Secondary Logo

Journal Logo


Exercise Intolerance in Heart Failure: Central Role for the Pulmonary System

Lalande, Sophie1; Cross, Troy J.2,3; Keller-Ross, Manda L.4; Morris, Norman R.5; Johnson, Bruce D.3; Taylor, Bryan J.6

Author Information
Exercise and Sport Sciences Reviews: January 2020 - Volume 48 - Issue 1 - p 11-19
doi: 10.1249/JES.0000000000000208
  • Free
  • Journal Club

Key Points

  • Engorgement of the bronchial vasculature due to increased pulmonary vascular pressure and pulmonary congestion results in greater bronchial conductance and airflow obstruction in heart failure.
  • Patients with heart failure show an abnormally high work of breathing during exercise, largely due to an excessive ventilation and a greater resistive load of breathing.
  • Skeletal muscle mass and an overactivation of group III/IV afferents contribute to ventilatory inefficiency during exercise in patients with heart failure.
  • Pulmonary hypertension limits exercise tolerance in patients with heart failure by negatively affecting ventilation-perfusion matching, pulmonary gas exchange, and oxygen delivery to exercising muscles.
  • A less negative inspiratory intrathoracic pressure and a more positive expiratory intrathoracic pressure automatically improve the cardiac response to exercise in patients with heart failure.


Chronic heart failure (HF) is a complex multiorgan syndrome characterized by pathophysiological changes in cardiac, vascular, musculoskeletal, endocrine, and pulmonary function. Exercise limitation is a hallmark of HF (1), and an increasing degree of intolerance to exercise is associated with poor prognosis. However, despite the typical pathophysiological sequelae of HF, no relationship exists between cardiac function (e.g., left ventricular ejection fraction, left ventricular volumes, and cardiac output) and exercise tolerance in HF patients (2). Indeed, it is now clear that the physiological underpinning of exercise intolerance in HF is multifactorial and that several noncardiac mechanisms play a significant role. Of each of the major organ systems negatively affected by HF, the pulmonary system is the most centrally integrated into the syndrome. As HF progresses, several pulmonary abnormalities may become apparent, such as increased physiological dead space, mild ventilation-to-perfusion mismatch, increased bronchial conductance and congestion, impaired lung diffusing capacity, and pulmonary hypertension (Fig. 1). It is our contention that the “dysfunctional” pulmonary system is a key determinant of exertional dyspnea and exercise intolerance in HF. In this review, we will present up-to-date evidence and mechanistically discuss our current understanding of the central role that the pulmonary system plays in exercise limitation in patients with HF.

Figure 1
Figure 1:
Schematic representation of key pulmonary manifestations of HF and their contribution to exercise limitation. RV, right ventricular; V D/V T, dead-space ventilation; E/ CO2, minute ventilation to carbon dioxide ratio; V/Q, ventilation-to-perfusion ratio.


Changes in the caliber of the airway mucosal vasculature, or bronchial circulation, may contribute to changes in pulmonary function in HF (3). The bronchial circulation runs concurrent with the airways to the terminal bronchiole level and supplies blood flow to the highly vascularized bronchial mucosa (3,4). The bronchial mucosal capillaries are located within 100 μm of the airway lumen. With this close proximity, it has been hypothesized that engorgement of the bronchial vasculature, reflected by an increased bronchovascular conductance, could result in a displacement of the tissue toward the adjacent lumen airspace, thereby narrowing the internal diameter of the compliant distal airways and contributing to the increased airway resistance seen in HF (3–5). Bronchial blood flow is regulated through α-adrenoreceptor and cholinergic autonomic vasoconstrictor mechanisms as well as by nitric oxide–dependent vasodilatory pathways (3,4). There is evidence that airway vascular conductance is sensitive to changes in pulmonary vascular pressures, inflammatory mediators, and increased firing from cardiac stretch receptors (3). Given that many of these changes are pathophysiological alterations associated with HF, the vascular tone and conductance of the bronchial circulation may well be altered in this patient group. Indeed, as cardiac filling pressures rise and pulmonary congestion progressively develops in HF, blood flow may back up into the bronchial circulation and influence the airway caliber, resulting in airflow limitation.

The development of a modified soluble gas technique has allowed the noninvasive measurement of bronchial blood flow in HF (4,6). Briefly, measuring the disappearance or uptake of inert and highly soluble dimethyl ether gas over time during a series of breath-hold maneuvers allows estimation of bronchial blood flow; this technique has been validated against direct bronchial blood flow measurements in animals using radiolabeled microspheres (4). To test the hypothesis that bronchovascular engorgement may affect pulmonary function in HF, Ceridon et al. (7) examined the relationship between bronchial blood flow, bronchovascular conductance, and pulmonary function in individuals with HF by measuring bronchial blood flow and bronchovascular conductance at rest using the modified soluble gas technique in 30 individuals with HF (New York Heart Association (NYHA) class I–IV; ejection fraction, 26 ± 7%) and 12 age-matched controls (ejection fraction, 61 ± 6%). Resting bronchovascular conductance was more than 40% higher in HF when compared with control. Moreover, there was a positive relationship between airflow obstruction and bronchovascular conductance in HF only. That is, greater airflow obstruction was associated with higher bronchovascular conductance in HF.

