Beyond Spirometry: Linking Wasted Ventilation to Exertional Dyspnea in the Initial Stages of COPD

Abstract

Exertional dyspnea, a key complaint of patients with chronic obstructive pulmonary disease (COPD), ultimately reflects an increased inspiratory neural drive to breathe. In non-hypoxemic patients with largely preserved lung mechanics – as those in the initial stages of the disease – the heightened inspiratory neural drive is strongly associated with an exaggerated ventilatory response to metabolic demand. Several lines of evidence indicate that the so-called excess ventilation (high ventilation-CO₂ output relationship) primarily reflects poor gas exchange efficiency, namely increased physiological dead space. Pulmonary function tests estimating the extension of the wasted ventilation and selected cardiopulmonary exercise testing variables can, therefore, shed unique light on the genesis of patients’ out-of-proportion dyspnea. After a succinct overview of the basis of gas exchange efficiency in health and inefficiency in COPD, we discuss how wasted ventilation translates into exertional dyspnea in individual patients. We then outline what is currently known about the structural basis of wasted ventilation in “minor/trivial” COPD vis-à-vis the contribution of emphysema versus a potential impairment in lung perfusion across non-emphysematous lung. After summarizing some unanswered questions on the field, we propose that functional imaging be amalgamated with pulmonary function tests beyond spirometry to improve our understanding of this deeply neglected cause of exertional dyspnea. Advances in the field will depend on our ability to develop robust platforms for deeply phenotyping (structurally and functionally), the dyspneic patients showing unordinary high wasted ventilation despite relatively preserved FEV₁.

Keywords:

• Dsypnea
• emphysema
• mild COPD
• gas exchange
• blood flow

Introduction

Dyspnea, either induced or worsened by exertion [1], is the primary determinant of disablement and reduced quality-adjusted-life-years in patients facing the devastating consequences of chronic obstructive pulmonary disease (COPD) [2]. Most patients suffering from COPD show a post-bronchodilator forced expiratory volume in one second (FEV₁) within normal range or only mildly reduced (herein termed “mild” COPD) [2–4], suggesting only minor disease [5]. However, there is growing recognition that a sizable fraction of these patients experience out-of-proportion dyspnea relative to the severity of the lung-mechanical abnormalities [2–4,6–9]. Extensive work carried out by our group [8,10–27] (reviewed in refs. [9,28]) showed that rather than primarily driven by mechanical constraints, exertional dyspnea is strongly related to high ventilation (V̇E) relative to carbon dioxide output (V̇CO₂) (excess ventilation) [16,28], signaling increased wasted ventilation in the physiological dead space (VD_phys), i.e. a high VD fraction of the tidal volume (high VD/VT ratio) [18–20,29].

In this concise review, we initially provide a primer on the determinants of gas exchange efficiency in health and inefficiency in COPD. We focus on increased wasted ventilation rather than venous admixture since the latter is not a prominent future in most patients with mild COPD. Based on current neurobiological concepts of exertional dyspnea, we discuss how wasted ventilation can translate into breathlessness in individual patients. The structural underpinnings of wasted ventilation in mild COPD are reviewed: we bring a new perspective (impaired perfusion of non-emphysematous tissue) in addition to emphysema as a cause of VD_phys. To further advance the field, we outline some critical unanswered questions on the seeds and consequences of wasted ventilation in these patients. We give special attention to how physiological measurements can be combined with respiratory functional imaging and metrics of emphysema extension and distribution by computed tomography (CT) to improve our understanding of this poorly understood cause of exertional dyspnea.

A primer on gas exchange efficiency in health: rest and exercise

The human lungs are admirably designed to meet their fundamental task – gas exchange – with precision and efficiency. For instance, alveolar ventilation (V̇A),/capillary perfusion (Q̇c) matching is favored by gravitational and non-gravitational mechanisms [30,31]:

• Due to the lungs’ weight, both the density of capillaries (high Q̇c) and the compliance of the dependent tissue (high V̇A) increases from apex to base;

• The gravity also acts on the weight of the pulmonary blood, increasing the flow across the basal regions;

• The effect of gravity on blood flow is greater compared to that exerted on airflow, i.e., V̇A/Q̇c decreases slightly from top to bottom [32]; and,

• “Matched” geometries of the airways and arteries, allowing a preferential distribution of V̇ and Q̇ to the same secondary pulmonary lobules.

