Sex Differences in the Pulmonary System Influence the Integrative Response to Exercise

Abstract

Healthy women have proportionally smaller lungs and airways compared with height-matched men. These anatomical sex-based differences result in greater mechanical ventilatory constraints and may influence the integrative response to exercise. Our review will examine this hypothesis in healthy humans in the context of dynamic whole-body exercise.

In Brief

Sex differences in the morphology of the pulmonary system influence the integrative response to dynamic whole-body exercise.

Key Points

• Sex-based differences in pulmonary system anatomy result in a greater work of breathing and respiratory muscle oxygen uptake in women during exercise.
• Similar to men, women can develop arterial hypoxemia during exercise and altering the work of breathing in women influences both blood flow and fatigue to the respiratory and locomotor muscles.
• Expiratory flow limitation may be more common in highly trained women, but there are no sex differences in subjects of average fitness.
• The functional consequences of sex differences in the structure of the pulmonary system seem to be maintained over the course of the healthy aging process.
• The effects of size, sex, hormones, and training status need to be considered in a context-dependent manner when designing and interpreting studies regarding the effect of sex differences in respiratory system morphology on the integrative response to exercise.

INTRODUCTION

The pulmonary system can be considered relatively “overbuilt” in terms of its capacity to meet the ventilatory and metabolic demands of dynamic whole-body exercise. This notion holds true across a range of exercise intensities in healthy humans, whereby arterial blood gas homeostasis is preserved and the energetic cost of breathing is not excessive. However, it is now clear that pulmonary system limitations, such as meeting the maximal capacity to generate expiratory flow (e.g., expiratory flow limitation (EFL)) and high work of breathing (W_b), can be present in otherwise healthy individuals and influence the integrative response to exercise (¹). Unfortunately, the majority of early studies investigating the influence of the pulmonary system limitations during exercise were performed exclusively or predominantly in men, and this sex-based bias in the literature has persisted (²). However, a growing body of literature points to important sex differences in the anatomy and physiology of several organ systems. Notably, recent work has shown that women have smaller lungs and airways than do men. In our brief review, we present the rationale for our overarching hypothesis that sex differences in pulmonary system morphology influence the integrative response to whole-body exercise (Fig. 1). We place emphasis on investigations that have studied healthy humans across the lifespan and limit our interpretation to dynamic whole-body exercise (e.g., cycling or running) rather than static, isolated, or small muscle mass exercise (e.g., handgrip). Other modalities of whole-body exercise (e.g., swimming) will not be discussed specifically because of the limited data owing to technical difficulties in many common measurements.

Figure 1: Schematic of our working hypothesis. The (+) and (−) symbols indicate that one item has a positive and negative effect (respectively) on the subsequent item.

WHAT WE KNOW

Sex Differences in Pulmonary System Morphology

The pulmonary system is composed of the lungs, airways, rib cage, and respiratory muscles (Fig. 1, see top panel, Morphological differences). Although the gross anatomy of the pulmonary system does not differ based on sex, several morphological differences have been identified. First, women have smaller lungs than men, even when matched for sitting height (a surrogate for thoracic volume) (³). Second, women have proportionally smaller airways than men, which has been observed at several points along the airway tree. Women have smaller cross-sectional areas of the trachea (⁴) and large conducting airways (generations 1–5; i.e., bronchi) than men matched for lung size (⁴) or height (⁵). In addition to differences in size, the shape of the lungs and rib cage differs based on sex. In men, the thorax is relatively shorter and broader, particularly at the bases, than in women (⁶). Moreover, women have “prismatic” lung geometry, with the bases and apices of the lungs being similar in size, whereas men have “pyramidal” lung geometry, with the lung bases being larger than the apices (⁶). Although the differences in the morphology of pulmonary system have little to no influence on breathing mechanics or blood gas homeostasis at rest, they have significant effects on the integrative response to exercise (⁷).