Indirect evidence appears to support the hypothesis of bronchovascular engorgement being related to altered airway function at rest and during exercise in HF. Cabanes et al. (5) reported that in individuals with HF, inhalation of methacholine resulted in airway hyperresponsiveness, similar to the response observed in bronchial asthma. Methacholine causes both constriction of the airways and dilation of the bronchial blood vessels, and the authors suggested that the constriction of the airways was partly mediated by the dilation of the bronchial blood vessels or bronchial engorgement. Indeed, administration of methoxamine, an α1-agonist known to constrict bronchial blood vessels, fully prevented the methacholine-induced hyperresponsiveness. Furthermore, the protective effect of methoxamine was in turn blocked by the administration of an α-adrenergic antagonist. In a separate study (8), individuals with a history of HF treated with methoxamine had an improved exercise tolerance and reduced exertional breathlessness. These findings support the contributing role of bronchovascular engorgement to the exertional dyspnea and exercise intolerance observed in patients with HF.


Diffusing capacity of the lungs for carbon monoxide (DLCO) represents the rate at which the lung can take up the inhaled test gas carbon monoxide. Although there was no relationship between DLCO and bronchial blood flow in HF (9), DLCO is clearly reduced in HF (10). DLCO represents the sum of two subcomponents, the conductance of the alveolar-capillary membrane and pulmonary capillary blood volume. Alterations in either the conductance of the alveolar-capillary membrane or pulmonary capillary blood volume will affect DLCO. Using a novel single-breath technique (11), several studies examined DLCO and subcomponents at rest and during exercise in HF individuals with both preserved ejection fraction (HFpEF) and reduced ejection fraction (HFrEF) (12,13). When compared with an age-matched healthy group, Olson et al. (12) found that both resting DLCO and the change in DLCO during exercise were significantly less in HFrEF (ejection fraction, 24 ± 2%). Moreover, the rise in DLCO relative to the increase in cardiac output during exercise was lower for the HFrEF group, suggesting a reduced alveolar-capillary recruitment. In a separate study, Olson et al. (13) also found DLCO to be significantly lower for HFpEF patients at rest and during exercise. Interestingly, although pulmonary capillary blood volume was lower compared with healthy individuals at rest and peak exercise in HFpEF, the initial relative rise in pulmonary capillary blood volume from rest to low-intensity exercise (20 W) was significantly greater in HFpEF than in healthy individuals (Fig. 2). Presumably, this rapid increase in pulmonary capillary blood volume was related to a greater increase in pulmonary venous pressures coupled to a higher ventilation, due to a higher breathing frequency, and an increased ventilatory drive relative to exercise workload in patients with HFpEF in comparison with healthy individuals. From 20 W to peak exercise, there was no further increase in pulmonary capillary blood volume in HFpEF patients despite an increase in cardiac output, indicating that the ability to recruit greater pulmonary capillary blood volume becomes rapidly saturated in HFpEF. The authors speculated that increases in venous return at the onset of exercise cannot be accommodated by the left ventricle such that right ventricular output transiently exceeds left ventricular output, and blood pools in the pulmonary circulation. Healthy individuals, on the other hand, had a significant increase in both pulmonary capillary blood volume and cardiac output throughout exercise. Thus, patients with HF display altered lung diffusing capacity at rest and during exercise, indicating that a lower pulmonary capillary blood volume and an impaired ability to transfer gases through the alveolar-capillary membrane contribute to exercise intolerance in this population.

Figure 2
Figure 2:
A. Pulmonary capillary blood volume (V C) as a function of cardiac output (c) during exercise. Circles, healthy individuals; squares, HF individuals with preserved ejection fraction (HFpEF). *P < 0.05 versus healthy individuals. B. Percent change from rest to matched submaximal absolute workload (20 W). Black bar, healthy individuals; white bar, patients with HFpEF. [Adapted from (13). Copyright © 2016 Elsevier. Used with permission].


The work of breathing (Wb) is inordinately higher in HF patients during physical exertion (14). Exercise hyperventilation and mechanical derangement primarily contribute to this greater overall mechanical cost of breathing in HF. Exercise hyperventilation is a hallmark of HF, and several mechanisms have been proposed as mediators of this excessive ventilatory response, including, but not limited to: (i) alveolar ventilation-perfusion mismatching, (ii) skeletal muscle hypoperfusion and deconditioning, (iii) juxtacapillary receptor stimulation consequent to pulmonary congestion and/or hypertension, (iv) heightened central/peripheral chemosensitivity, and (v) increased firing of group III/IV skeletal muscle afferents (i.e., the “skeletal muscle” hypothesis) (15–18). Notwithstanding these many different mechanisms, the following is clear: the HF patient breathes more than a healthy individual during exercise, and this hyperventilation begets an increase in the overall Wb at a given exercise intensity. However, not only is total Wb higher at a given work rate, respiratory muscle work also is greater at a given level of ventilation in this population (i.e., mechanical derangement).

There are two principal loads imposed on the respiratory muscles during spontaneous breathing: the resistive and elastic loads (inertive loads are often neglected due to their small magnitude). Patients with HF experience an increased elastic load to breathing. Certainly, dynamic lung compliance is systematically lower compared with healthy individuals and appears to worsen during physical exertion (19). Such increased lung “stiffness” has been attributed to competition between lung and cardiac tissue for intrathoracic space (e.g., cardiomegaly), the engorged pulmonary and/or bronchial vasculature, pulmonary interstitial edema (particularly during exercise), and remodeling of the lung parenchyma (7,20). Patients with HF may also present with a relatively stiffer chest wall (14). Taken together, the stiffer lungs and chest walls of HF patients contribute to a greater elastic Wb at moderate expired minute ventilations during graded exercise (14).

