., Glasgow, UK) and asked to limit physical activity for the entire duration of the fluid restriction. Local environmental temperature and humidity were recorded throughout 28 h using a portable logger (Hygrochron, iButton, Maxim Integrated, CA) and were later used to estimate absolute water content of inspired air and airway surface liquid loss (11,20). Expired water content was assumed to equal 33 mgH 2 O·L À 1 of air (11). The difference between inspired and expired absolute water content was calculated and multiplied by resting V _ E . Total water loss over the duration of the fluid restriction trial was estimated assuming negligible variations in V _ E over the 28 h period.

Exercise-induced dehydration.
Participants completed 2 h of low-intensity exercise in hot conditions [37 C, 50% relative humidity (RH)] with total fluid restriction. The exercise protocol was identical to that used in our previous study (6), alternating 20 min of cycling on a stationary bike (Excalibur Sport, Lode, Groningen, The Netherlands) and 10 min of stepping. At the end of each bout of cycling and stepping, the following measurements were taken: heart rate (FT1, Polar Electro Oy, Kempele, Finland), tympanic temperature (Thermoscan Exactemp 6022, Braun, Germany), overall rating of perceived exertion [RPE; on a scale of 6-20 (21)] and rating of breathing discomfort [on a scale of 1-10 (22)]. During the final 5 min of each bout of cycling/stepping, V _ E and oxygen uptake (V _ O 2 ) were recorded breath-by-breath (Vyntus CPX, Carefusion, Germany), with mean V _ E over the final minute used for analysis. Mean V _ O 2 over 2 h of exercise was used for estimation of airway surface liquid loss, based on calculations provided by Mitchell et al. (23): where m _ e is the rate of evaporative water loss in the expired air (g·min À1 ) and V _ O 2 is the oxygen uptake (L·min À1 STPD).

Systemic rehydration.
Participants gradually rehydrated by ingesting water at room temperature, mixed with 3 g NaCl·L H 2 O À1 to improve fluid retention (24). The volume of fluid ingested (liters) matched the loss of body mass (kg). Participants ingested water in three equal boluses (550 ± 176 mL) over 15 min, with a 5-min break between boluses to perform spirometry.

Local rehydration.
An ultrasonic nebulizer (UltraNeb, DeVilbiss Healthcare Ltd., UK) and isotonic saline (0.9%) were used to locally rehydrate the airways at a measured output of 1.4 ± 0.2 mL·min À1 . Participants were required to breathe tidally through a two-way nonrebreathing valve (Series 1410, Hans Rudolph, Inc., KS) with a nose clip in place. Participants were exposed to three 15-min bouts of nebulization, with 5 min breaks in between.

Sample Size
Sample size was based on our previous work that investigated the impact of exercise-induced dehydration upon pulmonary function in athletes with mild asthma (6). We hypothesized that 1) while still present, the reduction in FVC (i.e., primary outcome measure) following dehydration will be less severe in the healthy participants recruited in the present study compared with individuals with asthma (6); and 2) the change in FVC will be independent to the mode of dehydration. To detect a 200 mL difference in FVC between pre-and postdehydration, with a standard deviation of the difference in the response of matched pairs of 50 mL, we calculated that a sample size of eight participants would give $80% power for an alpha level of 5%.

Statistics
Statistical analyses were performed using dedicated software (SPSS version 26, SPSS, IBM Corp., Armonk, NY). All data were normally distributed, as confirmed by the Shapiro-Wilk test. To assess changes in spirometry, plethysmography, and hydration within and between trials, three-way repeated measures ANOVA were used (with mode of dehydration, mode of rehydration, and time as main factors). Post hoc Bonferroni-adjusted pairwise comparisons were used where significant main or interaction effects were detected. A within-subjects repeated measures correlation (25) was used to determine the relationship between changes in hydration and pulmonary function. Statistical significance was set at P < 0.05. Descriptive statistics are shown as means ± SD, unless otherwise stated. Differences between modes of dehydration/rehydration are expressed as means and 95% confidence intervals (CI).

