Proceedings of the American Thoracic Society

Expiratory flow limitation is the pathophysiologic hallmark of chronic obstructive pulmonary disease (COPD), but dyspnea (breathlessness) is its most prominent and distressing symptom. Acute dynamic lung hyperinflation, which refers to the temporary increase in operating lung volumes above their resting value, is a key mechanistic consequence of expiratory flow limitation, and has serious mechanical and sensory repercussions. It is associated with excessive loading and functional weakness of inspiratory muscles, and with restriction of normal VT expansion during exercise. There is a strong correlation between the intensity of dyspnea at a standardized point during exercise, the end-expiratory lung volume, and the increased ratio of inspiratory effort to volume displacement (i.e., esophageal pressure relative to maximum: Vt as a % of predicted VC). This increased effort–displacement ratio in COPD crudely reflects the neuromechanical dissociation of the respiratory system that arises as a result of hyperinflation. The corollary of this is that any intervention that reduces end-expiratory lung volume will improve effort–displacement ratios and alleviate dyspnea. In flow-limited patients, bronchodilators act by improving dynamic airway function, thus enhancing lung emptying and reducing lung hyperinflation. Long-acting bronchodilators have recently been shown to reduce hyperinflation during both rest and exercise in moderate to severe COPD. This lung deflation allows greater Vt expansion for a given inspiratory effort during exercise with consequent improvement in dyspnea and exercise endurance.

The pathophysiologic hallmark of chronic obstructive pulmonary disease (COPD) is expiratory flow limitation, whereas the most common symptom is dyspnea (the perception of respiratory discomfort). Expiratory flow limitation arises because of the combined effects of reduced elastic lung recoil and increased airways resistance. Dyspnea is the primary symptom limiting exercise in patients with more advanced disease, and often leads to avoidance of activity, with consequent skeletal muscle deconditioning. The relationship between the physiologic impairment—as traditionally measured by FEV1—and the characteristic symptom of COPD, however, is not straightforward. In this review, the evidence that lung hyperinflation provides a mechanistic link between expiratory flow limitation and dyspnea is examined, with a view to explaining how bronchodilators relieve symptoms.

Lung hyperinflation, defined as an abnormal increase in the volume of air remaining in the lungs at the end of spontaneous expiration, is present in COPD because of the effects of increased lung compliance as a result of the permanently destructive changes of emphysema and expiratory flow limitation. Although there are no formal epidemiologic studies, clinical experience suggests that hyperinflation develops slowly and insidiously over many years, similar to the decline in FEV1. Consequently, patients may not perceive the negative results of hyperinflation until the disease is quite advanced, mainly because the respiratory system adapts to the mechanical disadvantages caused by hyperinflation. For instance, the chest wall reconfigures to accommodate the over-distended lungs, and the diaphragm partially preserves its ability to generate pressure during resting breathing despite its shortened operating length (13). However, these compensatory mechanisms quickly become overwhelmed when ventilation rate is acutely increased, for example, during exercise. A patient with severe COPD, faced with a flight of stairs, may only be able to climb four or five steps before experiencing intolerable dyspnea and has to stop.

As expected, the relationship between dyspnea intensity (measured using the modified Borg scale) and V̇o2 (a measure of oxygen demand) during a symptom-limited incremental cycle ergometry test is notably different between patients with COPD and normal subjects (Figure 1) (4). We, and others, have shown that patients with COPD started to experience dyspnea at a much lower V̇o2 than healthy subjects. Furthermore, they experienced severe dyspnea (Borg score, ⩾ 5) and had to stop exercising at a much lower peak symptom-limited V̇o2 than did healthy subjects, who, in comparison, had not yet attained noticeable levels of respiratory difficulty (Borg score, < 0.5).

This difference in the sensory experience of exertional dyspnea in patients with COPD compared with normal subjects is accompanied by changes in ventilatory mechanics. Compared with normal subjects, patients with COPD have a heightened ventilatory response to exercise (Figure 2), reflecting greater ventilation–perfusion abnormalities (high fixed physiologic deadspace). In general, the breathing pattern in COPD is more rapid and shallow at any given ventilation compared with the breathing pattern in health (Figure 3) (4).