Inspiratory and expiratory resistive Wb is higher in HF patients compared with healthy controls at standardized minute ventilations during graded exercise (14). The larger resistive load to breathing may be due to the incipient pulmonary congestion affecting airway geometry and distribution of radial and interfacial forces (5), an enhanced bronchomotor tone and/or hyperresponsiveness (5), and/or increased collapsibility of the smaller airways during expiration (i.e., expiratory flow limitation) (21,22). In addition, pulmonary congestion and an enhanced bronchomotor tone may also serve to increase lung tissue resistance (23). When the elastic and resistive components of respiratory work are expressed relative to the total Wb incurred during exercise, inspiratory and expiratory resistive Wb seems to make larger contributions to the overall mechanical cost of breathing compared with healthy individuals (Fig. 3). By extension, the contribution of inspiratory and expiratory elastic Wb to the total mechanical cost of breathing is lower in patients with HF at a given minute ventilation during exercise.

Figure 3
Figure 3:
The relative contributions of resistive and elastic respiratory work to the total work of breathing (Wb) during graded exercise. Both panels illustrate the general relationship between each component of the Wb and minute ventilation during graded exercise for patients with heart failure (HF) and healthy controls, respectively. Fine stippling () represents inspiratory resistive Wb. Coarse stippling () denotes expiratory resistive Wb. The horizontal hatching () represents inspiratory elastic Wb, whereas vertical hatching () denotes expiratory elastic Wb. Please note that the relative contribution of inspiratory and expiratory resistive Wb to the total mechanical cost of breathing appears larger for HF patients than controls during graded exercise. [Adapted from (14). Copyright © 2012 European Respiratory Society. Used with permission].

The larger mechanical cost of breathing results in larger oxygen and blood flow requirements of the active respiratory muscles (24). Seeing that HF is often accompanied by an impaired cardiac output response to exercise, little reserve exists for the heart to meet the oxygen and blood flow demands of both the respiratory and locomotor muscles, simultaneously. As such, the respiratory and locomotor muscles may compete for an adequate share of the prevailing cardiac reserve, hastening the onset of exertional fatigue in HF patients during exercise. Indeed, reducing the overall Wb seemingly improves oxygen delivery to the locomotor muscles and increases exercise duration in this population (25). It is therefore important to consider the mechanical cost of breathing as a contributing factor to exercise intolerance in HF. Not only is the mechanical cost of breathing inordinately greater in HF patients compared with healthy individuals, this higher overall Wb appears largely due to differences in the resistive rather than elastic loads of breathing.


Skeletal muscle group III/IV afferents are a major contributor to excessive exercise ventilation in HF patients. Mechano- and metaboreceptors sense changes in muscle length, volume, and by-products of muscle contraction and feedback through group III/IV afferents via the dorsal horn of the spinal cord to the medullary respiratory centers (i.e., rostral ventral medulla, caudal ventrolateral medulla, and nucleus tractus solitarius) to increase ventilation (26,27). It is well known that patients with HF experience significant changes to the skeletal muscle, and these alterations likely result in excessive group III/IV afferent feedback and, therefore, greater ventilation during exercise.

The “skeletal muscle hypothesis” suggests that cardiac patients experience skeletal muscle myopathies, in part, due to alterations in blood perfusion to the muscle, which contribute to skeletal muscle fatigability (28), exercise intolerance and, subsequently, symptoms of dyspnea and fatigue (Fig. 1) (18). These muscle myopathies include muscle wasting and a shift in fiber type distribution, leading to a greater metabolic buildup during exercise in HF (18,29–31). As such, there is an early dependence on anaerobic metabolism, excessive early depletion of high-energy phosphate bonds, and excessive early intramuscular acidification during exercise, which leads to an exaggerated metaboreflex in patients with HF (32). Mechanoreflex sensitivity is shown to be enhanced in animal models of HF (33), but this is less clear in patients with HF as it is more difficult to isolate the mechanoreflex in humans (33). Collectively, both the animal and human literature suggest that mechano- and metaboreflex, and subsequently group III/IV skeletal muscle afferent activity, is exaggerated during exercise in these patients (29,32). The overactivation of group III/IV afferent activity causes aberrant increases in ventilation in patients with HF (Fig. 1). Our work (34,35) and the work of others (29) confirm the importance of the contribution of the skeletal muscle to an enhanced ventilatory response and impaired exercise tolerance in patients with HF.

Work by Scott et al. (29) demonstrated that when activating the metaboreflex by postexercise circulatory occlusion in the lower extremity, ventilation was elevated in patients with HF compared with age-matched controls. These investigators then demonstrated that the excessive elevation in ventilation in HF was associated with elevated hydrogen ion levels and increased lactate concentration (36). Furthermore, an infusion of sodium bicarbonate to reduce serum hydrogen ion concentration completely abolished the ventilatory response. The aforementioned findings clearly demonstrate the importance of the metabolic contribution to the ventilation response in patients with HF. During passive movement of the lower extremity to activate the mechanoreflex, a small increase in ventilation was observed in HF and controls, but the increase was similar between groups (29,37). Both animal and human studies suggest that metabolites, such as adenosine triphosphate and/or lactic acid, can sensitize muscle mechanoreceptors (33), but the direct response of this metabolically induced mechanical sensitization on ventilation is unclear. Collectively, these studies suggest that during exercise, the overactivation of group III/IV afferents causes an exaggerated ventilatory response in patients with HF.