Work of Breathing

The W_b is the mechanical work necessary to achieve a given minute ventilation (V˙_E) with a given breathing pattern (i.e., tidal volume, breathing frequency, and lung volume). The W_b is commonly measured by determining the area within an esophageal pressure-volume curve, with the former being measured using a balloon catheter placed in the esophagus to estimate pleural pressure (⁸). The main determinants of the W_b are related to viscoelastic and resistive forces, with the latter being heavily influenced by airway geometry. Given the aforementioned sex differences in the morphology of the pulmonary system, we would expect the mechanical W_b during exercise to be greater in women. In one of the first studies to compare the W_b in men and women, Wanke et al. (⁹) demonstrated that for a given V˙_E, the mechanical work of the respiratory muscles was significantly higher in women than in men. The authors concluded that the greater W_b was due to differences in absolute body and lung size, as well as differences in breathing pattern. In a series of subsequent studies, we have observed that for a given V˙_E, the total W_b is higher in women than men in those who are highly trained (i.e., women’s maximum oxygen uptake (V˙O_2max) 60 ± 5 mL·kg⁻¹·min⁻¹) (¹⁰), of average fitness (i.e., V˙O_2max 48 ± 6 mL·kg⁻¹·min⁻¹) (¹¹), and over a range of ages (i.e., 20–80 yr) (^11,12) (Fig. 2). The fact that W_b is higher in women of varying ages and fitness levels is important when considering the generalizability of this finding and its relevance to health. Typically, the difference in W_b between the sexes only becomes significant at V˙_E of 50–70 L·min⁻¹ (¹³). A V˙_E of 50–70 L·min⁻¹ in women most likely corresponds to a condition where flow in the large conducting airways (i.e., generations 0–5) is predominantly turbulent, rather than laminar. High levels of turbulent flow paired with airway branching would result in a rapid rise in airway resistance and likely explain the characteristic exponential rise in W_b as V˙_E increases (Fig. 2). The absolute V˙_E where the deflection point in the W_b-V˙_E relationship occurs varies between individuals of the same sex, which corresponds with the considerable variability in airway size (⁵). When the W_b is partitioned into the constituent components (i.e., resistive and elastic), it is only the resistive work that is greater in women and this difference occurs at V˙_E > 60 L·min⁻¹ (^11,14), which lends support to the theory that smaller airways result in a sex difference in respiratory work. That the elastic component of the W_b is similar between the sexes suggests that there are no sex-based differences in the structures of the lungs or in their inherent elasticity and provides physiological evidence for the argument that smaller airways in women are responsible for a higher W_b relative to men.

Figure 2: The relationship between work of breathing and absolute minute ventilation (VE) during incremental cycle exercise. Each curve was modeled from the equation found in the corresponding manuscript (¹²) and has been extrapolated to the average peak minute ventilation within each group. Data for older men are displayed as thick gray lines, whereas data for older women are displayed as thick black lines. Data for younger men are displayed as thin gray lines, whereas data for older women are displayed as thin black lines. *P < 0.05, main effect of age, comparisons made between all older and all younger subjects, regardless of sex. †P < 0.05, comparisons made between all men and all women, regardless of age. No significant interaction effect was observed. [Adapted from (¹²). Copyright © 2018 the American Physiological Society. Used with permission.]

Oxygen Cost of Breathing

Although the total mass of the respiratory musculature is relatively small compared with that of the locomotor muscles, respiratory muscles also require appropriate oxygen delivery to sustain contractions and generate an appropriate V˙_E. Like other skeletal muscles, contracting respiratory muscles demonstrate a linear relationship between work and oxygen uptake (V˙O₂). Therefore, if women have a greater W_b for a given V˙_E compared with men, it is expected that oxygen uptake of the respiratory muscles (V˙O_2RM) would be higher. Although the reported absolute values were excessive (i.e., >1–2 L·min⁻¹) due to methodological limitations, a higher V˙O_2RM in women than in men was demonstrated as early as 1962 (¹⁵). The main technical difficultly with resolving relatively small differences in V˙O_2RM is that matching and maintaining respiratory work is difficult to achieve owing to the need to precisely match pressure, volume, and flow in real time (¹⁶). Thus, an accurate assessment of V˙O_2RM involves having subjects mimic their exercise breathing patterns under resting conditions while simultaneously receiving real-time feedback of pressure, flow, and volume (¹⁶). In a classic study aimed at determining V˙O_2RM in a healthy cohort of individuals by using the mimicking technique, the lone female subject had the highest W_b and subsequently the greatest absolute and relative V˙O_2RM, despite breathing at a considerably lower absolute V˙_E than men (¹⁷). The female subject’s absolute V˙O_2RM at 70%–100% of V˙O_2max was 2.4–3.9 mL per L of V˙_E, whereas the men’s average was 1.8–2.9 mL per L of V˙_E at the same fractions of V˙O_2max. The relative V˙O_2RM at V˙O_2max also was higher in the woman compared with the men (15 vs 10 ± 2% of total V˙O₂, respectively). To explore the possibility of any sex differences in V˙O_2RM, we used a similar approach involving multiple trials over a range of V˙_E and found that on average, V˙O_2RM was significantly higher in women at a V˙_E > 55 L·min⁻¹ (¹⁴). Others have reported no difference in the V˙O_2RM between the sexes (¹⁸); however, their study investigated V˙_E of 40–60 L·min⁻¹ and at that level of V˙_E, we also did not observe differences between the sexes. The greater V˙O_2RM in our study was due to women having significantly greater W_b at a given V˙_E, but also significantly lower efficiency of the respiratory muscles throughout a range of ventilations (¹⁴). In this case, efficiency refers to mechanical efficiency, defined as the ratio between W_b and V˙O_2RM. The lower respiratory muscle efficiency in women is most likely due to anatomical differences, which result in altered biomechanics of the respiratory muscles, sex differences in respiratory muscle recruitment patterns, the presence of flow limitation, or a combination thereof; however, this theory has yet to be tested experimentally.