In healthy subjects, end-expiratory lung volume (EELV) and inspiratory capacity (IC) are maintained throughout exercise. During exercise, both the rate and depth (Vt) of respiration are normally increased to accommodate the increased metabolic demand. In COPD, the rate of lung emptying, which is dictated by the product of compliance and resistance (i.e., time constant), is often substantially delayed. In many patients, the expiratory time available during spontaneous resting breathing is insufficient to allow EELV to decrease to its relaxation volume, resulting in lung overinflation. This situation is aggravated further as expiratory time shortens during exercise, resulting in further increases in EELV. This phenomenon has been termed “air trapping,” or dynamic hyperinflation, and refers to the temporary and variable increase in EELV above its baseline value. During exercise, the change in EELV (from rest to peak) in a large population averaged 0.4 L, with considerable variation in the range (5). Changes in EELV are reflected by changes in IC, as total lung capacity (TLC) remains unchanged during exercise.

Although Vt is comparable between groups at rest, it comprises a larger proportion of IC in patients with COPD compared with normal subjects. Because of hyperinflation during the increased demands of exercise, Vt can only increase marginally in patients with COPD, and reaches a plateau (Figures 3 and 4). At this point, further increases in ventilation can only be achieved by increasing breathing frequency, which unfortunately rebounds to cause greater hyperinflation in a vicious cycle.

In addition, hyperinflation markedly increases the tidal inspiratory pressure or effort (expressed as a percentage of maximal inspiratory pressure [Pimax]) required to generate an increase in Vt (expressed as a percentage of the VC) in patients with COPD compared with normal subjects (Figure 4) (4). This is because the relationship between pleural pressure and lung volume during a static maneuver from TLC to residual volume (RV) is sigmoidal rather than linear (Figure 5). In healthy subjects, breathing at rest and during exercise takes part in the central linear portion of the pressure–volume relationship, which means that relatively small changes in tidal pressure will produce comparatively large changes in Vt. In younger individuals, EELV actually declines (IC increases) as expiratory muscles are recruited during exercise, thus conveying a distinct mechanical advantage (Figure 5). In COPD, Vt must “cycle” close to the TLC on the steep portion of the pressure–volume curve, where higher pressures are required for any given volume expansion. At end-expiration in health, the chest wall recoil is outwardly directed and, therefore, assists the inspiratory muscles in thoracic displacement. When Vt is positioned close to TLC in COPD, the inspiratory muscles must contend with the increased lung elastic recoil pressure at end-expiration. In addition, at end-expiration, the chest wall, in contrast to health, is inwardly directed in an expiratory direction, opposing the action of the inspiratory muscles. Combined inward recoil of the lung and chest wall at end-expiration essentially leads to an inspiratory threshold load on the inspiratory muscles, which must be overcome before respired flow is reversed. This threshold load can be substantial during exercise in COPD. Consequently, increasingly higher fractional tidal inspiratory pressures (or effort) must be generated for any given Vt expansion. Breathing at the upper part of the pressure–volume curve weakens the inspiratory muscles, which are not designed to function near TLC, and undoubtedly contributes to both the intensity and the quality of dyspnea.

Dyspnea includes a number of qualitatively distinct sensations that vary in intensity, and differ between patients with COPD and healthy subjects. In a study designed to compare the qualitative aspects of dyspnea in patients with COPD and age-matched healthy subjects, study participants were asked to perform a symptom-limiting exercise test and then describe their sensation of dyspnea using qualitative descriptors (4). Both healthy subjects and patients with COPD chose descriptors of “increased work/effort” and “heaviness” of breathing; however, only patients with COPD consistently chose descriptors denoting “unsatisfied inspiratory effort” (i.e., “can't get enough air in”), “inspiratory difficulty,” and “shallow breathing.”

The distinctive qualitative sensations of dyspnea in patients with COPD suggest that they receive altered peripheral sensory afferent information from multiple mechanoreceptors in the ventilatory muscles, chest wall, lung, and airways, which indicates to them that the mechanical response of the ventilatory system is insufficient or inappropriate for the effort expended. The sense of heightened effort is believed to be conveyed via corollary discharge relayed from the motor cortex to the sensory cortex in the forebrain. In normal subjects, both at rest and during exercise, there is a harmonious matching of effort to ventilatory output. This is because Vt is positioned on the linear part of the pressure–volume curve (Figures 4 and 5).