Young normal subjects show narrow distributions of V̇A/Q̇c ratios. In contrast, the distribution widens with aging [33], largely due to reduced ventilation of dependent lung regions secondary to small airway closure, i.e. the top-to-bottom decrease in V̇A/Q̇c is more pronounced in the elderly. Regardless of age, the aforementioned mechanisms of V̇A/Q̇c matching act to minimize (Figure 1) [35]:

Figure 1. Selected ventilatory and gas exchange responses to incremental cardiopulmonary exercise testing. Proportional decreases in dead space (VD)/tidal volume (VT) ratio (i.e. the physiological dead space) (a) and ventilation (V̇E)/carbon dioxide output (V̇CO₂) (b) ratios maintain arterial carbon dioxide partial pressure (PaCO₂) close to resting value during mild-to-moderate exercise (c). The V̇E/V̇CO₂ response contour is established by both the slope and the intercept of the linear V̇E-V̇CO₂ relationship (d). V̇E increases out of proportion to V̇CO₂ after the respiratory compensation point (RCP) (b–d) leading to respiratory alkalosis (c) to compensate for progressive lactic acidemia. Note the increases in the lowest V̇E/V̇CO₂ when the lactate threshold is reached at a low exercise intensity, i.e. before the stabilization of V̇E/V̇CO₂ at the “true” nadir. Given eucapnia during exercise in most dyspneic patients with mild COPD, their high V̇E-V̇CO₂ largely reflects a high physiological dead space, i.e. wasted ventilation. Reproduced, with permission of the publisher, from: Neder et al. [34].

• The amount of end-capillary blood with a low PO₂ added to the arterial blood (venous admixture) caused by a “true” shunt (no V̇A, preserved Q̇c) or a shunt “effect” (low V̇A/Q̇c, usually due to areas of low V̇A with normal Q̇c), and

• The volume of air not exposed to gas exchanging units which fills the VD_phys (wasted ventilation) secondary to anatomical VD, “alveolar” VD (preserved V̇A, no Q̇c), VD “effect” (high V̇A/Q̇c, usually due to normal V̇A but low Q̇c) and any contribution of shunt to increase the arterial partial pressure for CO₂ (PaCO₂).

Although high-intensity exercise is associated with increasing V̇A/Q̇c inhomogeneity and a progressive widening in alveolar-arterial O₂ difference (P(A-a)O₂) [36], there is little change in P(A-a)O₂ at exercise intensities more relevant to subjects’ functioning in daily life, i.e. mild-moderate [37],. Crucially, VD/VT decreases (↓) markedly on exercise because VT increases (↑) out of proportion to conducting airway VD due to radial traction, and the higher compliance of the alveoli over that of the airways [38]. Since PaCO₂ is maintained (↔) within ±3 mmHg from rest, V̇_E/V̇CO₂ decreases in tandem with VD/VT (Figure 1) [39]: Equation [1] $$\leftrightarrow {\text{PaCO}}_2=\frac{1}{\downarrow \frac{{\dot{\mathrm{V}}}_{\mathrm{E}}}{{\dot{V}\mathit{\text{CO}}}_2}\times \left(1-\left(\frac{\uparrow {\mathrm{V}}_{\mathrm{D}}}{\uparrow \uparrow {\mathrm{V}}_{\mathrm{T}}}\right)\right)}$$Equation [1]

Gas exchange inefficiency and wasted ventilation in mild COPD

As long proposed by Riley and Cournand [40], the term “gas exchange inefficiency” is herein used to characterize all abnormalities that conspire against the ideal lung, i.e. a lung without V̇A/Q̇c inequalities in which an ideal PAO₂ can be estimated (assuming no P(A-a)CO₂ and a constant respiratory exchange ratio). Specifically, such a definition does not include impaired gas transfer residing at the level of the alveolar-capillary membrane, i.e. “diffusion” limitation. COPD is characterized by a marked heterogeneity in the nature and degree of dysfunction of individual structural components of pulmonary gas exchange (in)efficiency) [41]. Therefore, considerable variability exists in the relative contribution of factors conspiring to increase wasted ventilation as the disease progresses [42] or between individuals showing similar spirometry. In this context, capillary obliteration/dysfunction in areas with alveolar destruction (emphysema) or areas with relatively preserved lung parenchyma may decrease Q̇c relative to V̇A, reducing local PACO₂ at a given V̇CO₂ (see What is the structural basis of wasted ventilation in the initial stages of COPD?) (Figure 2) [44].