Perhaps more important than the absolute V˙O_2RM (i.e., mL·min⁻¹) is the fraction of whole-body V˙O₂ dedicated to the respiratory muscles at a given relative intensity. The importance of the relative oxygen cost of breathing is demonstrated via the Fick equation, whereby in the absence of a change in oxygen extraction at the level of the muscle, V˙O₂ is directly proportional to blood flow. Therefore, the fraction of whole-body V˙O₂ devoted to the respiratory musculature likely reflects a similar percentage of cardiac output. Accordingly, we demonstrated that at submaximal and maximal exercise intensities, the relative oxygen cost of exercise hyperpnea was significantly greater in women (¹⁴) (Fig. 3). At maximal exercise, we found that men dedicated ~9% of whole-body V˙O₂ to their respiratory muscles (a value in close agreement with others (¹⁷)), whereas women dedicated ~14% (¹⁴). It is important to note that there is significant overlap between sexes, with many men having achieving values above the average of the women (Fig. 3), which likely relates to interindividual differences in W_b (most likely due to airway size differences) and the presence of EFL, which in and of itself increases V˙O_2RM (¹⁴). That women dedicate ~5% more of their whole-body V˙O₂ to their respiratory muscle is significant when considering the role respiratory muscle energetics play in the development of locomotor fatigue and blood flow distribution (see below).

Figure 3: Box and whisker plot, showing individual subject data and group mean ± SE for oxygen uptake of the respiratory muscles (V˙O_2RM) as a percentage of whole-body oxygen uptake at maximal exercise in women and men (%V˙O_2max). Squares in the box and whisker plot represent 5th and 95th percentiles and the horizontal line is the median. *Significantly greater compared with men (P < 0.05). (Reprinted from (¹⁴). Copyright © 2015 John Wiley and Sons. Used with permission.)

WHAT WE THINK WE KNOW

Expiratory Flow Limitation

Given that women have smaller lungs and airways than men, their maximal expiratory flows at a given fraction of vital capacity are lower than height-matched men (³). It follows that during exercise, women may be more likely to reach their maximum capacity to generate expired flow (¹⁹), a concept known as EFL. Indeed, data from women of a wide range of fitness levels suggest that women, particularly those with high cardiorespiratory fitness, are predisposed to EFL during exercise (²⁰); however, this study did not explicitly compare the sexes. When direct comparisons between highly-trained men and women are made, women are more likely to experience EFL at peak exercise than men (¹⁰). Nevertheless, it is unclear whether women of average fitness levels also have an increased propensity toward EFL during exercise relative to men. Recent work in larger samples (n = 20–146) of healthy women and men of varying fitness levels indicates that the occurrence of EFL is similar between the sexes (i.e., ~55%–60%) (^11,12). However, subgroup analysis from one of the aforementioned studies revealed that in subjects with a V˙O_2max of >125% predicted, 80% of the women developed EFL compared with 50% of the men (¹¹). Thus, it seems that in healthy individuals, sex differences in pulmonary system morphology influence the propensity toward EFL during exercise only when ventilatory demands are high enough to reach the maximal ventilatory capacity of the respiratory system, as is the case in individuals with high cardiorespiratory fitness. However, in those of average fitness, the ventilatory demands of exercise are relatively low and therefore sex differences in pulmonary system morphology do not influence the likelihood on developing EFL during exercise (²⁰).

Exercise-Induced Arterial Hypoxemia

In some men with high aerobic capacities, arterial blood gas homeostasis is not necessarily maintained during exercise; a phenomenon termed exercise-induced arterial hypoxemia (EIAH) (²¹). Although our understanding of the causes of EIAH in men is incomplete (²¹), even less is known about EIAH in women owing to a paucity of data (²²). There are studies in which temperature-corrected arterial blood gases were obtained in women performing exercise, but the associated findings and conclusions are conflicting. For example, some have argued that because of smaller lungs, airways, and less alveoli, women are more prone to EIAH than men (^23,24), whereas others have shown no difference in the occurrence and severity of EIAH between the sexes (^22,25). To date, only one study has directly compared arterial blood gas values in men and women performing strenuous exercise and no differences were noted (²²). Here, men and women had comparable ventilation-perfusion ratios (based on the multiple inert gas exchange technique) during exercise (²²). An important consideration when interpreting the findings of this study is that the subjects were intentionally matched for lung size. Indeed, others have shown that sex differences in pulmonary diffusion also are eliminated when matched for lung size (²⁶). The problem with matching for lung size is that, on a population level, women have smaller lungs and by matching for lung size, one would be eliminating a known sex effect. Therefore, although these studies provide insight into sex differences in pulmonary gas exchange, it is difficult to draw definitive conclusions regarding the effect of sex on the prevalence of EIAH. Another potential issue when assessing EIAH is the specific mode of exercise used (e.g., running vs cycling), which has been shown to influence blood gases (^25,27). Interestingly, studies supporting sex differences in EIAH have used a treadmill (^23,24), whereas those who show no difference between the sexes used cycle ergometry (^22,25). The potential sex difference in occurrence and severity of EIAH could be the result of differences in exercise modality between studies, but this theory awaits rigorous testing.