Dehydration Protocols
Fluid restriction.

Exercise.
Participants cycled at 70 ± 9 W. Physiological (heart rate, tympanic temperature, and V _ E ) and perceptual responses (overall RPE and rating of breathing discomfort), as well as estimated total water loss from the airways were not different between trials (Table 1). Compared with fluid restriction, the estimated total water loss from the airways during exercise was smaller (P < 0.001), whereas the estimated rate of water loss was larger (P < 0.001). The mean differences between exercise and fluid restriction were 262 mL (95% CI: 201-323 mL) for total water loss and 28 mL·h À1 (95% CI: 22-33 mL·h À1 ) for rate of water loss.

Hydration Status
Data for hydration status are shown in Table 2. Data are means ± SD; n = 10. EX-Systemic, exercise with systemic rehydration trial; EX-Local, exercise with local rehydration (nebulized isotonic saline) trial. The overall rating of perceived exertion was rated on a scale of 6-20, while breathing discomfort was rated on a scale of 0-10. All P > 0.05 for between-trial differences.

Effect of dehydration.
Both modes of dehydration induced a mild level of dehydration, with a reduction in body mass of 2.5 ± 0.4% after exercise and 2.7 ± 0.7% after fluid restriction (both P < 0.001 vs. baseline). No significant difference was noted between trials [mean difference in body mass loss between exercise and fluid restriction: 0.1 kg (95% CI: -0.1-0.3 kg)]. The reduction in body mass was associated with an increase in P osm in all trials (P < 0.001), but no difference between modes of dehydration [mean difference in dehydration-induced P osm change between exercise and fluid restriction: 0.8 mosmol/ kgH 2 O (95% CI: -1.8-3.3 mosmol/kgH 2 O)]. U osm increased following exercise and fluid restriction (P < 0.001), but the increase was greater following fluid restriction (P = 0.001). Mean difference in dehydration-induced U osm change between exercise and fluid restriction was 408 mosmol/kgH 2 O (95% CI: 267-549 mosmol/kgH 2 O). Plasma volume, hemoglobin and hematocrit were not different after exercise or fluid restriction compared with baseline.

Effect of dehydration.
As illustrated in Fig. 1, mild dehydration induced by exercise and fluid restriction led to a reduction in FVC (P < 0.001) but not FEV 1 . The change in FVC was not different between modes of dehydration (exercise: -173 ± 169 mL; fluid restriction: -131 ± 70 mL), with a mean difference of 43 mL (95% CI: Data are means ± SD; n = 10. FR-Systemic, fluid restriction with systemic rehydration (oral fluid intake); EX-Systemic, exercise with systemic rehydration; FR-Local, fluid restriction with local rehydration (nebulized isotonic saline); EX-Local, exercise with local rehydration. a P < 0.05 vs. baseline; b P < 0.05 vs. baseline and dehydration; c P < 0.05 vs. FR-Local and EX-Local at the corresponding time point; d P < 0.05 vs. dehydrated; e P < 0.05 vs. EX-Systemic and EX-Local at the corresponding time point.

Whole Body Plethysmography
Whole body plethysmography data are presented in Table 3.

Correlation Analysis
The percent change in body mass in response to dehydration and rehydration (60 min) showed a moderate positive correlation with the change in FVC (r = 0.643, P < 0.001;

DISCUSSION
The findings from this study show negative alterations in pulmonary function in mildly dehydrated healthy adults following both 2 h of exercise in the heat and 28 h of fluid restriction. The observed reduction in FVC combined with an elevated RV and FRC suggest that the dehydrationinduced pulmonary impairment is primarily localized to the small airways. Dehydration-induced alterations in pulmonary function were reversed following acute systemic rehydration (via oral fluid intake) but not following local rehydration of the airways (via nebulized isotonic saline). Systemic hydration, via plasma osmolality, may therefore play a regulatory role in the maintenance of small airway patency in healthy humans.