Figure 2. Putative mechanisms by which the structural abnormalities associated with COPD development may increase alveolar ventilation (V̇a)/capillary perfusion (Q̇c) relationship, increasing the wasted ventilation in the physiological dead space. The relative contribution of these abnormalities vary substantially amongst patients with similar spirometric findings justifying the use of more elaborated pulmonary function tests in addition to imaging to improve disease phenotyping. Reproduced, with permission of the American Thoracic Society. Copyright © 2023 American Thoracic Society. All rights reserved. Neder [43]. Annals of the American Thoracic Society is an official journal of the American Thoracic Society.

Using the multiple inert gas exchange technique (MIGET) (Table 1), Wagner et al. showed that a rightward shift of V̇A/Q̇c distribution to high values was more commonly observed in patients with a preponderance of emphysema over airway disease [58]. It remains unclear what the sources of afferent stimuli are that allow the respiratory controller to estimate how much “extra”- V̇E is required to overcome an enlarged VD_phys to maintain eucapnia (Equation 1Equation [1] $$\leftrightarrow {\text{PaCO}}_2=\frac{1}{\downarrow \frac{{\dot{\mathrm{V}}}_{\mathrm{E}}}{{\dot{V}\mathit{\text{CO}}}_2}\times \left(1-\left(\frac{\uparrow {\mathrm{V}}_{\mathrm{D}}}{\uparrow \uparrow {\mathrm{V}}_{\mathrm{T}}}\right)\right)}$$Equation [1] ) [39]. One potential explanation suggests that intra- or between-breaths fluctuations in PCO₂ of the arterial blood leaving the lungs would stimulate the peripheral and/or central chemoreceptors [43]. In several pathological circumstances involving a high VD_phys, therefore, PaCO₂ is reduced [18,59–61] with the resulting low PCO₂ set-point further increasing the ventilatory drive in a vicious circle (Figure 3) [62]. Regardless of the precise determinants, excess ventilation during cardiopulmonary exercise testing (CPET) is anticipated in COPD patients showing the combination of (a) hypocapnia (hyperventilation), (b) increased “alveolar” VD and VD “effect” and (c) a low VT with (b) and (c) jointly conspiring to increase VD_phys [63]: Equation [2] $$\uparrow \frac{{\dot{\mathrm{V}}}_{\mathrm{E}}}{{\dot{V}\mathit{\text{CO}}}_2}=\frac{1}{\downarrow {\text{PaCO}}_2\times \left(1-\left(\frac{\uparrow {\mathrm{V}}_{\mathrm{D}}}{\downarrow {\mathrm{V}}_{\mathrm{T}}}\right)\right)}$$Equation [2]

Figure 3. A putative explanation by which increased physiological dead space (wasted ventilation) and alveolar hyperventilation may trigger exertional dyspnea. The combination of pulmonary fibrosis and emphysema (CPFE) is depicted for illustrative purposes. In the presence of increased (⇑) airspace (alveolar ventilation (V̇a)) relative to capillary perfusion (Q̇c)) (wasted ventilation), less CO₂ from the mixed venous blood is unloaded into the right alveolus. Consequently, the alveolar carbon dioxide tension (PACO₂) diminishes (⇓) markedly but PCO₂ at the end of the capillary (c’) may rise transitorily, subsequently enriching the PCO₂ of the arterial blood sensed by the central and/or peripheral chemoreceptors. This may add to the other sources of afferent stimuli to the pontine–medullary respiratory centers (red circle). The inspiratory neural drive (IND) is heightened; accordingly. Of note, IND is also modulated by cortical (black circle) and limbic (blue circle) influences. Excessive ventilation of the left alveolus with preserved V̇a/Q̇c relationship, triggered by increased chemostimulation or other source(s) of additional ventilatory stimuli, leads to alveolar hyperventilation and hypocapnia. A low arterial Pco₂ may downward shift the CO₂ set-point, impair the cerebrovascular reactivity (CVR) to CO₂, and increase the central and peripheral chemoreceptor gains. Hypoxemia (arterial partial pressure for O₂ (PaO₂)), if present, sums up with these excitatory inputs to further increase the IND in a vicious circle that fuels breathlessness. In mild COPD, the evidence accrued to date does not support a major contribution of hyperventilation and/or low PaO₂ to the patient high IND, suggesting a dominant role for the wasted ventilation.