It is generally agreed upon that untrained men do not demonstrate significant levels of hypoxemia during exercise (²¹). However, there are now multiple reports of relatively untrained women (i.e., within 15% of predicted V˙O_2max) demonstrating EIAH (^23,24,28). We have proposed that mechanical ventilatory constraints (such as EFL) in some untrained women lead to EIAH, which often occurs at submaximal workloads (²⁴), although it is important to stress that not all subjects who demonstrate EFL develop EIAH and vice versa. Nonetheless, in those untrained women with EIAH, the presence of EFL results in an inadequate increase in alveolar ventilation to offset the typical widening of the alveolar-arterial oxygen gradient. To determine the importance of EFL on gas exchange in these subjects, we eliminated EFL experimentally with a normoxic helium-oxygen inspirate (i.e., 21% O₂: 79% helium). Replacing nitrogen with helium as the backing gas reduces airflow resistance by promoting laminar flow and increasing maximal respiratory flows (²⁴). With EFL during exercise eliminated, subjects that had EIAH and EFL increased their V˙_E, which resulted in a higher arterial partial pressure of oxygen and oxyhemoglobin saturation (²⁴). We emphasize that breathing helium-oxygen gas only partially attenuated EIAH, indicating that EFL (or mechanical ventilatory constraint) is not the sole or dominant cause of EIAH. Overall, we surmise that during exercise, the presence of EFL in some untrained women can hinder adequate compensatory hyperventilation and, thus, exacerbate the typical gas exchange impairment and lead to the development of EIAH.

Locomotor and Diaphragm Fatigue

Muscular fatigue, defined as a condition during which there is a temporary decrease in the capacity of muscle force generation that is reversible by rest, is known to occur in locomotor muscles after intense exercise. Manipulating the W_b is known to influence locomotor and diaphragm fatigue during high-intensity exercise in men (²⁹). Given that women have a greater W_b for a given absolute V˙_E, it could be hypothesized that they would be especially susceptible to respiratory fatigue influenced by W_b. Two caveats to this theory are that women seem fatigue resistant in many skeletal muscles, and most of the sex differences in fatigue are not present during dynamic exercise (³⁰). When we evaluated quadriceps fatigue after exercise, we found no difference between the sexes in terms of the severity or duration of fatigue (³¹). Both sexes had an ~28% decline in quadriceps twitch force in response to supramaximal magnetic stimulation of the femoral nerve, and recovery was incomplete 60 min postexercise (³¹). On separate days, subjects performed two trials that were iso-time and iso-work compared with the control day where we (i) eliminated any EIAH via a mildly hyperoxic inspirate and (ii) reduced the W_b with a mechanical ventilator. When the exercise-induced fall in SaO₂ (oxyhemoglobin saturation) below resting levels was prevented, quadriceps fatigue was reduced similarly in both sexes and the degree of fatigue attenuation was linearly related to the nadir in SaO₂ during the control trial (³¹). In other words, regardless of sex, those who had the greatest amount desaturation had the most improvement in leg fatigue while breathing hyperoxic gas during exercise. Overall, the impact of improved arterial oxygenation (i.e., restoration to “normal” levels) on locomotor fatigue itself does not seem to differ based on sex, but on the severity of EIAH. When the W_b was minimized with a mechanical ventilator, both sexes had a similar improvement in quadriceps fatigue (³¹). However, quadriceps fatigue attenuation corresponded to a different degree of respiratory muscle unloading between the sexes, with less of a decrease in W_b (25% in women vs 40% in men) being needed in women to obtain the same reduction in fatigue as noted in men. The explanation for this discrepancy is the difference between the sexes in relative V˙O_2RM (Fig. 3). Based on our estimates of the relative cost of V˙O_2RM at maximum exercise (¹⁴), a 25% and 40% reduction in the W_b for women and men would result in 3.5% and 3.6% reduction in total V˙O₂ no longer associated with the respiratory muscles. If the change in V˙O_2RM as a percentage of total V˙O₂ is similar between the sexes, it follows that a comparable fraction of cardiac output will be redirected toward the locomotor muscles. A comparable increase in blood flow to the locomotor muscles would explain why the extent of quadriceps fatigue attenuation was not different between the sexes.

