Abstract
Asthma evokes several uncomfortable sensations including increased “effort to breathe” and chest “tightness.” We have tested the hypotheses that “effort” and “tightness” are due to perception of increased work performed by the respiratory muscles. Bronchoconstriction was induced by inhaled methacholine in 15 subjects with mild asthma (FEV1/FVC baseline = 81.9% ± 5.8; bronchoconstriction = 64.0% ± 8.6). To relieve the work of breathing, and thereby minimize activation of respiratory muscle afferents and motor command, subjects were mechanically ventilated. Subjects separately rated effort to breathe and tightness during mechanical ventilation and during spontaneous breathing. Bronchoconstriction produced elevated end-expiratory lung volume (279 ± 62 ml); in a control study, end-expiratory lung volume was increased equally in the absence of bronchoconstriction by increasing end-expiratory pressure. During bronchoconstriction, ratings of effort were greater during spontaneous breathing than during mechanical ventilation (p < 0.05). Ratings of tightness were unchanged by the absence of respiratory muscle activity (p = 0.12). Hyperinflation alone did not produce tightness or effort. We conclude that tightness is not related to the increase in respiratory work during bronchoconstriction.
Keywords: asthma; dyspnea; work; respiratory muscles
There are several sensations that are commonly associated with asthma (1-3), including the sense of increased “effort” or “work” of breathing and also air hunger, which are common to many respiratory disorders. But the sensation of chest “tightness” is unique to conditions associated with bronchospasm (1, 3, 4) and is one of the most common phrases used by patients with asthma to describe their symptoms (5).
The sense of “effort/work” to breathe has been ascribed to perception of the motor drive to the respiratory muscles (corollary discharge hypothesis) (6); effort/work may also be mediated by muscle afferents. In asthma, increased airway resistance and hyperinflation (to stiffer parts of the pressure– volume curve) lead to increased work of breathing (7). Furthermore, as a consequence of hyperinflation, the inspiratory muscles are shortened, so the muscle is less effective and more motor drive is required to perform the same work. The dyspnea of asthma is commonly ascribed to sensations of work or effort (8, 9). However, mechanical ventilation, which ought to relieve the work of breathing, rarely ameliorates the unpleasant sensations associated with severe asthma (10, 11) in contrast to its effectiveness in other respiratory diseases.
Several indirect lines of evidence suggest that the physiological origins of the sensation of chest “tightness” cannot be explained by increased work of breathing. When respiratory work is increased by loaded breathing, the sensations produced are qualitatively (4, 12) and quantitatively (13) different in comparison to bronchoconstriction, suggesting sensations of asthma are not simply a perception of increased resistive work. Furthermore, β-agonists reduce the sensation of chest tightness disproportionately to the improvement in lung function and work of breathing (14). So it appears that effort and tightness are mediated by different mechanisms. Others have suggested the sense of tightness seen with acute bronchospasm may emanate from pulmonary receptors (4, 12, 14, 15).
We tested the hypothesis that active contraction of respiratory muscles is essential for the generation of tightness during methacholine-induced bronchoconstriction. If the tightness of asthma arises from increased mechanical load and motor output, institution of mechanical ventilation would relieve the sensation. Alternatively, if tightness arises from pulmonary receptors, we would not expect it to be reduced by mechanical assistance. We assessed the intensity of the sensations of both chest tightness and effort of breathing during positive pressure ventilation versus spontaneous breathing during bronchoconstriction. We chose methacholine to induce bronchoconstriction because it has been reported to induce sensations similar to those felt during asthma (4, 16). In separate trials, we examined the role of hyperinflation in the generation of tightness during mechanical ventilation by reproducing, with positive end-expiratory pressure, the changes in end- expiratory volume observed during bronchospasm.
The study was approved by internal review boards at Beth Israel Deaconess Medical and Harvard School of Public Health. Written informed consent was obtained. We informed subjects that they would receive a variety of substances by nebulizer, and that these might have no effect on their lungs, might induce wheezing, or might reduce wheezing. Day 1 was devoted to practice and determination of individual response to inhaled methacholine; data were collected on Day 2.
We studied 15 subjects with mild asthma (age range 20–51 yr; eight males, seven females). No subject had taken β2-agonists for the 24 h preceeding the study and none were using corticosteroids. None had been hospitalized because of asthma within the prior 6 mo. Three subjects were excluded after initial testing: one had no response to methacholine, another had significant bronchoconstriction before starting the experiment, and the third experienced laryngeal stridor.
Before each day's testing, subjects described sensations experienced during typical episodes of asthma. All recalled experiencing both tightness and effort. Following this, the subject chose three phrases from the list shown in Figure 1 to best describe his or her spontaneous asthma experience.