Figure 4. Schematic representation of the key mechanisms by which alveolar ventilation (V̇a)/capillary perfusion (Q̇c) mismatching may conspire to increase the inspiratory neural drive (IND), eventually leading to exertional dyspnea and exercise intolerance in patients with mild COPD. The putative mechanisms that have been ruled out in previous studies are highlighted (reviewed in refs. [9,28]). Emphasis is given to the unknown contribution of impaired perfusion of non-emphysematous lung tissue (in addition to emphysema) to increased (↑) wasted ventilation in the physiological dead space (high dead space (VD)/tidal volume (VT) ratio), heightened dyspnea at given work rate (WR) in these patients was largely commensurate to preserved dyspnea-ventilation (V̇_E) relationship, in keeping with the pattern of “excessive breathing” (Figure 5). See text for further elaboration. Definition of abbreviations: HPV: hypoxic pulmonary vasoconstriction; PAP = pulmonary arterial pressure. Modified, with permission of the publisher, from: Neder et al. [111].

Figure 5. A simplified framework to expose the major “pulmonary” determinants of exertional dyspnea during incremental cardiopulmonary exercise testing as related to the pathophysiology of COPD. The key physiological abnormalities are highlighted: excess ventilation (high ventilation (V̇E)-(carbon dioxide output (V̇CO₂) relationship) and the attainment of critical inspiratory constraints (CIC). In most dyspneic patients with mild COPD, increased wasted ventilation in the physiological dead space (VD_phys) prompts a high, V̇E/V̇CO₂ ratio leading to a pattern of “excessive breathing” rather than “restrained” (or “constrained”) breathing, i.e. increased dyspnea at given work rate (WR) but within the expected range when expressed relative to the heightened V̇E. Definition of abbreviations and symbols: ⇑: high; ⇓: low’; V̇A: alveolar ventilation; Q̇c: capillary perfusion; Pa: arterial partial pressure; PET: end-tidal partial pressure; Hypervent: alveolar hyperventilation; CIC: end-inspiratory lung volume/total lung capacity≥ 0.9, tidal volume (VT)/inspiratory capacity ≥ 0.7, VT plateau, all reached at an abnormally-low work rate. Ranges of dyspnea severity based on percentiles distribution of scores at a given ventilation as a function of age and sex: 5th–25th: “mild”; 25th–50th: “moderate”; 50th 75th: “severe”; 75th 95th: “very severe” [14]. Reproduced, with permission of the publisher, from: Neder [66].

Figure 6. An illustrative example of how respiratory functional imaging (phase-resolved functional lung (PREFUL) ¹H magnetic resonance imaging for ventilation and perfusion [109–113] can be combined with computed tomography-based emphysema extension and distribution (parametric response mapping) to provide relevant insights into the genesis of out-of-proportion dyspnea in patients with mild COPD. This 54 years old woman with preserved FEV₁ and out-of-proportion decrease in lung diffusing capacity for carbon monoxide showed high ventilation-CO₂ output and an increased physiological dead space (VD_phys) on cardiopulmonary exercise testing. Despite trivial emphysema (red and pink), note extensive areas of perfusion deficit percent (QDP) in non-emphysematous regions of the lung which were well ventilated (low ventilation deficit percent (VDP)) at rest and during exercise (light green). These findings are in keeping with the notion that impaired pulmonary perfusion beyond expected by emphysema, may increase the wasted ventilation in these patients, contributing to exertional dyspnea.

Table 1. A simplified overview of main invasive and noninvasive physiologic measurements potentially sensitive to disturbances in pulmonary gas exchange efficiency in patients with mild COPD.