In health, inspiratory effort increases as ventilation rises during exercise, but the perception of unsatisfied effort is rarely reported. In contrast, in patients with COPD, there is increasing disparity between effort and ventilatory output (or neuromechanical dissociation) as exercise progresses, because Vt is positioned in the extreme upper nonlinear part of the pressure–volume curve due to dynamic hyperinflation. Although patients with COPD try to meet ventilatory demand by increasing breathing frequency, Vt expansion is constrained by the progressive encroachment of EELV from below and the finite TLC from above. Therefore, they experience intolerable dyspnea very quickly during exercise, as there is “no room to breathe.” This so-called “intolerable dyspnea threshold” seems to be at the level at which the inspiratory reserve volume (i.e., the difference between TLC and EELV) approaches less than 500 ml (Figures 4 and 5) (6).

Increased neural drive has previously been linked to the perception of dyspnea, both in asthma and in COPD (79). Accumulating evidence suggests that neuromechanical dissociation due to dynamic hyperinflation is the basis for the inspiratory difficulties experienced by patients with COPD during exercise. For instance, Borg score at a standardized V̇o2 during a symptom-limited incremental cycle ergometry test is strongly correlated with pressure/Pimax:Vt/VC ratio (r = 0.86, p < 0.001) and EELV/TLC (r = 0.69, p < 0.001; Figure 6) (4). The pressure/Pimax:Vt/VC ratio is also strongly related to EELV/TLC at a standardized V̇o2 (r = 0.78, p < 0.001). These correlations have been confirmed in further studies (5, 1012). The corollary of these associations is that an improved ability to expand Vt (due to a reduction in EELV) for a given muscular effort should ameliorate dyspnea.

A number of strategies have been shown to reduce hyperinflation in patients with COPD. As mentioned previously, in flow-limited patients with COPD, the extent of dynamic hyperinflation depends on the prevailing level of ventilation. Therefore, interventions, such as oxygen therapy and exercise rehabilitation, which reduce ventilatory demand, may also reduce the rate of air trapping and dynamic hyperinflation (see article by Casaburi and Porszasz in this issue, pp. 185–189) (1315). Breathing techniques, such as pursed-lip breathing (16), also help to deflate the lungs. Lung volume reduction surgery is another obvious approach to reducing hyperinflation (17, 18). In addition, continuous positive airway pressure and noninvasive pressure support may counteract the negative effects of hyperinflation on the inspiratory muscles (19, 20). The scope of this review, however, is limited to the role of bronchodilators in reducing air trapping and hyperinflation.

Bronchodilators work by improving dynamic airway function, allowing improved lung emptying with each breath. Therefore, the time constant for lung emptying is shortened because airways resistance is diminished. This permits the patient to achieve the required alveolar ventilation during rest and exercise at a lower operating lung volume and, thus, at a lower oxygen cost of breathing. In other words, Vt is positioned at a lower operating lung volume (i.e., EELV is decreased and IC is increased). By deflating the lungs, bronchodilators effectively improve ventilatory muscle performance through greater Vt expansion. Exercise can proceed for a longer duration before the mechanical limitation to ventilation (i.e., critically low inspiratory reserve volume) is reached.

What effect do these improvements in lung mechanics have on dyspnea? We conducted a mechanistic crossover study in 23 patients, in which exertional dyspnea was measured using the modified Borg scale during a constant work-rate cycle ergometry test at 75% of maximal work capacity after a 2-wk treatment with the long-acting β2-agonist salmeterol (50 μg) or placebo (6). Dyspnea intensity at the point of symptom limitation was not different between groups (somewhat severe to severe); however, patients in the salmeterol group exercised for longer before reaching the same degree of dyspnea as patients in the placebo group. The reason for this improvement was that, at rest, as well as throughout exercise, IC was significantly greater and EELV was significantly reduced in the salmeterol group versus the placebo group. This reduction in hyperinflation permitted patients taking salmeterol to significantly increase Vt throughout exercise compared with those on placebo. The increased ability to expand Vt after bronchodilators correlated closely with improved dyspnea ratings at a standardized time during exercise.