Effects of Dehydration
This study shows that mild systemic dehydration, induced by both exercise and fluid restriction, leads to a reduction in FVC ($150 mL) and elevations in RV ($220 mL) and FRC ($130 mL) in healthy adults. These findings are in line with our previous work that demonstrated negative alterations in FVC, RV and FRC in athletes with mild asthma following 2 h of exercise in the heat (6). That we were able to replicate our previous findings in a healthy population suggests that dehydration-induced pulmonary impairment is a general phenomenon that is present irrespective of the presence of pulmonary or airway disease. The reduced severity of the pulmonary function alterations (mean reduction in FVC following exercise $170 mL vs. $300 mL in asthmatic individuals (6), with only 30% of our healthy participants reaching the "clinical threshold" of 200 mL (26) suggests that, while still affected, the airways of healthy individuals have a higher tolerance to systemic dehydration in comparison with individuals with preexisting lung conditions.
Our spirometry results are in contrast to those previously obtained in healthy individuals (see introduction). Govindaraj (5) reported that mild dehydration (2.0 ± 0.9% loss of body mass) induced by 16 h of fluid restriction caused negligible changes in FVC but was associated with a significant reduction in FEV 1 ($180 mL). While we cannot explain the differences observed in FEV 1 , the absence of a detectable change in FVC may be explained by the fact that in the study by Govindaraj (5) only 5 out of the 20 participants lost >2% body mass, whereas all our participants reached this threshold. According to Cheuvront and Kenefick (1), a day-to-day change in body mass >2% provides 95% probabilistic certainty that dehydration has occurred. A further study conducted by Javaheri et al. (7) showed improvements in pulmonary function (incl. FEV 1 , FEV 1 /FVC, and flow rates at all lung volumes) following moderate dehydration (i.e., 4.0 to 4.5% loss of body mass) induced by diuretics in a small sample (n = 6) of healthy men. The use of chlorthalidone could however explain the divergent findings, as diuretics cause iso-osmotic hypovolemia, whereas exercise and fluid restriction lead to hyperosmotic hypovolemia (13).

Effects of Rehydration
A novel finding of the current study is that systemic rehydration was effective at restoring selected lung volumes and Data are means ± SD; n =10. FR-Systemic, fluid restriction with systemic rehydration (oral fluid intake); EX-Systemic, exercise with systemic rehydration; FR-Local, fluid restriction with local rehydration (nebulized isotonic saline); EX-Local, exercise with local rehydration. a P < 0.05 vs. baseline; b P < 0.05 vs. dehydration; c P < 0.05 vs. FR-Local and EX-Local at the corresponding time point, d P < 0.05 vs. EX-Systemic and EX-Local at the corresponding time point; e P < 0.05 vs. baseline and dehydration. capacities (i.e., FVC, RV, and FRC). The positive effect of oral fluid intake on FVC was noted after only 15 min, which suggests a rapid reversal of the pulmonary alterations. Previously, dehydration-induced alterations in FVC, RV, and FRC were not restored following 40 min ad libitum water intake in individuals with asthma (6). The use of a matchedvolume rehydration strategy [with 100% of fluid replaced vs. 61 ± 19% in previous work (6)], together with administration of a hypertonic solution known to improve fluid retention (24), enabled us to return body mass close to baseline within an hour. In contrast, following nebulized isotonic saline rehydration, body mass was maintained below (-1.7 ± 0.5 kg) and P osm above (9 ± 4 mosmol/kgH 2 O) baseline, and FVC, RV, and FRC were not restored. Our findings therefore suggest that oral hypertonic fluid intake, but not nebulized isotonic saline solution, is an effective strategy to reverse dehydrationinduced pulmonary alterations.

