A systematic review and meta-analysis of pulmonary function in microgravity

Several different pulmonary variables were explored during this review. Gas exchange parameters including oxygen consumption (V̇O₂) and carbon dioxide production (V̇CO₂), respiration frequency, pulmonary flow rates like forced expiratory flow (FEF), mean inspiratory flow (MIF) and peak expiratory flow rate (PEFR), pulmonary diffusing capacity, minute ventilation (V̇E), breathing time like expiratory time, inspiratory time and breath-hold durations, pressure, compliance, resistivity, and respiratory work were analyzed.^17–44 The greatest number of variables measured were found in parabolic flights while the least were found in long-duration spaceflight missions (Table 3).

Thirteen studies demonstrated respiration frequency measurements in 1G and 0G (Fig. 3). The forest plot showed no change in respiration frequency between 1G and 0G across parabolic, short- and long-duration spaceflight, as demonstrated by both the common effect model and random effects model. However, a greater respiration frequency was observed in 0G than 1G during short-duration spaceflight breathing an inspired oxygen partial pressure (PiO₂) of 148 mmHg²⁰ and when standing at 1G (Prisk et al., 1995). A greater respiration frequency was also reported in 0G during short-duration spaceflight than in 1G when awake (Elliott et al., 2001), and during in-flight month 5 of a long-duration spaceflight (Baevsky et al., 2007).

Figure 3.

Forest plot for respiration frequency in 0G compared to 1G (i.e., a right-ward shift indicated a larger measure in 1G). Effect sizes are reported as Hedge’s g with 95% confidence intervals.

Eight studies reported minute ventilation measurements in 1G and 0G across all flight types (Fig. 4). During parabolic flight and short-duration spaceflight, there was no overall change in minute ventilation in 0G compared to 1G. However, greater minute ventilation at 1G during short-duration (+1.16 [0.08; 2.25]) and long-duration (+0.58 [0.17; 0.99]) spaceflight was reported when standing at 1G (Prisk et al., 1995; Prisk et al., 2006). Additionally, the hypoxic ventilatory response (initiated by reducing SaO₂ to 75%) produced a greater minute ventilation in 1G than 0G (+1.60 [0.08; 3.13]) (Prisk et al., 2000).

Figure 4.

Forest plot for minute ventilation in 0G compared to 1G (i.e., a right-ward shift indicated a larger measure in 1G). Effect sizes are reported as Hedge’s g with 95% confidence intervals.

Six studies measured V̇O₂ in 0G and 1G (Fig. 5). No differences were observed between short-duration spaceflight in 0G and during recumbent cycling in 1G, in which the subject is in a laid-back reclining position on the bicycle (Trappe et al., 2006), at various work rates (Stegemann et al., 1997), and at maximal exercise (Levine et al., 1996). However, V̇O₂ was greater at 0G than 1G in all measurements taken during parabolic flights (Limper et al., 2014). On the contrary, V̇O₂ was greater at 1G than 0G for the measurements taken during long-duration spaceflight (Prisk et al., 2006). As for V̇CO₂, three studies reported measurements in 1G and 0G, and there were no differences between 0G during short-duration spaceflight and 1G. However, a greater V̇CO₂ in 1G during long-duration spaceflight is observed in comparison to 0G (Prisk et al., 2006).

Figure 5.

Forest plot for parameters of pulmonary cellular respiration in 0G compared to 1G (i.e., a right-ward shift indicated a larger measure in 1G). Effect sizes are reported as Hedge’s g with 95% confidence intervals.

Three studies reported data on forced expiratory flow (FEF) in 0G and 1G and showed no difference between the two conditions for both parabolic and long-duration spaceflight (Fig. 6) (Prisk et al., 2006; Guy et al., 1991; Dutrieue et al., 2003). Four studies explored mean inspiratory flow (MIF) in 0G and 1G. Data obtained during parabolic flight and during light and deep sleep in short-duration spaceflight showed no changes between 0G and 1G (Dellaca et al., 2004; Estenne et al., 1992; Sa et al., 2009). However, another study’s data obtained during short-duration spaceflight showed a greater MIF in 1G than 0G (Prisk et al., 1995). On the other hand, a greater value in 0G than 1G was observed for MIF during REM sleep measurements (Sa et al., 2009). The common effects model generated shows an overall greater MIF in 0G than 1G. Three studies reported peak expiratory flow rate (PEFR) in 1G and 0G across all flight types and showed overall no differences between the two gravity conditions (Prisk et al., 2006; Guy et al., 1991; Elliott et al., 1996).