Measurement	Rationale/Advantages	Main Limitations
Noninvasive
O2 deficit [45]	• ⇑ values indicate a low arterial PO2 (estimated from the O2 dissociation curve) at a given alveolar PO2, providing a noninvasive approximate of P(A-a)O2, i.e. poor efficiency in O2 transfer.[46]	• Inaccuracies in PaO2 estimation due to under/overestimation of PaCO2 based on PETCO2 • The estimation of PaO2 becomes unreliable at a SpO2 > 94%
PETCO2	• ⇓ values might indicate lower mean mixed expired “alveolar” PCO2 due to the diluting effects of areas of wasted ventilation which, be definition, have lower PCO2	• Whereas a superficial and fast breathing pattern may reduce PETCO2, areas of hypoventilation may increase PETCO at a given PaCO2 (“normalizing” PETCO2)
Volumetric capnography [47]	• Slower kinetics/lower amplitude of expired PCO2 increase toward PaCO2 (or a close estimate) signals high VDphys • Once Fowler dead space (∼ anatomical) is taken into consideration, VA can be estimated	• Influenced by acute hypo or hyperventilation • Changes in cardiac output increase variability • Airflow obstruction prolongs the time to complete tidal exhalation, overestimating VDphys
DLCO [48]	• ⇓ values indicate reduced rate of CO transfer which is influenced by inhomogeneities in V̇A/Q̇c matching (amongst others) • Using different FIO2, the contribution of “membrane” and “blood” components to DLCO can be portioned out [49]	• Sensitive to total ventilation, mixed venous CO, diffusion limitation, [hemoglobin} and blood volume (including in units with normal V̇A/Q̇c) • Sensitive, but not specific
KCO [50]	• ⇓ values indicate reduced gas transfer related to its distribution volume (VA) • Increases curvilinearly as VA decreases	• Impaired ventilation distribution may decrease VA at a given TLC, leading to KCO normalization despite extensive abnormalities in gas exchange efficiency • Complex KCO-VA relationship [51] • See DLCO limitations
DLNO [52]	• ⇓ values relative to DLCO biased to indicate a higher contribution of the “membrane” diffusing capacity compared with the “blood” component of pulmonary gas transfer	• Might be also influenced by the membrane component; unclear clinical advantage over DLCO • See DLCO limitations
Exercise V̇E:V̇CO2	• ⇑ wasted ventilation and/or venous admixture (if associated with low PaO2) requires a higher V̇E to control PaCO2 at a given metabolic demand	• Might reflect alveolar hyperventilation (rather than a high VDphys), requiring concomitant measurements (or a close estimate) of PaCO2
Invasive
PaO2	• ⇓ age-corrected values secondary to ⇓ V̇A/Q̇c and “true” shunt	• It should be related to PAO2 to provide an index of gas exchange inefficiency
P(A-a)O2 [46]	• ⇑ age-corrected values indicate that less O2 has been added to arterial blood at a given mixed expired alveolar O2 pressure	• In isolation, it cannot differentiate diffusion limitation, ⇓ V̇A/Q̇c, and shunt as cause of hypoxemia • ⇑ FIO2 variably increases P(A-a)O2
P(a-ET)CO2 (and P(a-ET)CO2/ PaCO2)	• ⇑ values indicate that the arterial blood contains more CO2 than expected for a given mixed expired alveolar CO2 pressure, excluding the effect of anatomical and apparatus dead space on wasted ventilation	• Negative values may occur on exercise since VT and mixed venous PCO2 increases, and lung volume decreases as expiration progresses [53] • See PETCO2 limitations
Bohr dead space [54]	• ⇑ values indicate a lower mean mixed exhaled CO2 pressure at given mean alveolar CO2 pressure, signaling an increased volume of air within the conducting airways, i.e. it is biased to reflect anatomical dead space	• A gas sample captured late within a single exhalation. do not necessarily reflect the true mean alveolar gas composition • Ignores P(A-a)CO2, leading to lower dead space compared with Bohr-Enghoff VDphys
Bohr-Enghoff dead space (VDphys) [55]	• ⇑ values indicate a lower mean mixed exhaled CO2 pressure at a given PaCO2 signaling increased contribution of air which was not exposed to gas exchanging units (wasted ventilation) • Since mean PaCO2 might be higher than mean PACO2 in disease, VDphys is sensitive to intrapulmonary shunt and diffusion impairment, being strongly influenced by V̇A/Q̇c heterogeneity	• ⇑ values might result from any mechanism that increases P(A-a)CO2, being an index of overall gas exchange inefficiency • ⇑ ventilation of areas of ⇑ V̇A/Q̇c reduces PACO2 at given PaCO2 leading to higher dead space compared to Bohr formulation
Shunt by the 100% O2 test [56]	• As any ventilated alveoli which receives 100% O2 would contribute to PaO2, ⇑ values indicate that a higher fraction of cardiac output has not been exposed to 100% O2 (either intra- or extra-pulmonary, i.e. “anatomical” shunt)	• Alveolar collapse induced by 100% O2 may increase the estimated shunt • Hemodynamic effects of 100% O2 may reduce anatomical shunting, particularly during exercise
Shunt fraction (Q̇s/Q̇t)	• ⇑ values indicate that a higher fraction of cardiac output has the same (low) O2 content of mixed venous blood • Conceivably, the best estimation of venous admixture	• Requires central catheterization • Unpredictable increases with higher cardiac output on exercise