Fig. 1. Descriptors chosen to describe sensations experienced during bronchoconstriction due to asthma and to methacholine.
[More] [Minimize]Following each experiment subjects described the sensations experienced, then chose three phrases from the list shown in Figure 1. They were also asked whether symptoms following methacholine were similar to those experienced during spontaneous asthma.
Subjects in a semirecumbent position were mechanically ventilated via a mouthpiece (Siemens, 900C). Before the experiment subjects practiced relaxing the respiratory muscles during mechanical ventilation. Healthy subjects usually require slight hypocapnia and/or increased tidal volume in order to completely relax the respiratory muscles. Minute ventilation was therefore set to a value about twice normal (0.16 L/kg body weight) and rate was adjusted for comfort. Inspired gas comprised 30% O2 with added CO2, with the balance N2. Because of the increased ventilation, we raised CO2 to maintain end tidal Pco 2 (Pet CO2 ) 3–5 mm Hg below previously measured resting values. We inspected airway pressure during inspiration and flow during expiration to ascertain whether respiratory muscles were relaxed (17).
Subjects performed several 6-min trials on the ventilator. During the third and fourth minutes, a nebulizer (Misty Neb; Allegiance Healthcare, IL) in the inspiratory limb of the ventilator circuit delivered progressively more concentrated doses of methacholine with each succeeding trial (see Figure 2). After each trial, FEV1 and inspiratory capacity (IC) were measured. The procedure continued until FEV1 fell to ⩽ 50% predicted, subjects rated more than moderate tightness or effort (see below), or there was evidence of respiratory muscle contraction. We then administered 2.5 mg of nebulized albuterol; the subject was told only that this was the “next dose.”
Fig. 2. Summary diagram of one trial (saline); scores of tightness were less than 3% of full scale after saline inhalation. Subjects were asked to rate both “effort” (E ) and “tightness” (T ) at the end of each minute once they had become relaxed on the ventilator, and they scored both sensations again once they resumed spontaneous breathing. The methacholine dose order was saline (diluent), 0.025 mg, 0.125 mg, 0.25 mg, 1.25 mg, 2.50 mg, 5.00 mg, 7.50 mg, and 10.0 mg. Changes in EEV were calculated from the means of five breaths before the first and last scores of each trial.
[More] [Minimize]In an additional trial positive end-expiratory pressure (PEEP) was increased to reproduce the hyperinflation measured during the highest dose of methacholine (with nebulized saline to simulate a similar procedure). The increase in end-expiratory volume during bronchoconstriction was estimated from the reduction in inspiratory capacity; baseline static lung compliance (calculated from ventilator pressure and volume measurements) was used to estimate the amount of PEEP required to produce an equivalent level of hyperinflation in the absence of bronchoconstriction.
Tidal volume and end-expiratory volume (EEV) were measured with Respitrace bands placed around the rib cage and abdomen, calibrated by a previously described method (18). Tidal O2 and CO2 were measured by mass spectroscopy (1100 Medical Gas analyzer, Perkin-Elmer, Pomona, CA). Arterial O2 saturation and heart rate were estimated using an infrared pulse oximeter probe placed on the index finger (Biox 3740; Omeda, Louisville, CO). Airway pressure (Paw), was measured by a transducer integral to the ventilator, calibrated according to manufacturer's instructions.
Respiratory sensations were rated on a 100-mm visual analogue scale (VAS), implemented by an LED strip controlled by the subject with a linear potentiometer. The ends of the VAS were marked “None” (indicating no sensation) and “Extreme” (an intolerable amount). Before the start of the experiment subjects placed the words “Slight,” “Moderate,” and “Severe” at points they thought appropriate in between the ends of the VAS.
All subjects had volunteered the word “Tightness” in describing their spontaneous asthma; we instructed them to rate the tightness sensation they recalled from their experience with asthma. In preliminary experiments we found that the concept of respiratory effort or work was more confusing to subjects in the context of mechanical ventilation. This sensation was therefore defined explicitly using the following script since subjects in prior studies have demonstrated a variety of interpretations of the phrase (19, 20). “One of the sensations we will ask you to rate during this study is the sense of ‘effort or work of breathing.' By this, we mean the effort or work you are expending with your breathing muscles to inflate your chest. To help you understand this, think of the following analogy with your arm muscles. If you lift your arm, you must make an effort and do some work with your muscles. If you lift your arm while you are holding a weight, you require even more effort or work. If I lift your arm for you, you do not need to make an effort or do work with your muscles.” Although many physiologists conceive of respiratory work and effort as two different quantities, in our experience naive subjects are unable to make this distinction (20).