It is now well established that healthy male subjects of varying aerobic capacity performing sustained exercise (i.e., >8–10 min) at intensities greater than 80%–85% of V˙O_2max develop diaphragm fatigue, as assessed by supramaximal stimulation of the phrenic nerves (³²). There is a strong relationship between diaphragm force output during exercise and the magnitude of fatigue implying an association with the amount of generated pressure and total work performed by the diaphragm (³³). For example, when diaphragm work is reduced (via mechanical ventilation), exercise-induced diaphragm fatigue is prevented (²⁹) and diaphragm fatigue is known to be reversed by increasing phrenic artery blood flow (³⁴). Importantly, a high level of respiratory muscle force output during exercise is not solely responsible for diaphragmatic fatigue. Babcock et al. (³⁵) found that diaphragmatic fatigue did not occur when subjects replicated (at rest) the components of breathing as achieved during exercise at 95% V˙O_2max (³⁵). They interpreted this to mean that the effects of locomotor muscle activity, including a competition for blood flow distribution and the presence of significant acidosis, in tandem with high levels of diaphragmatic work lead to exercise-induced diaphragm fatigue.

As previously mentioned, there seems to be a sex difference in skeletal muscle fatigue (³⁰), where women often are found to be more fatigue resistant than men. Guenette et al. (³⁶) sought to determine if the anatomical and functional sex differences in the pulmonary system would make the female diaphragm more prone to fatigue during exercise. In a group of highly-trained subjects performing high-intensity exercise (i.e., 90% peak power) to exhaustion, they found that the magnitude of diaphragm fatigue was significantly greater in men (³⁶). That women had less severe diaphragm fatigue is consistent with observations in limb muscles, but an explanation is not intuitive given the higher propensity for EFL and a higher W_b seen in highly-trained women relative to their male counterparts. Before attempting to address the potential causes for a sex difference in diaphragm fatigue, it is important to emphasize the lack of data in the literature. Greater numbers of female subjects (across a range of fitness levels and exercise conditions) should be tested, using objective measures of diaphragm fatigue (e.g., phrenic nerve stimulation), before claims that sex differences exist on a population level can be made. With this in mind, several unconfirmed possibilities could be used to explain the observation that women develop less severe diaphragm fatigue. First, Guenette et al. (³⁶) found that the absolute load on the diaphragm was lower in women. Presumably, the lower absolute mechanical load results in less mechanical compression of the local vasculature and less intramuscular occlusion of blood flow, which are known to contribute to diaphragmatic fatigue (³⁷). Second, there are sex differences in locomotor muscle fiber type, composition, and substrate utilization that influence fatigue (³⁰). To our knowledge, this concept has not been investigated with respect to the diaphragm and remains an interesting hypothesis. Third, during time-to-exhaustion exercise, men seem to use their diaphragm to a greater degree, whereas women increasingly recruit “extra-diaphragmatic” inspiratory muscles (^36,38). It is possible that the ventilatory “strategy” used by women serves to minimize diaphragmatic work and, thus, minimize fatigue (see section Respiratory Muscle Activation).

Work of Breathing Influencing Blood Flow

When the W_b is experimentally lowered in trained men during high-intensity exercise, blood flow to the contracting leg muscles increases (³⁹). The proposed mechanism behind the blood flow redistribution is a sympathetically mediated reflex originating from the respiratory musculature (¹). We have recently provided direct support for this concept by showing that when W_b is lowered during exercise, sympathetic vasomotor outflow is reduced and, importantly, we observed the effect in both sexes (⁴⁰). It would then stand to reason that women also should redistribute blood flow when W_b is altered. We found that manipulating the W_b during intense exercise results in a redistribution of skeletal muscle blood flow in both men and women (⁴¹). In these subjects, when the W_b was increased or decreased, respiratory muscle blood flow also increased and decreased, respectively. Simultaneously, locomotor blood flow also changed, but in an opposite manner, whereby increasing W_b leads to a decrease in quadriceps blood flow (⁴¹). Given that we have shown that the respiratory muscles command a greater fraction of total V˙O₂ in women and that less of a change in W_b is needed to manipulate active locomotor blood flow (³¹), one could speculate that any blood flow redistribution would be exaggerated in women. However, we caution that determining if women demonstrate a heightened blood flow redistribution requires the determination of bulk locomotor blood flow (i.e., thermodilution) coupled with considerable respiratory muscle unloading (i.e., via proportional assist ventilator).

Exercise Performance

The influence of W_b and diaphragm fatigue on exercise performance has been addressed in several ways in male subjects. First, using a mechanical ventilator, Harms et al. (⁴²) noted a 14% increase in exercise tolerance with inspiratory muscle unloading and a 15% decrease with inspiratory muscle loading compared with control conditions. Second, studies have isolated the independent effects of diaphragm fatigue on exercise tolerance by inducing fatigue at rest before a subsequent exercise bout. Fatiguing the diaphragm resulted in significant reductions (14%–23%) in exercise tolerance (⁴³). Third, by breathing a low-density gas mixture such as heliox it is possible to unload the respiratory muscles during exercise (²⁴). Unfortunately, few studies have determined the contribution of a high W_b and diaphragm fatigue to whole-body exercise performance in women. In one study, we used a helium-oxygen mixture to reduce the W_b and minimize EFL to evaluate 5-km cycling time trial performance in men and women (⁴⁴). We found that the associated performance gains were small, but significant, and similar between men and women (⁴⁴). It should be emphasized that our reduction in the W_b with helium-oxygen was substantially less than what can be achieved with a proportional assist ventilator (^31,39) and diaphragm fatigue was not assessed. In our view, questions regarding diaphragm fatigue, exercise performance, and potential sex-based differences remain, for the most part, unresolved.