Symbols and Abbreviations: ⇑: high; ⇓:low; []: concentration; CO: carbon monoxide; CO₂: carbon dioxide; FI_: inspired fraction; K: transfer coefficient; MIGET: multiple inert gas exchange technique; NO: nitric oxide; O₂: oxygen; Pa: arterial partial pressure; PA: alveolar partial pressure; PET: end tidal partial pressure; RER: respiratory exchange ratio; Sp: hemoglobin saturation by pulse oximetry; Q̇c: capillary blood flow; Q̇s: shunted blood; Q̇t: cardiac output; TL: lung transfer factor; TLC: total lung capacity; VA: alveolar volume; V̇A: alveolar ventilation; VD_phys; physiologic dead space; V̇CO₂: pulmonary carbon dioxide output; V̇_E: minute ventilation.

Reproduced, with permission of the publisher, from: Neder et al. [57] (On-Line Supplement).

The following potential explanations for increased afferent stimuli and excess ventilation have been carefully excluded in our investigations involving patients with mild COPD: (a) high pulmonary arterial pressures; (b) metabolic acidotic drive [64], since excess ventilation emerged well before the so-called lactate threshold (in fact, it was frequently seen at rest)[65]; (c) hypoxemia[66]; (d) reduced CO₂ set-point as overt respiratory alkalosis was an exception rather than a rule[18]; (e) low VT leading to a high VD/VT as patients dynamically hyperinflate only at near maximum exercise; and, (f) heightened CO₂ chemosensitivity[26]. Thus, the only extant explanation for excess ventilation is increased areas of high V̇A/Q̇c; in fact, enlarged VD_phys was the closest correlate of excess ventilation in our studies (Figure 4) [10,14,18,67].

It should be acknowledged, however, that in practice, VD/VT is calculated by Enghoff’s modification of the seminal Bohr’s formulation. Due to the complexities in measuring a representative PACO₂, Enghoff proposed using PaCO₂ instead. However, the higher the CO₂ volume added to the arterial blood relative to PACO₂, the greater the overestimation of VD_phys by the Enghoff formulation. The bottom line is that spuriously high Enghoff’s VD/VT might be seen in patients showing a sizeable shunt and those with more extensive V̇A/Q heterogeneity [68]. Although it might be argued that this is likely more relevant in patients with advanced COPD, it may have contributed to enlarged VD_phys in our studies with milder disease [10,14,18,67].

Translating wasted ventilation into exertional dyspnea in mild COPD

Dynamic exercise is associated with a manyfold increase in V̇E. Remarkably, the sense of breathing effort/work does not increase substantially in healthy individuals. For instance, most non-trained men and women report only mild-moderate respiratory discomfort up to near-maximum CPET [69]. A downward shift in the end-expiratory lung volume to encroach on the expiratory reserve volume allows VT to be positioned on the most compliant portion of the sigmoid pressure-volume relationship of the respiratory system [70]. Optimization of the lungs’ dynamic compliance causes a harmonious coupling of the inspiratory neural drive (IND) from bulbo-pontine and cortical respiratory control centers and VT expansion, mitigating the breathing discomfort [9].