Mechanical ventilation was started and sustained until the subject became reaccustomed to the procedure. Trials were started after the subject achieved respiratory muscle relaxation as indicated by the inspiratory pressure trace (21). Subjects rated sensations of tightness and effort every minute at times shown in Figure 2. The experiment comprised several consecutive trials (the number of methacholine doses depended on each subject's response). In each trial the dose (of placebo, methacholine, or albuterol) was given during mechanical ventilation, followed by a period during which subjects rated sensations during mechanical ventilation. Mechanical ventilation was then stopped, and subjects again rated sensation after two to four breaths of spontaneous breathing (through the mouthpiece alone, with no inspired CO2) after mechanical ventilation was stopped. Pulmonary function testing was then performed.
The PEEP hyperinflation trial was performed either before the methacholine inhalation challenge (five subjects) or following reversal of bronchoconstriction (seven subjects). Two subjects (4 and 7) failed to complete this trial because they did not adequately relax during PEEP. If the hyperinflation trial was done before the administration of methacholine, the target for the change in EEV was based on data from the practice day.
Inhalation of methacholine gave rise to sensations of both “tightness” and “effort” in all subjects. Initial analysis (ANOVA) showed a difference in the effect mechanical ventilation had upon the two sensations. There was no significant difference in the tightness ratings made on or off the ventilator (mean difference = 6% of VAS; p = 0.12; paired t test, Figure 3). Effort ratings were greater during spontaneous breathing than during mechanical ventilation (mean difference = 16% of VAS; p = 0.0032, paired t test; Figure 3).
Fig. 3. Sensations of tightness (left panel) and effort to breathe (right panel) made during mechanical and spontaneous ventilation with the highest level of bronchoconstriction. Individual's data are represented by the dashed lines; the group mean and SEM. (n = 12) are represented by the solid lines. *Significant change in sensation between ventilatory states.
[More] [Minimize]Mean baseline FEV1/FVC was 81.9% (SD ± 5.8%). Mean FEV1/FVC after the highest dose of methacholine was 64% (SD ± 8.6%). Only data where FEV1 had been reduced by at least 15% of baseline values (n = 24) were used in analysis of sensations to prevent the possibility that ratings of zero sensation would skew the data. (Results were similar when data were screened using sensation instead of using FEV1.)
Peak airway pressure was slightly more variable after bronchoconstriction. The average standard error of peak pressure was increased after the highest dose of methacholine (1.312 cm H2O) compared with that after saline (0.864 cm H2O), but this difference was not significant (p = 0.10, paired t test). Only 7 of 75 trials showed significantly greater levels of variation of peak Paw after methacholine compared with saline (p < 0.05, Bartlett's test). Because this suggests increased inspiratory muscle activity, these seven trials were not included in further analysis.
The last 10 breaths of each trial were examined for evidence of “expiratory braking,” that is, inspiratory efforts made during expiratory flow (Figure 4). Expiratory braking was seen in some breaths in 11 of the 12 subjects. Linear regressions demonstrated no relationship between the duration or magnitude of these inspiratory efforts with either the level of bronchoconstriction or tightness rating.
Fig. 4. Examples of expiratory flow during mechanical ventilation from two subjects. Both panels include the last 10 breaths of a trial with similar levels of bronchoconstriction being achieved. The left panel shows the subject remained relaxed; the right panel shows brief episodes of “expiratory braking” by active inspiratory efforts.
[More] [Minimize]Mean Pet CO2 (± SEM) during mechanical ventilation was 32.9 ± 0.2 mm Hg; this fell significantly within the first five breaths of spontaneous breathing to 30.5 ± 0.6 mm Hg (p < 0.05).
The increase in EEV (measured by Respitrace) was not significantly different (ANOVA) in the three conditions: following the highest dose of methacholine on the ventilator (279 ± 62 ml), during spontaneous breathing (381 ± 88 ml), and during PEEP without bronchoconstriction (532 ± 118 ml). Eleven subjects completed the trial with additional PEEP; one subject was unable to remain relaxed during periods of elevated PEEP. Bronchoconstriction caused significantly higher tightness ratings than PEEP (p < 0.001; t test; Figure 5). Only one subject rated tightness significantly above None during hyperinflation with increased PEEP, and he said that the sensation was different from the tightness he felt during bronchoconstriction.
Fig. 5. Tightness during mechanical ventilation and hyperinflation caused by bronchoconstriction and elevated PEEP. Data from individual subjects are represented by dashed lines; group mean data (± SEM; n = 11) are shown by the solid line. Tightness was significantly higher in the presence of bronchoconstriction.