Ventilatory efficiency, characterized by the relationship between minute ventilation and the production of carbon dioxide (E/CO2slope), is a powerful predictor of mortality, even more so than V̇O2peak, in patients with HF (38). Olson et al. (39) demonstrated that E/CO2slope was elevated in mild HF compared with controls during exercise and activation of metaboreflex with lower-extremity postexercise circulatory occlusion. Indeed, when group III/IV afferent activity was inhibited with fentanyl (a μ-opioid receptor agonist), E/CO2slope was reduced significantly, indicating that partial blockade of group III/IV afferent activity improved ventilatory efficiency during exercise in HF patients. As further evidence for the important role of skeletal muscle to the exercise ventilatory response, it was recently demonstrated that low muscle mass was strongly associated with a high E/CO2slope in patients with HF (Fig. 4A). Concurrent with muscle wasting, HF patients also demonstrate histological changes indicating a shift in fiber-type proportion and myosin heavy chain isoform expression. The loss of myosin heavy chain type I and gain of myosin heavy chain type IIx result in greater metabolic activity, reduced oxidative capacity, and, therefore, greater activation of group III/IV afferents. Subsequently, when afferent feedback is inhibited, the signals from group III/IV afferents to the nucleus tractus solitarius of the brainstem are attenuated, which will lower the ventilatory response during exercise (34,39). Furthermore, the greatest reductions in E/CO2slope with afferent blockade occurred in patients with the least leg muscle mass (Fig. 4B) and lowest V̇O2peak (34). Thus, in this study, patients with HF who experience exercise intolerance (low V̇O2peak) have the greatest improvement in breathing efficiency during exercise when afferent feedback is blunted (34). It should be noted, however, that it is likely not the low muscle mass per se that is causing the greater activation of group III/IV afferent activity, it is likely the histological changes that are concurrent with muscle wasting that influence group III/IV afferent activity in individuals with HF. In conclusion, the skeletal muscle is a major contributing factor to the control of ventilation and intolerance to exercise in HF patients. Clinically, this is relevant and is consistent with the evidence that exercise training of the skeletal muscle causes significant improvements in tolerance to exercise in HF.

Figure 4
Figure 4:
Relationship between leg muscle mass (%) and V E/V CO2slope during the placebo condition (A) and change in V E/V CO2 with afferent blockade (injections of fentanyl) (B). V E/V CO2slope during the placebo condition was associated with leg muscle mass in heart failure (HF) (r 2 = 0.58, P = 0.02), but not in controls (CTL) (r 2 = 0.30, P = 0.16). With blocking afferent feedback, the relationship between V E/V CO2slope and leg muscle mass was abolished in HF (r < 1.0, P = 0.96) and remained nonsignificant in CTLs (r 2 = 0.16, P = 0.32). [Adapted from (34). Copyright © 2016 Elsevier. Used with permission].


Pulmonary hypertension is a common complication of HF, often occurring as a “symptom” of the underlying condition. It is estimated that pulmonary hypertension develops in ~60% and ~70% of HF patients with severe left ventricular systolic dysfunction or isolated left ventricular diastolic dysfunction, respectively (40). The presence of pulmonary hypertension is associated with more severe symptoms, worse exercise tolerance, and poor prognosis in these patient populations (41). Exercise introduces an impressive stress to the lungs, where elevations in venous return increase pulmonary blood volume and blood flow by 50% and 300%, respectively. Although the healthy, highly compliant, and low-resistance pulmonary vasculature can readily accommodate these marked increases in blood volume and flow, this reserve is compromised in HF patients, particularly those with pulmonary hypertension (42). In agreement with others (43,44), our unpublished data (Taylor BJ, Borlaug BA, Miller AD, Olson TP, Johnson BD, 2017) suggest that in HFrEF patients: (i) the ratio of mean pulmonary arterial pressure to cardiac output is greater at peak exercise in HF with versus without pulmonary hypertension; (ii) pulmonary arterial compliance at peak exercise is significantly lower in HF with compared to without pulmonary hypertension; and (iii) a greater mean pulmonary arterial pressure‐to‐cardiac output ratio and lower absolute pulmonary arterial compliance at peak exercise are related to impaired exercise capacity across all HF patients (Fig. 5). These findings unequivocally demonstrate that the response to exercise in HF with pulmonary hypertension is characterized by an exaggerated rise in pulmonary vascular pressures and limited or no pulmonary vascular reserve, both of which are key determinants of exercise limitation in these patients. Mechanistically, pulmonary hypertension likely contributes to exercise intolerance in HF through a substantial increase in right ventricular afterload that restricts right ventricular contractile reserve and forward flow of blood from the right ventricle to the pulmonary vasculature and systemic circulation. This, in combination with impaired recruitment and distension of the lung vasculature secondary to pulmonary hypertension‐related remodeling of the pulmonary resistance vessels, negatively affects ventilation-perfusion matching, pulmonary gas exchange, and systemic oxygen transport to the exercising muscles, limiting a patient’s ability to exercise (Fig. 6).