Interpretation of Findings
A decrease in FVC alongside concomitant increases in RV, RV/TLC, and FRC is usually indicative of airway closure and air trapping (12). Our findings therefore suggest that systemic dehydration may selectively impair small airway function. Alterations in spirometry and plethysmography were noted conjointly with increases in P osm following both dehydration modalities. Together with the reversal of FVC, RV, FRC, and RV/TLC under systemic rehydration only (i.e., when P osm was normalized) and the significant association between P osm and lung volumes, our finding points toward P osm as a key determinant of the small airway impairment.
Pogson et al. (27) reported an inverse correlation between increased serum osmolality and decreased FVC and FEV 1 in a large population (>10,000) of patients suffering from chronic obstructive pulmonary disease. The authors suggested a causal relationship, mediated by airway epithelial cells, between increased serum plasma osmolality and reduced pulmonary function. Airway epithelial cells are "osmotic transducers" (28) that respond to changes in osmolality of both their extracellular and intracellular environments. Through controlled secretion and/or absorption of salt and water, airway epithelial cells preserve hydration of the airways and maintain water and osmolyte homeostasis in human lungs (29). In our dehydration trials, we postulate that airway epithelia "detected" the increase in P osm in bronchial vasculature, which, in turn, would have influenced water supply to the airways and altered the composition and/or content of the airway surface liquid. The common functional implication of pertubations to the airway surface liquid is peripheral airway instability and premature airway closure (12), which aligns with the lung volume changes observed in our participants. During the systemic rehydration trials (i.e., oral fluid consumption), the rapid normalization of P osm is likely to have facilitated the return of airway surface liquid to its hydrated state; this would have decreased surface tension and reopened the collapsed airways, thereby explaining the rapid restoration of lung volumes to baseline. That previous studies (30,31) have evidenced restoration of plasma volume and extracellular osmolality (i.e., P osm ) within the timeframe used for our rehydration trials (30 to 60 min) supports the idea that extracellular hypervolemia following fluid consumption facilitates the recovery of small airway patency, even if full recovery of intracellular and interstitial compartments may take up to 4 h following rehydration (32). During the local rehydration trials (i.e., nebulized isotonic saline), the mucus layer may have acted as a "liquid reservoir" (33), with no or little influence on airway surface liquid ionic composition and, thereby, no ensuing reversal of small airway collapse and no restoration of pulmonary function.
In airway epithelia, numerous mechanisms ensure efficient control of airway surface fluid "depth" and composition. These mechanisms involve both passive surface forces (dependent on hydration status of the mucus layer) and active ion transport mechanisms (located within the airway  lining cells) (9). In our experiment, while the fluid restriction trials would have provided a prolonged window (28 h) for active ion transport to take place, thereby favoring maintenance of airway hydration, the involvement of such mechanism during the shorter (2 h) exercise trials is questionable. However, in cultures of differentiated human airway epithelia, rapid (within minutes) and transient increases in paracellular sodium conductance have been observed (34). Ion flow through the epithelial cells may therefore have counteracted the effects of systemic dehydration on local (airway) fluid availability in both sets of trials. The extent of this effect and implication for the severity of airway impairment require further work.

