INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Dynamic hyperinflation has been defined as a dynamic increase in end-expiratory lung volume (EELV) and is considered to be one of the most important respiratory mechanical abnormalities in patients with chronic obstructive pulmonary disease (COPD) (1). Although dynamic hyperinflation has been frequently observed in patients with COPD not only during exercise (2) but also during resting breathing (5), its effects on breathing have not been studied sufficiently for the following reasons: (1) Patients with dynamic hyperinflation almost always have other severe respiratory mechanical abnormalities, so that the effects of dynamic hyperinflation cannot be easily isolated for study; (2) dynamic hyperinflation results in intrinsic positive end-expiratory pressure (PEEPi) that imposes an inspiratory threshold load (ITL) on the inspiratory muscles. Meanwhile, the increase in EELV itself puts the inspiratory muscles at a mechanical disadvantage. How these two factors, by increasing the demand on the inspiratory muscles while decreasing their capacity, separately affect breathing and its control remains undefined. It is clinically difficult to partition the effects of dynamic hyperinflation on breathing into contributions of PEEPi-imposed ITL and increasing operating lung volume itself.

As PEEPi and increasing EELV are, per se, inspiratory loads, dynamic hyperinflation may present a good example of how a predominantly expiratory load (expiratory airflow limitation in COPD patients) can be converted into a load on the inspiratory muscles. This leads us to think that external ITL and/or the induced acute hyperinflation from positive pressure breathing will impose loads similar to expiratory airflow limitation on the inspiratory muscles in healthy subjects, and therefore may mimic the pathophysiological consequences of dynamic hyperinflation on inspiration. Accordingly, the purpose of the present work was to study the separate and combined effects of external ITL and induced acute hyperinflation on breathing. One may assume that the PEEPi and increase in operating lung volume are interactive in their effects on breathing. Indeed, any change in breathing parameter observed during dynamic hyperinflation presumably reflects this interactive effect. What has not been described is the independent effect of PEEPi-imposed ITL and changes in operating lung volume on breathing. Therefore, the major questions we would like to answer are as follows: Which factorthe imposed ITL, or operating lung volume, or bothis responsible for the increased perception of inspiratory difficulty? If both play a role, which factor is the stronger predictor for the variability of inspiratory difficulty? These questions are clinically relevant because it has been shown that in stable patients with COPD, chronic dyspnea is closely correlated with the existence and severity of expiratory airflow limitation, presumably because the latter inevitably gives rise to dynamic hyperinflation (9).

	METHODS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Study Design

In healthy subjects, the pathophysiological consequences of dynamic hyperinflation, namely the dynamic increase in EELV and the resultant PEEPi, can be brought about by positive end-expiratory pressure (PEEP). When this is combined with the application of continuous positive airway pressure (CPAP) and external inspiratory threshold loading, partition of the effects of the dynamic increase in EELV and the PEEPi-imposed ITL on breathing becomes possible. The rationale for loading the respiratory system by CPAP, PEEP, and external ITL and studying breathing responses that occur during dynamic hyperinflation is summarized in Figure 1. The left panel of Figure 1 shows the schema of the Campbell diagram used in our recent work to measure PEEPi during spontaneous breathing with dynamic hyperinflation (10). The Campbell diagram is composed of the static pressure-volume relationship of the chest wall (chest wall curve) and the relationship between pleural pressure and volume during slow inspiration (lung curve) (11). The horizontal dashed line shows an increase in EELV by dynamic hyperinflation, which intersects the lung and chest wall curves at A and B, respectively. The pressure difference between points B and A represents the elastic recoil pressure of the respiratory system at that volume and therefore equals PEEPi (12). The right panel assumes the same lung and chest wall curves. Without affecting the position of the two curves (13, 14), applying an appropriate PEEP will increase EELV to the level of AB, so that end expiration is at point B and an inspiratory pressure given by BA (PEEPi) is required in order to start inspiratory flow. This situation is exactly the same as that shown in the left panel due to dynamic hyperinflation except for the mechanism by which it is produced. CPAP imposes a comparable, yet different type of load to the inspiratory muscles. Similar to PEEP, application of CPAP increases EELV. However, CPAP requires a less negative pleural pressure throughout inspiration. As shown in Figure 1, applying an appropriate CPAP will also result in an increase in EELV to the level of AB. However, there is no PEEPi hence no ITL that will be present because CPAP moves the inspiratory lung volume-pleural pressure relationship to the right. Thus, CPAP can be used to study the pure effect of increasing operating lung volume (shortening of the inspiratory muscles) on breathing. If a given level of external ITL as given by CD (where CD = AB) is applied, generation of an inspiratory pressure equal to DC is required before inspiratory flow can start because the external ITL moves the lung curve to the left. This situation is similar to that induced by dynamic hyperinflation (left panel) or by PEEP except EELV is not increased. In this way, the effects of only increasing EELV (CPAP) or ITL (external ITL) can be evaluated independently and compared with those influenced by both factors (PEEP), to assess the separate and integrated effect of the imposed ITL and volume load on breathing.