Dyspnea arises in patients with cardiopulmonary diseases due to a disparity between the heightened IND and the capacity of the respiratory system to respond appropriately [71]. Such an imbalance has been reflected in a consistent increase in several physiologic ratios that ultimately relate ventilatory demand to capacity [72], such as:

• Instantaneous V̇E/maximal ventilatory capacity (measured or estimated from FEV₁), recently termed dynamic ventilatory reserve [73];

• Diaphragm electromyography and esophageal pressure, both related to values at maximal inspiration [67]; and,

• VT/inspiratory capacity or vital capacity, i.e., the volumes theoretically available for tidal expansion [74].

The relationship of these ratios to dyspnea intensity is modulated by the presence or absence of mechanical constraints that are sufficiently high to hinder the translation of the IND into the mechanical act of breathing (VT and V̇E) (Figure 5):

• If the heightened IND can be fully converted to VT displacement and a higher V̇E, patients are more likely to use the words “increased work/effort” to describe their uncomfortable respiratory sensations [27]. Since the increased dyspnea ratings are rather commensurate to the heightened ventilatory demands, we descriptively describe this source of exertional dyspnea as excessive breathing; or

• If the heightened IND cannot be appropriately converted into VT displacement and a higher V̇E due to significant restrictive mechanical constraints, patients frequently report “unsatisfied inspiration” [27]. Thus, dyspnea ratings increase at any given work rate and V̇E compared with healthy subjects. (constrained (or restrained) breathing).

The corollary is that wasted ventilation contributes to a high IND, a high V̇E/V̇CO₂, and the CPET pattern of excessive breathing (Figures 4 and 5) in dyspneic patients with largely preserved FEV₁ [8,75].

What are the structural bases of wasted ventilation in mild COPD?

Emphysema as the sole determinant of wasted ventilation

Emphysema epitomizes a structural abnormality associated with increased VD_phys since it reduces the surface for gas exchange, and alveolar coalescence increases distal airspaces that are not perfused (“true” VD) [68,76,77]. Classical studies found that panlobular and centrilobular emphysema may lead to increased VD_phys since the earlier stages of COPD [81–83]. In fact, we [66] and others [78,79] found that wasted ventilation in emphysematous areas is associated with excess ventilation in mild-moderate COPD. Disproportionally impaired lung diffusing capacity for carbon monoxide (D_LCO) relative to FEV₁ has been associated with emphysema burden in several studies with mild COPD [2,80,81]. Moreover, at least part of the low D_LCO depicted by patients with incipient/minor emphysema has been ascribed to the “membrane” component [82]. Emphysema extension was related to resting V̇A/Q̇c heterogeneity measured by MIGET in mild COPD[83]. but this was not the case during exercise [83]. Importantly, however, excess ventilation was found in patients with only trace emphysema [24,66,84]. Since no study to date has assessed the topographic/regional distribution of V̇A/Q̇c under the modulating influence of emphysema, it remains unclear whether capillary destruction due to emphysema fully explains their increased wasted ventilation (Figure 2) (see Perspectives: amalgamating functional imaging and clinical physiology to advance the field) [68,77,85,86].

Reduced capillary blood volume in non-emphysematous lung

Tobacco-related injury/dysfunction of the lung microvasculature[87] and vascular compression/distortion by areas of gas trapping secondary to small airway disease[88] may increase Q̇c, leading to areas of high V̇_A/Q̇c (VD “effect”)[68,77]. Impaired release of endothelium-derived relaxing factors, reduced endothelial nitric oxide synthase, and increased expression of endothelin-1 have been described in the pulmonary arterioles of patients with mild-moderate COPD [89]. A recent review identified sixteen pathways associated with pulmonary vascular remodeling in COPD, most related to endothelial cell dysfunction and damage and only remotely linked to emphysema development [90]. Pulmonary vascular tone had a central role in regulating V̇A/Q̇c in non-hypoxemic patients with largely preserved FEV₁ [83]. In fact, impaired microvascular blood flow out-of-proportion to emphysema burden has been reported in these patients[91,92]; moreover, manipulations of Q̇c via increased venous return [93] and inhaled nitric oxide[94] variably lessened excess ventilation. We [66] and others [95] found impaired global pulmonary perfusion during exercise in patients with mild COPD showing excess ventilation. Our group [67] and other investigators [81,96] also described a strong association between a low DL_CO, a “window to pulmonary microcirculation” [97], and VD_phys in mild COPD. Recently, we reported that reduced pulmonary vascular volume adjusted by emphysema extent was associated with low DL_CO and heightened exertional ventilation in dyspneic smokers with minor emphysema [98]. Moreover, gadolinium-enhanced magnetic resonance imaging (MRI) studies found indirect evidence of reduced perfusion in non-emphysematous lung in smokers and patients with minor COPD [92,99].