In a recent mechanical study on the effects of the long-acting anticholinergic tiotropium (21), relief of exertional dyspnea was closely associated with the improvement in the ratio of expiratory effort to Vt, thus supporting the notion that neuromechanical coupling, as a result of dynamic hyperinflation, forms the basis (at least in part) for the quality and intensity of dyspnea during exercise in COPD.

Accumulating evidence suggests that sustained pharmacologic lung volume reduction with modern-day pharmacotherapy translates into improvements in dyspnea relief and exercise tolerance, as shown in larger-scale clinical trials. For example, a 4-wk treatment with tiotropium has been shown to significantly reduce trough (predose) FRC (a measure of lung hyperinflation) by 0.5 L and peak (postdose) FRC by 0.7 L compared with placebo in 81 patients with severe COPD (22).

Recently, the effects of tiotropium (18 μg once daily) versus placebo on measures of exercise tolerance were compared in 187 patients with severe COPD (23). The results of this 6-wk study showed that tiotropium provided a significant and sustained reduction in air trapping at rest and during exercise, which allowed for greater Vt expansion during exercise. The effects of these improvements in lung volume were reflected by significant increases in exercise tolerance. Furthermore, these improvements in air trapping and exercise tolerance were associated with reductions in exertional dyspnea. At a standardized time point during exercise, for example, Borg dyspnea scores were significantly reduced from baseline in the tiotropium group compared with the placebo group. This allowed patients in the tiotropium group to exercise for longer before their dyspnea became intolerable. These results have been confirmed in a further study (24). The results from these studies on the effects of long-acting bronchodilators support the hypothesis that lung hyperinflation, dyspnea, and exercise intolerance are closely linked in COPD.

Although expiratory flow limitation is the pathophysiologic hallmark of COPD, the main consequence of this is lung hyperinflation. During activity, acute-on-chronic hyperinflation has serious sensory and mechanical consequences. This dynamic hyperinflation likely contributes to both the intensity and the distinct qualitative sensations of dyspnea, particularly the distressing feeling of unsatisfied inspiration. Bronchodilator therapy relieves dyspnea by deflating the lungs, which reduces the elastic load on the inspiratory muscles. This, together with the reduction of resistive work, improves the functional performance of the inspiratory muscles. The attendant increased ability to expand Vt contributes to enhanced neuromechanical coupling of the respiratory system. Sustained pharmacologic volume deflation has the potential to impact positively on the important patient-centered outcomes of dyspnea and activity limitation in the long term.