View larger version (15K): [in this window] [in a new window]   Figure 1.   Schematic presentation of the Campbell diagram and the changes in position of the inspiratory lung volume-pleural pressure relationship (lung curve) during application of external ITL, CPAP, and PEEP in healthy subjects.

Subjects

Eight healthy laboratory staff (aged 30-41 yr) from whom informed consent was obtained, volunteered to be the subjects of the study. The scientific purpose of the study was unknown to all but one subject. The study was approved by the research ethics committee of the Montreal Chest Institute Research Center.

Measurements and Procedures

Respiratory flow was measured by a heated Fleisch No. 2 pneumotachograph (Lausanne, Switzerland) attached to a differential pressure transducer (Validyne Corp., Northridge, CA). Esophageal and gastric pressures (Pes and Pga) were measured by two Validyne differential pressure transducers via two separate conventional balloon-catheter systems, which were placed in the lower third of the esophagus and the stomach, respectively. Mouth pressure (Pmo) was measured by an additional pressure transducer (Validyne) through a tube connected directly to the mouthpiece.

The perceived breathing effort with each load was measured by the degree of inspiratory difficulty (15) using the modified Borg scale (16). This required the subjects to give a number between 0 to 10 representing their perceived inspiratory difficulty. The number 0 indicates no difficulty and 10 the maximal difficulty, with each doubling of the numerical reading representing the twofold increase in the perceived sensation (16).

Application of the External ITL, PEEP, and CPAP

The external ITL was applied by a custom-built constant negative pressure system that was applied directly to the inspiratory port of a unidirectional Hans-Rudolph nonrebreathing valve (type 2700; Hans-Rudolph Inc., St. Louis, MO). For each breath, inspiratory muscles have to generate an equal or greater negative pressure in the mouth side of the port in order to initiate inspiratory flow. This system proved to add no additional flow resistance during inspiration and therefore provides a pure inspiratory threshold load (17). The desired PEEP and CPAP were applied by a BiPAP Ventilatory Support System (BiPAP S; Respironics Inc., Murrysville, PA) attached to the breathing circuit.

Protocol

The subjects were seated on an armchair with their head and back firmly supported to keep the body posture unchanged during the experiment. They breathed through a mouthpiece while wearing a noseclip. Before the loading experiments, the subjects were asked to perform respiratory relaxation maneuvers by expiring passively from total lung capacity (TLC) to functional residual capacity (FRC) against a high expiratory resistance, in order to establish the individual static chest wall pressure-volume relationship (PVstat,w). This procedure was performed several times to determine the reproducibility (18).