Respiratory Muscle Activation

Exercise-induced increases in V˙_E place a significant burden on the respiratory muscles. Because women have a higher W_b and V˙O_2RM than men for a given V˙_E above ~55 L·min⁻¹ during exercise, it is possible that women use different patterns of respiratory muscle activation than men to share the “load” across a larger volume of muscle. Previous work demonstrated that women have a lower diaphragmatic contribution to overall inspiratory pressure generation during exhaustive constant-load exercise at 85% of V˙O_2max than do men (³⁶), leading to the hypothesis that women rely on “extra-diaphragmatic” inspiratory muscles during exercise to greater extent than do men to avoid or minimize the onset diaphragm fatigue. By using concurrent assessments of obligatory (i.e., diaphragm and scalene) and accessory (i.e., sternocleidomastoid) inspiratory muscle activation during exercise, a recent study noted that women rely on scalene, sternocleidomastoid muscles to greater extent than do men during exercise to exhaustion at 85% of V˙O_2max (³⁸). We recently expanded on this finding by quantifying inspiratory muscle activation in combination with measures of inspiratory pressure generation during incremental cycle exercise (⁴⁵). We found that at across a range of absolute (i.e., in L·min⁻¹) and relative (i.e., as a % of maximal V˙_E) V˙_E, women rely on scalene and sternocleidomastoid muscles to a greater extent than do men. We also noted that extra-diaphragmatic inspiratory muscle activation was inversely associated with the diaphragmatic contribution to overall inspiratory pressure generation, implying that increased scalene and sternocleidomastoid activation likely serves to assist the diaphragm (⁴⁵). We caution, however, that our findings are observational in nature and stress that differential muscle activation between the sexes is unlikely to be a conscious “choice” and could be the result of morphological differences or other untested factors. The functional significance of sex differences in respiratory muscle activation is currently unknown, but could theoretically influence overall respiratory muscle efficiency, the perception of dyspnea, fatigue of these extra diaphragmatic muscles, and the response to respiratory muscle training.

Aging

Healthy aging involves significant changes to the structures of the pulmonary system that lead to a progressive decline in pulmonary function (^46,47). Specifically, older individuals have a reduced capacity to generate flow, a smaller vital capacity, decreased lung diffusion capacity, and weakened respiratory muscles when compared with their younger counterparts (⁴⁶). The age-related reduction in pulmonary function has a significant detrimental impact on breathing mechanics during exercise, as evidenced by a greater ventilatory response for a given absolute exercise intensity, an increase in W_b at a given V˙_E, a greater propensity toward EFL, and higher operating lung volumes than younger individuals (⁴⁶). Given the inherent effect of sex on the morphology of the pulmonary system, it stands to reason that healthy older women may be predisposed to mechanical ventilatory constraints of exercise hyperpnea. However, few studies have considered the combined effects of sex and healthy aging on the pulmonary system’s responses to exercise. We recently found that in those 60–80 yr old, women have greater W_b than men at a given V˙_E of >60 L·min⁻¹, a finding that is similar to studies in younger (20–40 yr) individuals (¹²). We also noted that older women were more likely to experience EFL during exercise than older men, which likely is attributed to the fact that healthy aging increases the ventilatory demands of exercise and reduces the maximal ventilatory capacity of the pulmonary system (¹²). As is the case with highly trained young individuals, the ventilatory demands of exercise in older adults commonly reach, and in some cases even exceed, the finite capacity of their pulmonary system, which results in EFL. The fact that older women are more likely to experience EFL during exercise in comparison to men is potentially due to sex differences in lung and airway size.

Given that older individuals have a greater W_b at a given V˙_E and breathe at higher lung volumes during exercise than younger individuals, they also have greater extra-diaphragmatic inspiratory muscle activation (⁴⁵). Moreover, the effect of sex on inspiratory muscle activation patterns during exercise is independent of age, implying that older women exhibit greater activation of the scalene and sternocleidomastoid muscles during exercise than younger women and older men (⁴⁵).

Although it seems that sex differences in the mechanical ventilatory response to exercise persist during the healthy aging process, several important questions remain unanswered. Clearly, more research is needed to clearly define the combined, and potentially interactive, effects of age and sex on the ventilatory response to exercise and the associated influence on the integrative response to exercise.