Figure 6.

Forest plot for parameters of pulmonary flow rates in 0G compared to 1G (i.e., a right-ward shift indicated a larger measure in 1G). Effect sizes are reported as Hedge’s g with 95% confidence intervals.

Six studies reported breathing time measurements in 1G and 0G (Fig. 7). A majority of measurements showed no difference between 0G during parabolic flight and short-duration spaceflight and 1G. However, one study showed greater breathing time in 1G than 0G during parabolic flight when a breath-hold was done at residual volume before test inspiration (Dutrieue et al., 2005). On the other hand, another study showed a greater breathing time in 0G in parabolic flight than 1G when inspiratory time is taken as a function of total breath time (% of preflight 1G measurement at the standing position) (Prisk et al., 1995).

Figure 7.

Forest plot for breathing time in 0G compared to 1G (i.e., a right-ward shift indicated a larger measure in 1G). Effect sizes are reported as Hedge’s g with 95% confidence intervals.

Three studies investigated pulmonary diffusing capacity in 0G and 1G across parabolic flight and short-duration spaceflight (Fig. 8). Measurements reported include diffusing capacity for carbon monoxide (CO) or nitric oxide (NO) and diffusing capacity per unit alveolar volume. Pulmonary diffusing capacity was greater at 0G than at 1G during all parabolic flight measurements (Vaida et al., 1997). However, membrane diffusing capacity, and normoxic diffusing capacity for CO and per unit alveolar volume outcome measures were greater in 0G than 1G in short-duration spaceflight (Prisk et al., 1993). Lung CO-diffusing capacity and CO-diffusing capacity per unit alveolar volume showed no differences between 0G in short-duration spaceflight and 1G (Verbanck et al., 1997). Additionally, single measurements were reported for components of pulmonary diffusing capacity including alveolar volume, pulmonary capillary blood volume, and pulmonary capillary blood flow. No differences were observed in alveolar volume and pulmonary capillary blood flow between 0G in parabolic flight and 1G, while pulmonary capillary blood volume was increased in 0G relative to 1G in short-duration spaceflight (Prisk et al., 1993).

Figure 8.

Forest plot for pulmonary diffusing capacity in 0G compared to 1G (i.e., a right-ward shift indicated a larger measure in 1G). Effect sizes are reported as Hedge’s g with 95% confidence intervals.

Fig. 9 depicts measurements of pressure, compliance, elastic work, exercise economy, exercise work, resistivity, and respiratory work. Five studies measured pressure parameters within the pulmonary system in 1G and 0G across all flight types. Greater pulmonary pressure and transdiaphragmatic pressure were observed in 0G in parabolic flight than 1G when measured at end of expiration (Bettinelli et al., 2002; Bettinelli et al., 2002) while greater end-expiratory gastric pressure was found in 1G (Edyvean et al., 1991). No differences were seen for measurements taken at 1G and during short-duration spaceflight, except that end-tidal PCO₂ when taken as % of preflight value at standing position was greater in 0G than in 1G (Prisk et al., 1995). However, another study noted no differences between 0G in short-duration spaceflight and 1G for that same measurement (Prisk et al., 2000). As for studies performed in long-duration spaceflight, maximum inspiratory pressure at functional residual capacity (FRC) is greater at 0G during both early and late in-flight timepoints than 1G. However, maximum inspiratory pressure at residual volume (RV) early and late in-flight and maximum expiratory pressure at total lung capacity (TLC) early in-flight are greater in 1G than 0G. There are no differences between 0G and 1G for maximum expiratory pressure at FRC during early and late in-flight, and maximum expiratory pressure at total lung capacity (TLC) late in-flight (Prisk et al., 2006).

Figure 9.

Forest plot for pulmonary pressure, compliance, and resistance parameters in 0G compared to 1G (i.e., a right-ward shift indicated a larger measure in 1G). Effect sizes are reported as Hedge’s g with 95% confidence intervals.