The loading experiments comprised external ITL of 2.5, 5, 10, 15, 20 cm H₂O and PEEP and CPAP of 2.5, 5, 7.5, 10, 12.5 cm H₂O. Each load lasted 2 min and was preceded by a recording of a short period of quiet breathing. The subjects did not know either the type or the level of the loads that were randomly applied. During each run, only a single level of one type of load was applied. Between the runs, the subjects were allowed a few minutes of rest. Before and at the end of each load, inspiratory capacity (IC) inspirations were performed to measure the change in EELV. Because external ITL may prevent the subjects from reaching the real TLC, during each external ITL run, the load was shut off as soon as the second IC effort was initiated. The IC maneuver was performed by "making a further maximal effort on top of a maximal inspiration" (19, 20). Immediately after each loaded run, the subjects reported the score for their sensation of inspiratory difficulty on the modified Borg scale.

Data Analysis

The flow and pressures were amplified (HP 7758B; Hewlett-Packard, Waltham, MA), digitized using a 12-bit analog-to-digital converter, and recorded at 200 Hz on a desktop computer. Breathing parameters including tidal volume (VT), frequency (f), minute ventilation (E), inspiratory time (TI), duty cycle (TI/Ttot), and mean inspiratory flow (VT/TI) were measured from the flow signal. During external ITL and PEEP loading, inspiratory effort started before the beginning of inspiratory flow, therefore, TI was measured from the beginning of rise in transdiaphragmatic pressure (Pdi, calculated as Pga  Pes) to the end of inspiratory flow (7). The change in EELV for each load was calculated as the "mirror image" of the change in IC. The change in end-inspiratory lung volume (EILV) was calculated as the sum of EELV and VT. The last 10 breaths during each loaded run were ensemble averaged using the point at the beginning of inspiratory flow as the reference point.

The inspiratory difficulty score during loading was compared for a given imposed ITL (between external ITL and PEEP), and for a given EELV (between PEEP and CPAP). The imposed ITL was measured as Pmo at the beginning of inspiratory flow during external ITL, and as the Pes value at the beginning of inspiratory flow relative to that on the individual PVstat,w for a given EELV during application of PEEP (Figure 2). In six subjects, the individual PVstat,w was obtained by fitting by eye the relationship between lung volume and Pes, recorded during repeated relaxation maneuvers. The criteria for respiratory muscle relaxation were zero Pdi, smooth declines of Pes and Pga during the procedure, and reproducibility of the relaxation characteristics (18). In two subjects, in spite of repeated efforts, respiratory muscle relaxation was considered unsuccessful according to the above criteria. The slope of PVstat,w was therefore adopted from the literature (21). Taken together, the mean slope of the PVstat,w used to calculate PEEPi during PEEP application was 0.209 ± 0.007 L/cm H₂O. For the six subjects who were able to perform satisfactory respiratory muscle relaxation, the measured static compliance of the total respiratory system was 0.112 ± 0.005 L/cm H₂O.

View larger version (18K): [in this window] [in a new window]   Figure 2.   Measurement of the imposed ITL during application of PEEP and external ITL. Point C represents the lung volume-Pes relationship at end expiration during quiet unloaded breathing. The thick line is the static pressure-volume relationship of the chest wall. On the clockwise breathing loop, Points A and B represent the beginning and end of inspiratory flow. During PEEP, EELV increases resulting in PEEPi. The PEEPi-imposed ITL was measured as the Pes value at the beginning of inspiratory flow (Point A) relative to that on the static pressure-volume relationship of the chest wall for a given EELV (left panel ). This is comparable with the imposed ITL measurement given by AC (right panel ) during external ITL application except that in the latter condition, EELV did not change. External ITL was also measured as the negative mouth pressure at the beginning of inspiratory flow, which almost always gave the same value as that obtained by the method shown in the right panel.

In order to ensure the reliability of IC maneuvers, the Pes values at each peak IC effort were measured. We found the following: (1) the coefficient of variation of Pes for each subject was between 5.57% and 11.30% (mean 7.68%); (2) the Pes at peak IC never went less negative than 30 cm H₂O except in one subject, whose mean Pes was 28 cm H₂O with a coefficient of variation of 6.14%; (3) there was no significant change in Pes at peak IC when the comparison was made among different loads and between before and the end of any loading run (Figure 3). These results suggest that the IC maneuvers were reliable in evaluating the changes in EELV.