Figure 5
Figure 5:
Unpublished data (Taylor BJ, Borlaug BA, Miller AD, Olson TP, Johnson BD, 2017) collected in 36 heart failure patients with reduced ejection fraction (HFrEF) show that 1) mean pulmonary arterial pressure–to–cardiac output ratio (mPAP/) is greater (A) and pulmonary arterial compliance (PAC) is lower (C) at peak exercise in pulmonary hypertension (PH)–HF (n = 22) compared with HF without PH (HF) (n = 14); and 2) a greater mPAP/ ratio (B) and lower absolute PAC (D) at peak exercise are significantly related to lower peak oxygen consumption (V̇O2peak) across all HF patients. **P < 0.01 HF versus PH-HF.
Figure 6
Figure 6:
The right ventricular (RV) and pulmonary gas exchange response to exercise in healthy (A) and in HF patients with pulmonary hypertension (PH-CHF) (B). In PH-HF: “exercise” PH, defined as mean pulmonary arterial pressure (mPAP) >30 mm Hg and total pulmonary vascular resistance (TPVR) >3 Wood units (WU), presents a substantial increase in RV afterload, limiting RV fractional area change (FAC) and forward flow of blood and impairing pulmonary capillary perfusion. Hypoperfused alveoli leads to alveolar ventilation-to-perfusion (V A/) mismatch, increased dead-space ventilatory (V D/V T), elevated ventilatory equivalent for CO2 ( E/ CO2), and decreased partial pressure of end-tidal CO2 (PETCO2) (i.e., impaired pulmonary gas exchange). LV, left ventricle; PA, pulmonary artery.

The pathogenesis of pulmonary hypertension in HF is complex. Two distinct subsets of pulmonary hypertension due to HF have been identified: isolated postcapillary pulmonary hypertension (IpcPH) caused by backward transmission of increased left ventricular filling pressure into the pulmonary circulation, and combined pre- and postcapillary pulmonary hypertension (CpcPH) caused by abnormalities in pulmonary arterial structure and function that are characterized by an excessive rise in pulmonary arterial pressure and pulmonary vascular resistance. The reader is directed toward Vachiéry et al. (40) for a detailed hemodynamic definition and differentiation of IpcPH and CpcPH. The presence of CpcPH more greatly impairs aerobic capacity in HF (45). Although the precise mechanisms that underpin the development of CpcPH remain debated, it may be related to the increasing severity and chronicity of the underlying HF (46). Moreover, the presence of CpcPH is thought to reflect “superimposed” pulmonary vasoconstriction secondary to decreased nitric oxide availability, increased endothelial expression of endothelin-1, and desensitization to natriuretic peptide‐induced vasodilation (47). Rather provocatively, we have previously suggested that the aforementioned increase in endothelin-1 concentration in CpcPH is hypoxemia mediated, and as such, there may be a role for systemic hypoxemia in the CpcPH component of pulmonary hypertension in HF (48). In 39 patients with stable systolic HF, it was found that, despite relative uniformity in HF disease severity, patients with CpcPH were hypoxemic relative to patients with no pulmonary hypertension or IpcPH (48). The transpulmonary pressure gradient (i.e., the severity of pulmonary hypertension) was inversely related to partial pressure of arterial oxygen and oxygen saturation in the CpcPH patients only. In addition, plasma endothelin-1 concentration correlated with partial pressure of arterial oxygen and transpulmonary pressure gradient in CpcPH patients only. These data provide compelling evidence that systemic hypoxemia plays a role in the CpcPH of HF, potentially via a hypoxia-induced increase in endothelial release of the vasoconstrictor endothelin-1.

It has also been reported that derangements in ventilatory and pulmonary gas exchange indices at rest and during submaximal and maximal exercise in HF are likely related to the development and severity of pulmonary hypertension (49,50). For example, at an equivalent oxygen consumption, end-tidal CO2 (PETCO2) and arterial oxygen saturation were greater in no pulmonary hypertension and IpcPH versus CpcPH patients (49). Conversely, dead-space ventilation and the E/CO2slope were lower in no pulmonary hypertension and IpcPH versus CpcPH patients (49). Interestingly, the exercise-induced change in dead-space ventilation, E/CO2slope, and PETCO2 correlated significantly with the change in mean pulmonary arterial pressure, diastolic pressure difference, and transpulmonary pressure gradient in CpcPH patients only (49). The mechanisms by which the elevated pulmonary vascular pressures associated with CpcPH worsen exercise pulmonary gas exchange in these patients are multifactorial, but likely include an inability to adequately increase pulmonary vascular perfusion during exercise, resulting in impaired ventilation-perfusion matching and a reduction in oxygen delivery to the working tissues with subsequent stimulation of ergoreceptors and/or chemoreceptors and a consequent relative hyperventilatory response to exercise (Fig. 6).