Three studies measured various compliance measurements including lung, chest wall, abdominal and dynamic compliance during parabolic flight (Fig. 9). No differences were seen between 1G and 0G in the various compliance measurements. However, both common effect model and random effects models show an increase in 0G relative to 1G. Moreover, pulmonary resistance measurements show no differences between 0G during parabolic flight and 1G (Dellaca et al., 2004; Edyvean et al., 1991).

Various single studies reported other pulmonary measurements essential to understanding changes in pulmonary function in space, but these did not fit the criteria for a meta-analysis. This includes resistivity (Hahn et al., 2013), respiratory work, elastic work (Dellaca et al., 2004), exercise work rate, and exercise economy (Levine et al., 1996). Greater resistivity is observed in 1G during parabolic flight than in 0G in measurements taken of the upper lung during spontaneous at normal tidal volume at end-expiration and end-inspiration, and during increased breathing at normal tidal volume at end-expiration. However, greater resistivity is seen in 0G in measurements taken of the lower lung during increased breathing at normal tidal volume at the end-expiration (Hahn et al., 2013).

Several work parameters were explored in 0G during parabolic flight including respiratory work and elastic work, while work rate at maximal exercise and economy outcome measures in 0G were obtained during short-duration spaceflight. No differences between 0G and 1G were observed in total inspiratory work, expiratory resistive work, and total resistive work. However, inspiratory resistive work was greater in 1G than in 0G. As for measurements of elastic work, no differences were seen between 1G and 0G in short-duration spaceflight for net elastic work performed by the inspiratory muscles and total elastic muscular inspiratory work. However, work necessary to overcome the elastic recoil of the lung and elastic work released by the chest wall were greater in 1G than in 0G (Dellaca et al., 2004). No differences were seen between 1G and 0G in short-duration spaceflight in the work rate measured at maximal exercise and economy at maximal V̇O₂ (Levine et al., 1996).

For most variables, the meta-analysis revealed no changes due to microgravity except for pressure parameters (Table 4). The majority of measurements had a high residual heterogeneity, suggesting a lot of variability between individual results. There was a significant change in oxygen consumption between the duration levels with greater levels in 0G during transient microgravity exposure and lower levels during sustained exposure. There was an effect of microgravity on pressure and pulmonary diffusing capacity.

This systematic review gathers all the available published parabolic flight and spaceflight studies with functional pulmonary outcomes to provide an overview of the changes in pulmonary system function in space relative to normal gravity on Earth. There were 28 papers that covered 0-1G exposure times from 18 seconds to a maximum of 180 days. The average duration for each category in the review was 21 ± 2 s (parabolic; n=115 individuals), 14 ± 4 d (short-duration; n=67 individuals), and 174 ± 8 d (long-duration; n=29 individuals). In a previous study we focused on anatomical changes with low gravity exposure (Ghani et al., 2023). In this study, the focus was on functional lung measures.

The lung comprises both fluid and gas dynamics, therefore, changes in gravity could alter pulmonary function, particularly the ventilation-perfusion ratio, leading to impaired physical performance and potential health issues. However, this systematic review found that all the measures demonstrated either no change between 1G and 0G or the changes were small in magnitude and of uncertain clinical relevance. Of the small but statistically significant changes these included: reductions in CO₂ production and elastic work, and increases in pulmonary diffusing capacity, pulmonary capillary blood volume, and pulmonary compliance in 0G compared to 1G.