View larger version (10K): [in this window] [in a new window]   Figure 3.   Pes values during IC efforts (zero flow) plotted against increasing magnitude of external ITL, PEEP, and CPAP. Open circles: before application of each load; closed circles: at the end of each loading run.

The results are presented as mean ± SE. The difference between values at a given load and at rest was examined by Dunnett's test. The difference in the results across the different types of load was evaluated by the general linear model analysis of variance. Pearson correlation and multiple step-forward regression analysis were used to examine the various factors contributing to the inspiratory difficulty scores. A value of p < 0.05 was considered to indicate statistical significance unless otherwise stated.

	RESULTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Figure 4 shows the experimental recordings during external ITL, PEEP, and CPAP loading from one subject. Please note that the unstable volume signal does not represent an unstable EELV during the course of loading, as it mainly reflects the electrical drift of the volume signal due to integration of flow. During external ITL, EELV was constant as reflected by the constant IC and constant end-expiratory Pes. At the maximal applied external ITL, EELV was 0.02 ± 0.02 L. During PEEP and CPAP applications, EELV increased, as shown by the decreased IC amplitude at the end of loading and the change in end-expiratory Pes in the positive direction. After an initial transition period, the level of the elevated EELV became stable for the rest of the load. This is illustrated by the stable end-expiratory Pes during the second half of the loading period. This was observed during both PEEP and CPAP applications in all subjects.

View larger version (34K): [in this window] [in a new window]   Figure 4.   Experimental recordings of flow, lung volume, Pes, and Pmo from one subject during external ITL, PEEP, and CPAP. The breathing signals were unstable during the transition from quiet unloaded breathing (QB) to the load and became stabilized thereafter. The changing baseline of the volume tracing mainly reflects artificial drift due to integration of flow. The changes in EELV under each condition can be measured by comparing the amplitude of the IC before and at the end of the load. IC was constant during external ITL, but reduced during PEEP and CPAP applications, suggesting the increase in EELV.

Figure 5 shows the mean perceived inspiratory difficulty under all loading conditions. The comparisons were made between applications of external ITL and PEEP for a given imposed ITL, and between applications of PEEP and CPAP for a given increase in EELV. The mean perceived inspiratory difficulty increased progressively with either increasing the imposed ITL or increasing EELV under all loading conditions (all r² > 0.95, p < 0.0001). However, the inspiratory difficulty score was always greater during PEEP than during external ITL and CPAP for a given imposed ITL and EELV, respectively (p < 0.001).

View larger version (11K): [in this window] [in a new window]   Figure 5.   The perceived inspiratory difficulty plotted as a function of the imposed ITL (left panel ) during external ITL (closed circles) and PEEP (open circles) applications, and as a function of the increase in EELV (right panel ) during CPAP (closed circles) and PEEP (open circles) applications. Results are mean ± SE. Linear regressions are shown for each set of the data.

In order to explore the separate contributions of the imposed ITL and the dynamic increase in operating lung volume to the perception of inspiratory difficulty, Pearson correlations between inspiratory difficulty score and the imposed ITL, EELV, and EILV were performed for each subject. Inspiratory difficulty score correlated significantly (p < 0.05) with the imposed ITL in seven of the eight subjects, with EELV in three subjects, and with EILV in five subjects (Table 1). Next, the step-forward regression of the inspiratory difficulty score to the above three independent variables was performed using 0.15 as the margin of acceptance. The imposed ITL was the first selected variable in six subjects and the second selected variable in the remaining two subjects, compared with EELV, which was the first selected variable in two subjects and the third selected variable in another two subjects, and with EILV, which was the second selected variable in five subjects (Table 2). When the results from all subjects were pooled with EELV and EILV normalized as percent of the individual's IC during unloaded breathing, inspiratory difficulty scores correlated significantly (p < 0.0001) with all the three independent variables, with r² of 0.40, 0.24, and 0.26 for the imposed ITL, EELV, and EILV, respectively. Step regression selected imposed ITL as the first significant variable (r² = 0.40), and EILV as the second significant variable, which increased the r² to 0.64. The coefficient of the fit (slopes) was 0.21 ± 0.02 cm H₂O¹ for the imposed ITL, and 0.051 ± 0.006 %IC¹ for EILV.