In patients with HF, altered interactions between the heart and lungs exist due to decreased lung and left ventricular compliance, increased cardiac size, high cardiac filling pressure, and altered receptor sensitivity to neural activation. Exercise further affects these cardiopulmonary interactions by stimulating an increase in the depth and frequency of breathing, which accentuates the fluctuations in intrathoracic pressure with peak inspiratory intrathoracic pressure becoming more negative and peak expiratory intrathoracic pressure becoming more positive with increasing exercise intensities. Patients with HF often exhibit excessive ventilation for a given exercise intensity, primarily characterized by an increased respiratory rate rather than tidal volume despite significant room to increase tidal volume by further encroaching on inspiratory reserve volume (21), possibly due to weakened inspiratory muscles (51). The avoidance of high lung volumes during exercise, often at the expense of seemingly unnecessary large positive expiratory intrathoracic pressures, results in significant wasted effort in patients with HF (21,52).

Inspiration and the accompanying decrease in intrathoracic pressure result in increased intraluminal pressure and left ventricular preload and afterload. Patients with HF tend to be less sensitive to changes in left ventricular preload due to a decreased left ventricular compliance (1); therefore, a reduction in left ventricular preload through a less negative intrathoracic pressure in patients with elevated filling pressures may decrease effective filling to a level more acceptable for left ventricular performance (53). In accordance with the strong inverse linear relationship between stroke volume and left ventricular afterload (54), an overriding influence of left ventricular afterload in patients with HF suggests that the negative intrathoracic pressure accompanying inspiration may reduce stroke volume in these individuals (53). Thus, a less negative inspiratory intrathoracic pressure and a larger positive expiratory intrathoracic pressure could preserve cardiac function during exercise in patients with HF.

The effect of manipulating inspiratory intrathoracic pressure and expiratory intrathoracic pressure on stroke volume during moderate-intensity exercise was investigated in patients with HFrEF (NYHA class I–II). During moderate-intensity exercise, inspiratory unloading using a ventilator (less negative intrathoracic pressure) increased stroke volume, possibly due to a reduction in left ventricular afterload, whereas inspiratory loading using resistors (more negative intrathoracic pressure) did not affect stroke volume in patients with HFrEF (55). Accordingly, expiratory loading (more positive expiratory intrathoracic pressure) elicited increases in stroke volume during moderate-intensity exercise in patients with HFrEF (56), which may be caused by a beneficial reduction in left ventricular preload. Thus, manipulation of intrathoracic pressure, through a less negative inspiratory intrathoracic pressure and a more positive expiratory intrathoracic pressure, automatically improves stroke volume in patients with HFrEF. Furthermore, inspiratory unloading increased exercise endurance in patients with HF (57). Thus, strategies designed to limit the negative cardiorespiratory interactions during exercise, such as volitionally altering breathing pattern or the use of positive pressure breathing during expiration, could be beneficial to patients with HF.


In summary, we propose that derangements in the pulmonary system, including engorgement of the bronchial vasculature, ventilatory inefficiency, elevated muscular Wb, inspiratory muscle weakness, impaired pulmonary gas exchange, and pulmonary hypertension, are central determinants of exercise intolerance in HF. Our work and that by others identify the importance of the pulmonary system in limiting exercise tolerance in HF, and highlight that treating the pulmonary manifestations of HF such as pulmonary pressure and congestion as well as adopting exercise training intervention to improve inspiratory muscle resistance/endurance and increase skeletal muscle mass may yield promising results in improving exercise tolerance in this population.


This study was supported by National Instituted of Health (NIH) grant HL071478. Bryan J. Taylor and Sophie Lalande were supported by the American Heart Association Postdoctoral Fellowships (12POST12070084 and 0920054G, respectively). Troy J. Cross was supported by an Australian Postgraduate Award. Manda L. Keller-Ross was supported by an NIH/NIAMS T32AR56950 grant.