The reduction in elastic work and improved pulmonary compliance result in easier breathing, as the respiratory muscles exert less effort (Fig. 9) (Cross et al., 2012). However, in individuals with pulmonary disease the implications of this can vary. Restrictive lung diseases (e.g., fibrosis, acute respiratory distress) are characterized by an increase in elastic work and a reduction in compliance (Mannino et al., 2003). Hypothetically, the effects of microgravity might help alleviate some of these issues by offsetting the underlying mechanisms. Conversely, in obstructive pulmonary diseases (e.g., asthma) where compliance is already high and elastic work low, microgravity might exacerbate air trapping (Devine, 2008). This can lead to hyperinflation of the lungs, increase the work of breathing, and potentially impair gas exchange by causing ventilation-perfusion mismatching (Rossi et al., 2015). While persons with overt restrictive or obstructive pulmonary disease are not going to be selected for current long duration space missions, prolonged exposure to lunar or Martian dust could cause airway inflammation that leads to obstructive pulmonary disease (Hnizdo and Vallyathan, 2003). This is of course a hypothetical risk and it is important to note that these measures have not been recorded during spaceflight and are currently limited to parabolic flight. As parabolic flight involves pre- and post-loading of >1G loads, further investigations are needed in sustained microgravity environments to observe the effects that this could have, particularly regarding people with obstructive pulmonary diseases.

An observable increase in pulmonary diffusing capacity in microgravity was present in all three studies that assessed it (Fig. 8). This would improve pulmonary gas exchange (Roughton and Forster, 1957). One study showed diffusing capacity to be supported by increased pulmonary capillary blood (Fig. 7) (Prisk et al., 1993). These factors act to improve oxygen uptake and CO₂ removal, thus helping ventilation-perfusion matching across the lungs. This benefit could compensate for pulmonary impairments, allowing gas exchange with less respiratory effort. However, no long-duration spaceflight studies have looked at this, so it is unclear whether this increase in pulmonary diffusing capacity is maintained or reduced over time, with the longest study evaluated here being 10 d (Verbanck et al., 1997).

The observed random effects decrease in CO₂ production was only significant in one of the four studies that assessed this (Fig 5) (Prisk et al., 2006). This could be consistent with a lower metabolic rate, supported by trends towards lower minute ventilation and oxygen consumption (Fig. 4; Fig. 5), with oxygen consumption lower with longer durations spent in space (Table 4). Indeed, the longer-duration study of Prisk et al. (2006) that evaluated these variables suggested the underlying reason could arise from astronauts doing relatively more activity in shorter-duration flights. It was also postulated, that while unlikely, a reduction in body mass over time could reduce metabolic rate, although this did not align with other measured parameters (Prisk et al., 2006).

Overall, the study highlights the adaptability of the lung, including in the longest duration 0G exposure studies that have reported minimal changes compared to pre-flight. The changes that were observed mostly lead to improved pulmonary function, by improving gas exchange and potentially providing a benefit for those with any underlying restrictive lung component. Obstructive lung disease could meet with some challenges, but whether these are ever likely to be present in short-and long-duration 0G exposures remains to be determined. Likewise, it is unknown whether the negative effects are offset by concurrent improvements in gas exchange.

The findings of this review as well as the many foundational studies done during NASA Spacelab missions and ESA parabolic flights showed that lung function remains surprisingly robust in space, with no major impairments in gas exchange or ventilation (Prisk, 2019; Prisk et al., 2008; West et al., 1997; Puchkova et al., 2025). A recent study provided a detailed investigation how simulated lunar gravity via head-up bed rest and microgravity via head-down bed rest minimally affected lung function in healthy human subjects. There was clinically insignificant decrease in key respiratory parameters like FVC and FEV₁ within the first hours of exposure and the changes were not pathological and remained within normal physiological ranges. Observed changes normalized over time, indicating the respiratory system’s ability to adapt. Importantly, no long-term impairments were observed, and physical training during the simulation helped accelerate recovery of lung function (Puchkova et al., 2025). These findings alongside the results of our review suggest that human lung function remains resilient in space-like environments, supporting the feasibility of extended missions with appropriate countermeasures like exercise.

Spaceflight reviews are inherently limited by the number of studies conducted, and particularly by the availability of repeated analyses of specific similar outcome variables. This is a consequence of both funding and time constraints. As a result, studies are rarely repeated, leading to inherent large heterogeneity in the study designs. This showed up in the current meta-analyses due to the impact of subtle differences between how variables were measured as well as the individual responses (Table 4). Additionally, the limited number of long-duration studies are the only evidence available that would be most indicative for long-duration missions to the Moon and Mars. More data is required from truly long-duration missions.