[More] [Minimize]Nine subjects said the intensity of tightness they felt during bronchoconstriction was the same during mechanical ventilation as it was during spontaneous breathing. The remaining three said that tightness was more intense when mechanical ventilation stopped (subjects 4, 9, and 11). Subjects 5 and 12 described a more diffuse distribution of the feeling of tightness during spontaneous breathing compared with a more localized, substernal sensation during mechanical ventilation, but they did not report a change in the intensity between spontaneous and mechanical ventilation.
Eight subjects described the effort to breathe as greater during spontaneous breathing than during mechanical ventilation during the highest level of bronchoconstriction. Subject 9 said more effort was needed during mechanical ventilation in order “to breathe against the ventilator” and consequently scored effort as lower during spontaneous breathing. Subject 6 said there was no difference in effort between the two states and scored as such.
The three most commonly chosen descriptive phrases were the same for asthma and methacholine-induced bronchoconstriction (Figure 1). These were, “My chest feels tight,” “I feel I cannot take a deep breath,” and “My breathing requires effort.” During the debriefing all subjects described the sensations they felt during the methacholine challenge as “similar to” or “the same as” those felt during an episode of asthma.
Methacholine-induced bronchoconstriction produced chest tightness and an increased sense of effort associated with breathing. The intensity of perceived effort to breathe was significantly reduced by mechanical ventilation. In contrast, the sensation of chest tightness or constriction was unchanged by mechanical ventilation.
These data are not consistent with the hypothesis that the perception of tightness in subjects with asthma requires active contraction of respiratory muscles. The present data demonstrate that tightness is unaffected by the reduction or removal of respiratory muscle work and, thus, is not generated by afferent or efferent activity associated with respiratory muscles.
We examined peak pressure variability to determine whether the respiratory muscles were relaxed. As expected with concurrent changes in lung mechanics, absolute inspiratory pressure increased with increasing bronchoconstriction. However, we did not detect a significant change in peak Paw variability, suggesting that respiratory muscles were relaxed. Despite this evidence, it is possible that there was residual activity within the respiratory muscles. There was evidence of inspiratory muscle activity during exhalations; this activity reduced expiratory flow and may have contributed to the increases in end-expiratory lung volume observed, but it was not related to either the level of bronchoconstriction or the magnitude of tightness sensation. Electromyography (EMG) could have been used to assess respiratory muscle activity. However, the use of airway pressure signals to assess respiratory muscle activity has been shown to be as good as EMG (21).
Although there may have been minor respiratory muscle contractions in some subjects, the work of breathing was dramatically reduced by ventilation. If tightness depended on respiratory muscle afferents, it should likewise have been reduced. No such effect was observed. Mechanical ventilation did reduce the sense of effort to breathe, consistent with the reduction in the work of breathing.
Normal subjects breathing at higher lung volumes also give reports of increased effort and dyspnea (22). At higher lung volumes the respiratory muscles are shortened and mechanically less effective, and lung compliance is reduced so more effort is required to breathe. It has been shown that hyperinflation associated with bronchoconstriction induces dyspnea, however the possibility that more than one sensation might arise during bronchoconstriction was not addressed (9). The present experiments show that hyperinflation is not the stimulus that induces tightness, however hyperinflation probably does induce a perception of increased effort.
Our findings are consistent with the hypothesis that tightness is caused by stimulation of airway receptors that respond to bronchospasm. Slowly adapting receptors are excited by contraction of airway smooth muscle; rapidly adapting (irritant) receptors and C-fibers may respond to local inflammation of the airways (reviewed in [23]). Support for this hypothesis comes from the fact that airway anesthesia with lidocaine reduces the sensations of chest irritation or tightness associated with bronchoconstriction (12, 15), leading to the conclusion that vagal activity contributes to the sensation. The present study provides novel support for the notion that pulmonary receptors are important in generating symptoms of asthma.
In summary, the role of respiratory muscles in the generation of the sensations associated with asthma has been controversial. The present data are consistent with the hypothesis that muscle afferents or corollary discharge play a role in the sensation of effort during bronchoconstriction. As such, increased muscle work may contribute to the discomfort and distress of asthma during spontaneous breathing. On the other hand, tightness of asthma does not appear to require respiratory muscle activity and is not relieved significantly by the institution of mechanical ventilation. Changes within the airway itself and the associated activity of pulmonary receptors provide an attractive alternative mechanism for the generation of tightness and the persistent respiratory discomfort following institution of mechanical ventilation in patients with status asthmaticus.
The authors would also like to thank Robert Lansing and Robert Brown for help in planning the study. We would also like to acknowledge Deborah Kinsman for technical assistance.