Methodological Considerations
The average $2.6% loss of body mass in the present study was well matched with the $2.3% loss in our previous study  of athletes with mild asthma (6) and across the four experimental trials. While the 28-h duration was required to induce the target degree of dehydration (19), it did not allow us to record pulmonary function at the same time of day within trials. To exclude diurnal variation as a confounder to the observed changes, we invited a subset of the participants (n = 5; 50% of our initial sample) to perform an additional "control" visit. Spirometry and whole body plethysmography were performed in a euhydrated resting state at matched time points to the experimental trials. The difference between morning and afternoon values was 85 ± 92 mL for FVC, -76 ± 200 mL for RV, and -54 ± 198 mL for FRC. As the directions of change were opposite to those found following dehydration, we can exclude diurnal variation as a confounding influence on the observed alterations in pulmonary function.
To account for the possible effect of evaporative water loss, through pulmonary ventilation, on airway surface liquid osmolarity (35), we estimated airway water losses during our dehydration trials. Airway water loss was greater during fluid restriction ($340 mL) compared with exercise ($80 mL) and accounted for 23% of total body water loss during the 28h fluid restriction period versus only 5% during the 2 h of exercise. The nonsignificant differences in pulmonary impairments between the two modes of dehydration suggest that evaporative water loss was not a significant contributor to the observed changes. However, as typically reported in the literature (36,37), relatively large variability was noted for some of our outcome measures (including RV and FRC). Therefore, to ascertain a lack of differential effect of exercise versus fluid restriction on dehydration-induced pulmonary impairment, our findings require replication using a larger sample and across a range of dehydration severities.
In the absence of a "gold standard" method for assessment of small airway function (38), we used a combination of highly standardized functional tests (i.e., spirometry and whole-body plethysmography). Alongside these functional tests, imaging techniques such as high-resolution computed tomography or magnetic resonance imaging could have helped to quantify small airway dysfunctions (38). An ultrasonic nebulizer was used to ensure a high flow rate and even distribution of water vapor was delivered to the airways; however, the rate of delivery ranged from 1.0 to 1.8 mL·min À1 . These values are lower than expected (39) and may have limited our ability to restore lung volumes and capacities. Prior work has shown that isotonic saline delivered as small droplets (as generated by the nebulizer used in the current study) penetrates to the lung periphery (40). We are confident therefore that our solution reached the small airways. However, our nebulized isotonic solution may have become hypotonic upon delivery, as the solution was only isotonic when delivered in a euhydrated state. In individuals with asthma, both hyper-and hypo-osmotic nebulized saline can compromise pulmonary function (41). Thus we cannot exclude the possibility that our local rehydration strategy modified airway surface liquid ion concentration, ultimately preventing restoration of pulmonary function.
Finally, it is possible that oral fluid consumption might have led to psychological benefits and, thereby, contributed to improved effort during volitional respiratory maneuvers. Cognitive task performance and mood have indeed been shown to improve following rehydration with oral fluid in healthy men (42). However, PEF, an effort-dependent variable, did not significantly change at any time in our study. We are therefore confident that the effort produced by our participants remained consistent throughout the trials and that any psychological effects were likely minimal.

Clinical and Functional Significance
Our findings have potential significance to both healthy and clinical populations. In particular, endurance athletes are at increased risk for exercise-induced dehydration and commonly report respiratory symptoms (including breathlessness and cough) while exercising (43). Older adults, especially those with pulmonary disease, often experience exertional breathlessness (44) and are particularly prone and vulnerable to dehydration (3). Further work is now needed to determine the impact of dehydration-induced pulmonary alterations on susceptibility to respiratory symptoms and to understand the risk of pulmonary function deterioration in dehydrated states. Whether dehydration, by increasing gas trapping, triggers or exacerbates dynamic lung hyperinflation, thereby promoting breathlessness during physically demanding tasks, remains to be determined. It is also conceivable that the changes induced by dehydration impair airway surface liquid and thus mucociliary clearance mechanisms. Further work is needed to explore the role of systemic dehydration on mucociliary dysfunction and pulmonary exacerbations.

Conclusions
Mild systemic dehydration was associated with a reduction in pulmonary function, primarily localized to the small airways. These changes occurred in healthy adults after both acute exercise in the heat and prolonged periods of fluid deprivation. Oral fluid consumption, but not nebulized isotonic saline, quickly reversed these alterations in pulmonary function. Future work is needed to explore the implications of dehydration-induced changes in pulmonary function in older adults, especially in those with pulmonary disease. In the meantime, oral rehydration appears to be the most effective strategy for reversing dehydration-induced pulmonary impairments.

DISCLOSURES
No conflicts of interest, financial or otherwise, are declared by the authors.

ENDNOTE
At the request of the authors, readers are herein alerted to the fact that additional materials related to this manuscript may be found at https://doi.org/10.6084/m9.figshare.12191496.v1. These materials are not a part of the manuscript and have not