1. Kitzman DW, Higginbotham MB, Cobb FR, Sheikh KH, Sullivan MJ. Exercise intolerance in patients with heart failure and preserved left ventricular systolic function: failure of the Frank-Starling mechanism. J. Am. Coll. Cardiol. 1991; 17(5):1065–72.
2. Franciosa JA, Park M, Levine TB. Lack of correlation between exercise capacity and indexes of resting left ventricular performance in heart failure. Am. J. Cardiol. 1981; 47(1):33–9.
3. Ceridon M, Wanner A, Johnson BD. Does the bronchial circulation contribute to congestion in heart failure? Med. Hypotheses. 2009; 73(3):414–9.
4. Wanner A, Mendes ES, Atkins ND. A simplified noninvasive method to measure airway blood flow in humans. J. Appl. Physiol. 2006; 100(5):1674–8.
5. Cabanes LR, Weber SN, Matran R, et al. Bronchial hyperresponsiveness to methacholine in patients with impaired left ventricular function. N. Engl. J. Med. 1989; 320(20):1317–22.
6. Morris NR, Ceridon ML, Beck KC, et al. Exercise-related change in airway blood flow in humans: relationship to changes in cardiac output and ventilation. Respir. Physiol. Neurobiol. 2008; 162(3):204–9.
7. Ceridon ML, Morris NR, Hulsebus ML, Olson TP, Lalande S, Johnson BD. Influence of bronchial blood flow and conductance on pulmonary function in stable systolic heart failure. Respir. Physiol. Neurobiol. 2011; 177(3):256–64.
8. Cabanes L, Costes F, Weber S, et al. Improvement in exercise performance by inhalation of methoxamine in patients with impaired left ventricular function. N. Engl. J. Med. 1992; 326(25):1661–5.
9. Ceridon ML, Morris NR, Olson TP, Lalande S, Johnson BD. Effect of supine posture on airway blood flow and pulmonary function in stable heart failure. Respir. Physiol. Neurobiol. 2011; 178(2):269–74.
10. Puri S, Baker BL, Dutka DP, Oakley CM, Hughes JM, Cleland JG. Reduced alveolar-capillary membrane diffusing capacity in chronic heart failure. Its pathophysiological relevance and relationship to exercise performance. Circulation. 1995; 91(11):2769–74.
11. Ceridon ML, Beck KC, Olson TP, Bilezikian JA, Johnson BD. Calculating alveolar capillary conductance and pulmonary capillary blood volume: comparing the multiple- and single-inspired oxygen tension methods. J. Appl. Physiol. 2010; 109(3):643–53.
12. Olson LJ, Snyder EM, Beck KC, Johnson BD. Reduced rate of alveolar-capillary recruitment and fall of pulmonary diffusing capacity during exercise in patients with heart failure. J. Card. Fail. 2006; 12(4):299–306.
13. Olson TP, Johnson BD, Borlaug BA. Impaired pulmonary diffusion in heart failure with preserved ejection fraction. JACC Heart Fail. 2016; 4(6):490–8.
14. Cross TJ, Sabapathy S, Beck KC, Morris NR, Johnson BD. The resistive and elastic work of breathing during exercise in patients with chronic heart failure. Eur. Respir. J. 2012; 39(6):1449–57.
15. Ponikowski P, Banasiak W. Chemosensitivity in chronic heart failure. Heart Fail. Monit. 2001; 1(4):126–31.
16. Gehlbach BK, Geppert E. The pulmonary manifestations of left heart failure. Chest. 2004; 125(2):669–82.
17. Rubin SA, Brown HV. Ventilation and gas exchange during exercise in severe chronic heart failure. Am. Rev. Respir. Dis. 1984; 129(2 Pt 2):S63–4.
18. Coats AJ. The “muscle hypothesis” of chronic heart failure. J. Mol. Cell. Cardiol. 1996; 28(11):2255–62.
19. Agostoni P, Pellegrino R, Conca C, Rodarte JR, Brusasco V. Exercise hyperpnea in chronic heart failure: relationships to lung stiffness and expiratory flow limitation. J. Appl. Physiol. 2002; 92(4):1409–16.
20. Olson TP, Beck KC, Johnson BD. Pulmonary function changes associated with cardiomegaly in chronic heart failure. J. Card. Fail. 2007; 13(2):100–7.
21. Johnson BD, Beck KC, Olson LJ, et al. Ventilatory constraints during exercise in patients with chronic heart failure. Chest. 2000; 117(2):321–32.
22. Johnson BD, Beck KC, Olson LJ, et al. Pulmonary function in patients with reduced left ventricular function: influence of smoking and cardiac surgery. Chest. 2001; 120(6):1869–76.
23. Barnas GM, Sprung J, Kahn R, Delaney PA, Agarwal M. Lung tissue and airway impedances during pulmonary edema in normal range of breathing. J. Appl. Physiol. (1985). 1995; 78(5):1889–97.
24. Margaria R, Milic-Emili G, Petit JM, Cavagna G. Mechanical work of breathing during muscular exercise. J. Appl. Physiol. 1960; 15:354–8.
25. Olson TP, Joyner MJ, Dietz NM, Eisenach JH, Curry TB, Johnson BD. Effects of respiratory muscle work on blood flow distribution during exercise in heart failure. J. Physiol. 2010; 588(Pt 13):2487–501.
26. Iwamoto GA, Waldrop TG, Bauer RM, Mitchell JH. Pressor responses to muscular contraction in the cat: contributions by caudal and rostral ventrolateral medulla. Prog. Brain Res. 1989; 81:253–63.
27. Li J, Hand GA, Potts JT, Wilson LB, Mitchell JH. c-Fos expression in the medulla induced by static muscle contraction in cats. Am. J. Physiol. 1997; 272(1 Pt 2):H48–56.
28. Keller-Ross ML, Larson M, Johnson BD. Skeletal muscle fatigability in heart failure. Front. Physiol. 2019; 10:129.
29. Scott AC, Francis DP, Davies LC, Ponikowski P, Coats AJ, Piepoli MF. Contribution of skeletal muscle ‘ergoreceptors’ in the human leg to respiratory control in chronic heart failure. J. Physiol. 2000; 529(Pt 3):863–70.