Supported by NIH Grant HL57916 and the Fanny S. Bienenstok Endowment Fund for Asthma.
| 1. | Simon PM, Schwartzstein RM, Weiss JW, Fencl V, Teghtsoonian M, Weinberger SEDistinguishable types of dyspnea in patients with shortness of breath. Am Rev Respir Dis142199010091014 |
| 2. | Elliott MW, Adams L, Cockcroft A, MacRae KD, Murphy K, Guz AThe language of breathlessness. Use of verbal descriptors by patients with cardiopulmonary disease. Am Rev Respir Dis1441991826832 |
| 3. | Mahler DA, Harver A, Lentine T, Scott JA, Beck K, Schwartzstein RMDescriptors of breathlessness in cardiorespiratory diseases. Am J Respir Crit Care Med154199613571363 |
| 4. | Moy ML, Weiss WJ, Sparrow D, Israel E, Schwartzstein RMQuality of dyspnea in bronchoconstriction differs from external resistive loads. Am J Respir Crit Care Med1622000451455 |
| 5. | Killian KJ, Watson R, Otis J, St Amand TA, O'Bryne PMSymptom perception during acute bronchoconstriction. Am J Respir Crit Care Med1622000490496 |
| 6. | Gandevia SC, Killian KJ, Campbell EJThe effect of respiratory muscle fatigue on respiratory sensations. Clin Sci601981463466 |
| 7. | Martin J, Powell E, Shore S, Emrich J, Engel LAThe role of respiratory muscles in the hyperinflation of bronchial asthma. Am Rev Respir Dis1211980441447 |
| 8. | Chapman KR, Rebuck ASInspiratory and expiratory resistive loading as a model of dyspnea in asthma. Respiration441983425432 |
| 9. | Lougheed DM, Webb KA, O'Donnell DEBreathlessness during induced lung hyperinflation in asthma: the role of the inspiratory threshold load. Am J Respir Crit Care Med1521995911920 |
| 10. | Corbridge TC, Hall JBThe assessment and management of adults with status asthmaticus. Am J Respir Crit Care Med151199512961316 |
| 11. | Hall GT, Gardner TD, Potter EKAttenuation of long-lasting effects of sympathetic stimulation after repeated stimulation. Circ Res671990193198 |
| 12. | Taguchi O, Kikuchi Y, Hida W, Iwase N, Satoh M, Chonan T, Takishima TEffects of bronchoconstriction and external resistive loading on the sensation of dyspnea. J Appl Physiol71199121832190 |
| 13. | Kelsen SG, Prestel TF, Cherniack NS, Chester EH, Deal ECComparison of the respiratory responses to external resistive loading and bronchoconstriction. J Clin Invest67198117611768 |
| 14. | Moy ML, Lantin ML, Harver A, Schwartzstein RMLanguage of dyspnea in assessment of patients with acute asthma treated with nebulized albuterol. Am J Respir Crit Care Med1581998749753 |
| 15. | Petit J. Sensory innervation of the airways. In: Porter R, editor. Breathing: Hering-Breuer centenary symposium. London: J & A Churchill; 1969, p. 111–114. |
| 16. | Rubinfeld AR, Pain MCConscious perception of bronchospasm as a protective phenomenon in asthma. Chest721977154158 |
| 17. | Prechter GC, Nelson SB, Hubmayr RDThe ventilatory recruitment threshold for carbon dioxide. Am Rev Respir Dis1411990758764 |
| 18. | Banzett RB, Mahan ST, Garner DM, Brughera A, Loring SHA simple and reliable method to calibrate respiratory magnetometers and Respitrace. J Appl Physiol79199521692176 |
| 19. | Lansing RW, Im BS, Thwing JI, Legedza AT, Banzett RBThe perception of respiratory work and effort can be independent of the perception of air hunger. Am J Respir Crit Care Med162200016901696 |
| 20. | Moosavi SH, Topulos GP, Hafer A, Lansing RW, Adams L, Brown R, Banzett RBAcute partial paralysis alters perceptions of air hunger, work and effort at constant PCO2 and Ve. Respir Physiol12220004560 |
| 21. | Ward ME, Corbeil C, Gibbons W, Newman S, Macklem PTOptimization of respiratory muscle relaxation during mechanical ventilation. Anesthesiology6919882935 |
| 22. | Killian KJ, Gandevia SC, Summers E, Campbell EJEffect of increased lung volume on perception of breathlessness, effort, and tension. J Appl Physiol571984686691 |
| 23. | Paintal ASVagal sensory receptors and their reflex effects. Physiol Rev531973159227 |