Areas of air trapping – a common finding in mild COPD – [100] are thought to be poorly ventilated, decreasing (instead of increasing) V̇_A/Q̇c [101]. However, patchy pockets of air trapping may theoretically compress adjacent arterioles/capillaries, increasing V̇_A/Q̇c in non-emphysematous tissue (Figure 2) [88] This scenario, however, remains highly speculative since no study to date has shown impaired perfusion adjacent to scattered regions of air trapping, and inhaled bronchodilators typically fail to improve VD_phys in these patients [102].

Perspectives: amalgamating functional imaging and clinical physiology to advance the field

Despite the marked advances in our understanding of the decisive role of wasted ventilation in eliciting out-of-proportion exertional dyspnea in patients with apparently minor COPD [8,10–27], several questions remain unanswered:

• Does high V̇A/Q̇c due to relative oligemia in areas with still preserved lung parenchyma (non-emphysematous) contribute substantially to increased VD_phys?

• If this proves true, are areas of high V̇A/Q̇c due to low Q̇c in non-emphysematous lung at least partially reversible with pulmonary vasodilators or, in fact, do they reflect largely irreversible vascular damage (or, alternatively, the compressive effects of patchy areas of air trapping (Figure 2)?

• If these abnormalities are reversible, is the expected improvement in VD_phys large enough to decrease excess ventilation and exertional dyspnea, i.e., complex endpoints that are characteristically multifactorial in COPD?[34]

• How are these responses modulated by comorbidities that are known to increase exertional ventilation, such as pulmonary hypertension, lung fibrosis, and diastolic dysfunction?

• Does the burden of high V̇A/Q̇c due to low Q̇c in non-emphysematous lungs hold any mechanistic influence in out-of-proportion exertional pulmonary hypertension in COPD? and,

• Do these abnormalities precede the development of emphysema in longitudinal studies – as long proposed by Liebow [103]?

A crucial limitation of pulmonary function tests aiming at interrogating lungs’ efficiency in gas transfer (Table 1) is their inherent lack of information on the regional distribution of V̇A and Q̇c as modulated by emphysema [104,105]. To fill this gap, there is a growing role for respiratory functional imaging approaches co-registered to emphysema distribution by CT [106–108], An illustrative approach may combine free-breathing phase-resolved functional lung (PREFUL) ¹H MRI [109–113] with computed tomography to provide quantitative, spatially-resolved ventilation, perfusion, and ventilation/perfusion deficits, taking into consideration the modulating influence of emphysema in patients showing or not increased VD_phys (Figure 6).

Conclusions

Restraining the functional assessment of dyspneic patients with COPD to spirometry is prone to provide an incomplete and potentially misleading picture of the actual extension of the disease. Such a caveat is particularly true in patients in the initial stages of when FEV₁ is largely preserved. In this concise review, we outlined how increased wasted ventilation – frequently well beyond expected from the severity of FEV₁ impairment and the emphysema burden – is relevant to patients’ breathlessness. In fact, DL_CO, a physiological biomarker that is known to be negatively influenced by VD_phys [114], remains the closest correlate of exertional dyspnea in these patients [67]. Moreover, bronchodilators, the cornerstone for treatment of the consequences of expiratory flow limitation (air trapping and high lung volumes) [115] with negligible effects on VD_phys [102] are poorly effective in improving exertional dyspnea in mild disease [116]. Advances in the field will therefore depend on our ability to develop robust platforms for comprehensively phenotyping (structurally and functionally), the dyspneic patients showing unexpectedly high VD_phys [57]. This will create the bases for testing existing (e.g. oral and inhaled vasodilators) and newer therapeutic approaches geared to improve V̇A/Q̇c in carefully selected patients showing the deleterious respiratory sensory consequences of increased wasted ventilation.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Additional information

Funding

J Alberto Neder has been funded by a New Clinician Scientist Program from the Southeastern Ontario Academic Medical Association (SEAMO), Canada. The funders had no role in data collection and analysis or preparation of the manuscript.