MATTERS OF CONTENTION

How to Make Sex-Based Comparisons

When comparing the sexes, there are several methodological concerns that influence the interpretations of results. Paramount among these concerns is the issue of size versus sex. In other words, are differences between the sexes due to a disparity in absolute size or due to sex specifically? For example, should V˙O₂ be compared between the sexes using absolute (L·min⁻¹) values or relative to body mass (mL·kg⁻¹·min⁻¹) or fat-free mass (mL·kg FFM⁻¹·min⁻¹)? The answer depends on the context of the question being asked. If an experimental question revolves around activities of daily living, then absolute comparisons would be most appropriate given the standardized metabolic requirement specific activities. Conversely, when studying athletes or generally younger healthy individuals who are performing intense exercise, relative exercise intensities would be more appropriate. Overall, we suggest that both absolute and relative comparisons should generally be presented together and that interpretation of results depends on the specific question at hand.

Second, the menstrual cycle results in repeated variations in circulating hormone concentrations that potentially influence the integrative response to exercise. For example, differences in estrogen can influence leg blood flow and blood pressure regulation (^48,49). Although the fluctuations in hormone concentrations are generally predictable, it is known that the absolute concentrations and the temporal pattern of change can vary greatly between and within women not taking any exogenous form of hormone (e.g., a contraceptive) (⁵⁰). Accordingly, without daily measures of hormone levels, standardizing women to a certain phase of the menstrual cycle based on self-report is not an effective approach to achieve a standardized hormonal concentration. Yet, despite the fact that resting chemosensitivity varies over the course of the menstrual cycle, exercise ventilation seems unaffected (⁵⁰). Most likely, the myriad of feedback and feedforward control during exercise overrides any effect of hormonal variation. Unfortunately, there is no perfect solution to account for cyclical changes in hormonal levels. Testing women at random points in the menstrual cycle potentially introduces variability. Alternatively, testing women only at specific points during their menstrual cycle compromises the generalizability of the findings. Overall, the determination of whether to account for the menstrual cycle depends on the research question being asked.

Dyspnea

Briefly defined as “the subjective experience of breathing discomfort,” dyspnea is a common sensory consequence of whole-body dynamic exercise. Several studies indicate that the perceptions of dyspnea during activities of daily living are greater in women than men across a wide range of ages (i.e., >38 yr old) (⁵¹). This finding also has been noted during incremental exercise at absolute exercise intensities or levels of V˙_E in a laboratory setting (^12,47,52). However, it is essential to note that when comparisons between the sexes are made at relative exercise intensities or levels of V˙_E, sex differences in the perceived intensity of exertional dyspnea are no longer present (^12,47,52). When interpreting the evidence concerning the effect of sex on the perceived intensity of dyspnea during exercise, two important considerations must be made. First, one must consider the notion that assessing the perception of dyspnea is founded on the principle of psychophysics, whereby a stimulus (e.g., exercise) is linked to a sensation (i.e., dyspnea) (⁵³). The salient question then becomes: are the sex differences in exertional dyspnea a product of sex per se, or simply a matter differences in the magnitude of the stimulus? The answer to this question depends on context. Assessing dyspnea at absolute exercise intensities is most appropriate when aiming to determine the sensory responses to exercise where the metabolic cost is relatively similar between individuals regardless of size or physical fitness (e.g., during activities of daily living). This approach often is used in studies involving older adults or patient populations for whom maintaining functional capacity (in absolute terms) is essential. Previous studies have shown that the perceived intensity of dyspnea is higher in older (60–80 yr) women than men during exercise at a standardized metabolic workload (^12,47). However, when comparing groups of individuals with relatively high functional capacities (e.g., healthy young adults), the results derived from comparing them at absolute exercise intensities may be biased by the effect of size or physical fitness. In this case, comparing the intensity of perceived dyspnea at relative exercise intensities (e.g., at discrete fractions of peak V˙O₂) is likely to be more informative.

To date, the vast majority of studies focusing on sex differences in exertional dyspnea have been observational, but determining the mechanisms underpinning sex differences in dyspnea during exercise requires an experimental approach. A possible mechanism is the effect of sex differences in pulmonary system morphology on mechanical ventilatory constraint during exercise (⁵⁴). It could be reasoned that the greater propensity toward EFL and the high W_b during exercise in older women relative to older men may increase the perception of dyspnea. To test this hypothesis, we reduced (using a helium-oxygen mixture) and increased (using an inspiratory resistor) the degree of mechanical ventilatory constraint, in a single-blinded fashion, while healthy older men and women (60–80 yr old) exercised at a standardized intensity (~74% of V˙O_2max) (⁵⁵). Despite significantly decreasing and increasing the degree of mechanical ventilatory constraint, the perceived intensity of dyspnea was unaffected, suggesting that the higher perception of dyspnea in older women relative to older men during exercise is not caused by sex differences in EFL and W_b.