There is a lack of recent investigations to re-examine the longstanding observation that microgravity produces little to no measurable effect on the lungs, meaning this conclusion remains based on limited and outdated evidence. Also, the most valuable information for preparation of manned missions to the Moon and Mars comes from long-duration missions, and there have been no recent investigations thoroughly examining pulmonary functional outcomes in long-duration human space travel. The age of the available literature highlights the urgent need for renewed research attention to pulmonary physiology in spaceflight. The lack of recent studies is a field-wide limitation.

Consistent with current selection criteria for astronauts or parabolic flights, all subjects in the included studies were fit and healthy people. Such subjects were either crew mission members, who have undergone rigorous medical screening and tests, or terrestrial people with good health. Both sets were within a relatively tight age range and so do not reflect the general human population. Hence, future versions of these studies that can include data of civilian astronauts who are not subjected to rigorous medical screening prior to spaceflight may exacerbate the individual variability and lead to more substantial effects. Also, people with lung diseases like asthma or chronic obstructive pulmonary disease may show different outcomes in microgravity, and hence, information regarding effects of microgravity towards these conditions could be potentially clinically significant as more data is collected in the future, given the increase in space tourism activities.

The studies included in this systematic review present a major gap in male and female participation, making the roles of sex in pulmonary function in space not well understood. 43% of the included studies included no female participants, while females only made up 32% in those that did. Only one study recruited an equal number of male and female participants (Limper et al., 2014). Overall, there were n=162 males and n=49 females included in the analysis of this systematic review. Sex and gender differences in pulmonary health studies should be considered a future priority, especially since females typically have smaller lungs and airways than males, resulting in lower total alveoli and alveoli surface area, which can impact pulmonary function (Dominelli and Molgat-Seon, 2022). Further, during exercise in 1G it is known that females have lower diffusion capacity, membrane diffusing capacity, and pulmonary capillary volume than males, limiting exercise tolerance (Bouwsema et al., 2017; Guzman and Summers, 1973). Females are also more likely to experience pulmonary hypertension, which could be exacerbated by the cephalic fluid shift seen in microgravity (Hughes-Fulford et al., 2024). Conversely, males are terrestrially at higher risk of developing pulmonary fibrosis and obstructive sleep apnea. However, the exclusion of women from many studies makes comparisons based on sex challenging, masking the sex-specific effects of spaceflight on pulmonary function. Such information could be important in informing future long-duration expeditions to the moon and Mars.

Human spaceflight research is limited by the financial and logistical costs associated with launching and conducting experiments in true microgravity, and technical constraints to research methods and data collection tools. Parabolic flights are composed of an average of 30 parabolas per flight, which includes periods of both microgravity and hypergravity that may act as a confounding variable (Golding et al., 2017). Therefore, the transient microgravity in parabolic flights may not be indicative of observations in sustained, stable microgravity. In addition to the inherent challenges of conducting human spaceflight experiments and recruiting participants, this systematic review is limited by the data available at different time-points in space (Table 3). These gaps make it difficult to confirm whether the reported changes are unique to each study or scientific fact. This is especially the case for long-duration spaceflight where almost two-thirds of the pulmonary functional variables included in this review were not measured during long-duration missions, which only accounted for three papers (Baevsky et al., 2007; Prisk et al., 2006; Verheyden et al., 2010).

This review does not provide information about the functional changes that happen to the pulmonary system upon return to Earth, as it focuses on comparisons solely between 0G in-flight and 1G at pre-flight and not post-flight. Specifically, exploring possible changes to pulmonary function during early and late post-flight time points could reveal a great deal about the effects of space on the lungs and provide further insights in recovery durations, complementing the findings of this review. This is particularly relevant to assessing the overall risk of spaceflight on the pulmonary system; according to the NASA Human System Risk Assessment Process, a key factor to determining risk is the time to return to baseline (reconditioning). While microgravity can cause changes to lung function, several studies noted that these changes were temporary and reversible upon return to Earth’s gravity, even after 6 months of exposure to microgravity (Prisk, 2019; Prisk et al., 2008; Prisk et al., 2006).^2,4,38 Therefore, since there were either no changes or the changes were reversible, the changes were unlikely to be clinically significant. However, if a person has a lung condition that alter mechanics, such as asthma, COPD, or idiopathic pulmonary fibrosis (IPF), then they are likely to have a different situation.