30. Sullivan MJ, Green HJ, Cobb FR. Skeletal muscle biochemistry and histology in ambulatory patients with long-term heart failure. Circulation. 1990; 81(2):518–27.
31. Sullivan MJ, Green HJ, Cobb FR. Altered skeletal muscle metabolic response to exercise in chronic heart failure. Relation to skeletal muscle aerobic enzyme activity. Circulation. 1991; 84(4):1597–607.
32. Keller-Ross ML, Johnson BD, Joyner MJ, Olson TP. Influence of the metaboreflex on arterial blood pressure in heart failure patients. Am. Heart J. 2014; 167(4):521–8.
33. Sinoway LI, Li J. A perspective on the muscle reflex: implications for congestive heart failure. J. Appl. Physiol. 2005; 99(1):5–22.
34. Keller-Ross ML, Johnson BD, Carter RE, et al. Improved ventilatory efficiency with locomotor muscle afferent inhibition is strongly associated with leg composition in heart failure. Int. J. Cardiol. 2015; 202:159–66.
35. Olson TP, Joyner MJ, Eisenach JH, Curry TB, Johnson BD. Influence of locomotor muscle afferent inhibition on the ventilatory response to exercise in heart failure. Exp. Physiol. 2014; 99(2):414–26.
36. Scott AC, Wensel R, Davos CH, et al. Skeletal muscle reflex in heart failure patients: role of hydrogen. Circulation. 2003; 107(2):300–6.
37. Ives SJ, Amann M, Venturelli M, et al. The mechanoreflex and hemodynamic response to passive leg movement in heart failure. Med. Sci. Sports Exerc. 2016; 48(3):368–76.
38. Arena R, Myers J, Aslam SS, Varughese EB, Peberdy MA. Peak VO2 and VE/VCO2 slope in patients with heart failure: a prognostic comparison. Am. Heart J. 2004; 147(2):354–60.
39. Olson TP, Joyner MJ, Johnson BD. Influence of locomotor muscle metaboreceptor stimulation on the ventilatory response to exercise in heart failure. Circ. Heart Fail. 2010; 3(2):212–9.
40. Vachiéry JL, Adir Y, Barbera JA, et al. Pulmonary hypertension due to left heart diseases. J. Am. Coll. Cardiol. 2013; 62(Suppl. 25):D100–8.
41. Butler J, Chomsky DB, Wilson JR. Pulmonary hypertension and exercise intolerance in patients with heart failure. J. Am. Coll. Cardiol. 1999; 34(6):1802–6.
42. Lewis GD, Bossone E, Naeije R, et al. Pulmonary vascular hemodynamic response to exercise in cardiopulmonary diseases. Circulation. 2013; 128(13):1470–9.
43. Lewis GD, Murphy RM, Shah RV, et al. Pulmonary vascular response patterns during exercise in left ventricular systolic dysfunction predict exercise capacity and outcomes. Circ. Heart Fail. 2011; 4(3):276–85.
44. Dragu R, Rispler S, Habib M, et al. Pulmonary arterial capacitance in patients with heart failure and reactive pulmonary hypertension. Eur. J. Heart Fail. 2015; 17(1):74–80.
45. Caravita S, Faini A, Deboeck G, et al. Pulmonary hypertension and ventilation during exercise: role of the pre-capillary component. J. Heart Lung Transplant. 2017; 36(7):754–62.
46. Miller WL, Grill DE, Borlaug BA. Clinical features, hemodynamics, and outcomes of pulmonary hypertension due to chronic heart failure with reduced ejection fraction: pulmonary hypertension and heart failure. JACC Heart Fail. 2013; 1(4):290–9.
47. Moraes DL, Colucci WS, Givertz MM. Secondary pulmonary hypertension in chronic heart failure: the role of the endothelium in pathophysiology and management. Circulation. 2000; 102(14):1718–23.
48. Taylor BJ, Mojica CR, Olson TP, Woods PR, Frantz RP, Johnson BD. A possible role for systemic hypoxia in the reactive component of pulmonary hypertension in heart failure. J. Card. Fail. 2013; 19(1):50–9.
49. Taylor BJ, Smetana MR, Frantz RP, Johnson BD. Submaximal exercise pulmonary gas exchange in left heart disease patients with different forms of pulmonary hypertension. J. Card. Fail. 2015; 21(8):647–55.
50. Lim HS, Theodosiou M. Exercise ventilatory parameters for the diagnosis of reactive pulmonary hypertension in patients with heart failure. J. Card. Fail. 2014; 20(9):650–7.
51. McParland C, Krishnan B, Wang Y, Gallagher CG. Inspiratory muscle weakness and dyspnea in chronic heart failure. Am. Rev. Respir. Dis. 1992; 146(2):467–72.
52. Olson TP, Beck KC, Johnson JB, Johnson BD. Competition for intrathoracic space reduces lung capacity in patients with chronic heart failure: a radiographic study. Chest. 2006; 130(1):164–71.
53. Lalande S, Johnson BD. Breathing strategy to preserve exercising cardiac function in patients with heart failure. Med. Hypotheses. 2010; 74(3):416–21.
54. Weber KT, Janicki JS, Hunter WC, Shroff S, Pearlman ES, Fishman AP. The contractile behavior of the heart and its functional coupling to the circulation. Prog. Cardiovasc. Dis. 1982; 24(5):375–400.
55. Lalande S, Luoma CE, Miller AD, Johnson BD. Effect of changes in intrathoracic pressure on cardiac function at rest and during moderate exercise in health and heart failure. Exp. Physiol. 2012; 97(2):248–56.
56. Lalande S, Luoma CE, Miller AD, Johnson BD. Expiratory loading improves cardiac output during exercise in heart failure. Med. Sci. Sports Exerc. 2012; 44(12):2309–14.
57. O'Donnell DE, D'Arsigny C, Raj S, Abdollah H, Webb KA. Ventilatory assistance improves exercise endurance in stable congestive heart failure. Am. J. Respir. Crit. Care Med. 1999; 160(6):1804–11.

heart failure; exercise tolerance; work of breathing; bronchial vasculature; ventilatory inefficiency; pulmonary hypertension; intrathoracic pressure

Copyright © 2019 by the American College of Sports Medicine