1. Gorman RB, McKenzie DK, Pride NB, Tolman JF, Gandevia SC. Diaphragm length during tidal breathing in patients with chronic obstructive pulmonary disease. Am J Respir Crit Care Med 2002;166:1461–1469.
2. Similowski T, Yan S, Gauthier AP, Macklem PT, Bellemare F. Contractile properties of the human diaphragm during chronic hyperinflation. N Engl J Med 1991;325:917–923.
3. Cassart M, Pettiaux N, Gevenois PA, Paiva M, Estenne M. Effect of chronic hyperinflation on diaphragm length and surface area. Am J Respir Crit Care Med 1997;156:504–508.
4. O'Donnell DE, Bertley JC, Chau LK, Webb KA. Qualitative aspects of exertional breathlessness in chronic airflow limitation: pathophysiologic mechanisms. Am J Respir Crit Care Med 1997;155:109–115.
5. O'Donnell DE, Revill SM, Webb KA. Dynamic hyperinflation and exercise intolerance in COPD. Am J Respir Crit Care Med 2001;164:770–777.
6. O'Donnell DE, Voduc N, Fitzpatrick M, Webb KA. Effect of salmeterol on the ventilatory response to exercise in chronic obstructive pulmonary disease. Eur Respir J 2004;24:86–94.
7. Bellofiore S, Ricciardolo FL, Ciancio N, Sapienza MA, Patane A, Mistretta A, Di Maria GU. Changes in respiratory drive account for the magnitude of dyspnoea during bronchoconstriction in asthmatics. Eur Respir J 1996;9:1155–1159.
8. Burki NK. Breathlessness and mouth occlusion pressure in patients with chronic obstruction of the airways. Chest 1979;76:527–531.
9. Lahrmann H, Wild M, Wanke T, Tschernko E, Wisser W, Klepetko W, Zwick N. Neural drive to the diaphragm after lung volume reduction surgery. Chest 1999;116:1593–1600.
10. Martinez FJ, de Oca MM, Whyte RI, Stetz J, Gay SE, Celli BR. Lung-volume reduction improves dyspnea, dynamic hyperinflation, and respiratory muscle function. Am J Respir Crit Care Med 1997;155:1984–1990.
11. Di Marco F, Milic-Emili J, Boveri B, Carlucci P, Santus P, Casanova F, Cazzola M, Centanni S. Effect of inhaled bronchodilators on inspiratory capacity and dyspnoea at rest in COPD. Eur Respir J 2003;21:86–94.
12. Man WD, Mustfa N, Nikoletou D, Kaul S, Hart N, Rafferty GF, Donaldson N, Polkey MI, Moxham J. Effect of salmeterol on respiratory muscle activity during exercise in poorly reversible COPD. Thorax 2004;59:471–476.
13. O'Donnell DE, Bain DJ, Webb KA. Factors contributing to relief of exertional breathlessness during hyperoxia in chronic airflow limitation. Am J Respir Crit Care Med 1997;155:530–535.
14. O'Donnell DE, D'Arsigny C, Webb KA. Effects of hyperoxia on ventilatory limitation during exercise in advanced chronic obstructive pulmonary disease. Am J Respir Crit Care Med 2001;163:892–898.
15. Somfay A, Porszasz J, Lee SM, Casaburi R. Dose-response effect of oxygen on hyperinflation and exercise endurance in nonhypoxaemic COPD patients. Eur Respir J 2001;18:77–84.
16. Bianchi R, Gigliotti F, Romagnoli I, Lanini B, Castellani C, Grazzini M, Scano G. Chest wall kinematics and breathlessness during pursed-lip breathing in patients with COPD. Chest 2004;125:459–465.
17. Ciccone AM, Meyers BF, Guthrie TJ, Davis GE, Yusen RD, Lefrak SS, Patterson GA, Cooper JD. Long-term outcome of bilateral lung volume reduction in 250 consecutive patients with emphysema. J Thorac Cardiovasc Surg 2003;125:513–525.
18. Ingenito EP, Loring SH, Moy ML, Mentzer SJ, Swanson SJ, Reilly JJ. Physiological characterization of variability in response to lung volume reduction surgery. J Appl Physiol 2003;94:20–30.
19. O'Donnell DE, Sanii R, Younes M. Improvement in exercise endurance in patients with chronic airflow limitation using CPAP. Am Rev Respir Dis 1988;138:1510–1514.
20. Keilty SE, Ponte J, Fleming TA, Moxham J. Effect of inspiratory pressure support on exercise tolerance and breathlessness in patients with severe stable chronic obstructive pulmonary disease. Thorax 1994;49:990–994.
21. O'Donnell DE, Webb KA. The effects of tiotropium on ventilatory mechanics during exercise in COPD. Am J Respir Crit Care Med 2005;2:A494.
22. Celli B, ZuWallack R, Wang S, Kesten S. Improvement in resting inspiratory capacity and hyperinflation with tiotropium in COPD patients with increased static lung volumes. Chest 2003;124:1743–1748.
23. O'Donnell DE, Flüge T, Gerken F, Hamilton A, Webb K, Aguilaniu B, Make B, Magnussen H. Effects of tiotropium on lung hyperinflation, dyspnea and exercise tolerance in patients with COPD. Eur Respir J 2004;23:832–840.
24. Maltais F, Hamilton A, Marciniuk D, Hernandez P, Sciurba F, Richter K, Kesten S, O'Donnell DE. Improvements in symptom-limited exercise performance over eight hours with once-daily tiotropium in patients with COPD. Chest 2005;128:1168–1178.
Correspondence and requests for reprints should be addressed to Denis E. O'Donnell, M.D., F.R.C.P.(I), F.R.C.P.(C), Division of Respiratory and Critical Care Medicine, Respiratory Investigation Unit, Department of Medicine, Queen's University, 102 Stuart Street, Kingston, ON, K7L 2V6 Canada. E-mail:


No related items