Dyspnea is, by definition, a subjective or relative measure and a given rating of perceived dyspnea (e.g., “5” on the Borg category-ratio 10 scale) may occur as a result of different stimulus in different individuals or different groups of individuals. Thus, although it is tempting to speculate that sex differences in the magnitude of perceived dyspnea during exercise are linked to sex differences in pulmonary system morphology, our experimental data in healthy older men and women suggest that this is not the case (⁵⁵). It is possible that some other physiological mechanism is causing sex differences in exertional dyspnea, but it is essential to note that psychosocial effects on symptom perception cannot be discounted (⁵⁶). Indeed, the mechanisms of dyspnea are complex and multifactorial, and the evidence relating to sex on the perception of dyspnea remains equivocal.

FUTURE DIRECTIONS

We have made the case that there are important sex differences in pulmonary system morphology that result in functional differences; Figure 1 provides a schematic of the available evidence to support our hypothesis. However, there is much left that is unexplained and definitive conclusions are still forthcoming. Here, we briefly present areas that, with further attention, could strengthen our understanding of the complexities of the integrative nature cardiopulmonary physiology of exercise and sex differences.

Linking Anatomy to Function

Imaging studies of the airways (^4,5) have shown a sex difference whereby the luminal areas of the large airways are smaller in women, even when matched for lung size or height. It has been assumed that airway size differences explain the sex differences in W_b and V˙O_2RM, and by association the effect on the integrative physiological response to exercise. A difficulty here is that anatomical and functional measures have been made in separate studies making the interpretation difficult. Measures of airway anatomy and functional measures in the same individuals are therefore necessary to draw a direct link between the airway size differences and the mechanical as well as metabolic cost of breathing during dynamic whole-body exercise.

Gas Exchange

Maintaining arterial oxygenation is an important determinant of V˙O_2max and exercise performance. There may be sex differences in the mechanisms of EIAH, but this is not definitive. We suggest that there is a need to obtain additional temperature-corrected arterial blood gases in women performing exercise to appropriately characterize the presence or absence of sex-based difference in gas exchange derangements.

Diaphragm Fatigue

Additional studies are required to fully understand the nature of sex differences in diaphragm fatigue. It is not known under what conditions the female diaphragm is more resistant to fatigue. Manipulation of a number of variables is necessary to determine the contributing factors such as (i) exercise intensity, (ii) absolute versus relative diaphragmatic work, and (iii) compromised O₂ delivery (i.e., hypoxia or reduced blood flow). To our knowledge, the central versus peripheral contribution to diaphragm fatigue and how this might differ based on sex also has not been addressed.

Blood Flow Redistribution

From a pulmonary mechanics standpoint, it could be hypothesized that women would have a greater degree of blood flow distribution when subjected to high levels of respiratory muscle work. There is now some evidence to suggest that in response to isolated respiratory muscle work, women demonstrate a blunted respiratory muscle metaboreflex (^57,58). It is known that there are sex differences in sympathetic nerve activity (i.e., muscle sympathetic nerve activity) at rest and during small muscle mass exercise. It is unknown how activation of the respiratory muscle metaboreflex modulates vasomotor outflow in men versus women.

Clinical Implications

We have made the case that healthy women are more susceptible to pulmonary limitations during exercise when compared with healthy men. If correct, then this may mean that women with cardiorespiratory disease may be particularly vulnerable to the negative consequences associated with pulmonary limitations. How these differences may relate to clinical populations merits brief comment. There are reports of sex differences in the diagnosis, severity, treatment, and outcomes of many pulmonary diseases (⁵⁹). For example, in chronic obstructive pulmonary disease, sex differences in disease phenotype have been observed, where women seem to have more pronounced peripheral muscle dysfunction relative to men (⁶⁰). The aforementioned exercise limitations could influence the effectiveness of currently prescribed pulmonary rehabilitation, which also have noted sex differences (^61,62). For example, inspiratory muscle training can improve inspiratory muscle strength and reduce dyspnea in those with chronic obstructive pulmonary disease (⁶³). We have shown that older healthy women have greater dyspnea at absolute workloads than men; therefore, there seems to be potential for respiratory muscle training in older women with COPD. What is clear, however, is the lack of research in the rehabilitation setting investigating sex difference in women (⁶⁴), which impedes progress and optimal treatment. Overall clarification of the factors underlying the sex differences in dyspnea and exercise limitation will enable the establishment of evidence-based guidelines for the treatment of activity-related dyspnea and inform the development of sex-specific treatment plans for the management of cardiopulmonary disease.

CONCLUSIONS

There are notable sex-based differences in pulmonary system anatomy. Principally, women have smaller conducting airways than height- or lung size-matched men. As a result of these smaller airways, women have a greater W_b and V˙O_2RM during exercise. However, if and how these differences in structure and function result in changes in the integrative response to exercise is still relatively unknown. Like men, women can experience EFL, develop EIAH, redistribute blood flow when W_b is manipulated, and their respiratory muscle energetics influence locomotor fatigue. But whether there is a sex difference in these integrative responses is unclear. Finally, one must consider the effects of size, sex, hormones, and training status when designing and interpreting studies regarding sex differences on the integrative response to exercise.