The effects of loading on breathing parameters are summarized in Table 3. In order to evaluate whether and to what extent the various breathing parameters (including VT, f, E, TI, TI/Ttot, and VT/TI as well as the total Pes swing) contributed independently to the perceived inspiratory difficulty during loading, with the pooled data, we first performed the multiple regression of each of these parameters against the three independent variables, the imposed ITL, EELV, and EILV. We found that each of the breathing parameters was related to at least one of the three independent variables. Then, we calculated the residuals for each parameter and entered them as additional independent variables to regenerate the step-forward regression for inspiratory difficulty scores. As shown in Table 4, adding the residuals of the total Pes, TI/Ttot, E, f, and VT slightly but significantly increased the fit.

	DISCUSSION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Dynamic hyperinflation has been associated with perception of increased breathing effort (4, 9, 15, 22). The adverse effect of dynamic hyperinflation on breathing can be broken into two components, the inspiratory threshold load due to PEEPi and the increase in operating lung volume itself. The latter puts the inspiratory muscles in a mechanically unfavorable condition (23). How PEEPi-imposed ITL and increase in EELV independently contribute to breathing sensation is unknown. In dynamically hyperinflated patients, CPAP has been shown to counterbalance PEEPi without significant influence on lung volume. Some studies (22, 24) have suggested that CPAP significantly reduced breathing effort or dyspnea in patients with evident dynamic hyperinflation, implying the importance of the PEEPi-imposed ITL in contributing to breathing sensation. Others (25), however, have indicated that CPAP had little effect on dyspnea in spite of its effect on reducing the work of breathing, implying the increased operating lung volume to be a major source of the breathing-related sensation. None of the previous studies has attempted to partition breathing sensation into contributions from PEEPi and changes in EELV. This was, therefore, the major aim of the current study.

In the present study, in order to separate the effect of the imposed ITL and operating lung volume on inspiratory difficulty, increase in ITL alone without changing EELV and increase in EELV alone without concomitant PEEPi were evaluated separately during external ITL and CPAP applications, respectively (Figure 1). Inspiratory difficulty scores under these two conditions were compared with those when both factors (increase in imposed ITL and EELV) were applied simultaneously during PEEP application. Our major findings are as follows. The perceived inspiratory difficulty increased significantly with both increasing ITL alone (external ITL) and increasing EELV alone (CPAP), and was always greater during the combined increase in both the imposed ITL and EELV (PEEP) than with increases in either factor alone, suggesting that the imposed ITL and EELV each contribute independently and significantly to the perceived inspiratory difficulty. However, the imposed ITL was found to be a stronger factor than operating lung volume in predicting the variability of the perceived inspiratory difficulty, no matter whether the analysis was performed on individual or pooled results.

It is not surprising that the imposed ITL was selected as the strongest predictor for inspiratory difficulty. The characteristic feature of ITL or PEEPi that differs from resistive and elastic loads is its requirement that inspiratory muscles overcome the load before initiating flow, a condition known as neuromechanical uncoupling of the ventilatory pump (3, 22, 26). Lougheed and coworkers (22) showed that balancing PEEPi while assisting inspiratory muscles by CPAP considerably relieved breathlessness, whereas just assisting inspiratory muscles without balancing PEEPi by intermittent positive airway pressure (IPAP) had much less effect on breathlessness during asthmatic attacks. It is believed that neuromechanical uncoupling of the ventilatory pump due to PEEPi-imposed ITL results in feelings of unrewarded inspiratory effort/difficulty, which forms an important component of sensory feedback mediating generally unpleasant sensations of breathing (15, 22, 26).

Although the analysis of the pooled data did not select EELV as a significant independent variable, it by no means implies that EELV contributed nothing to inspiratory difficulty under our experimental conditions. The results in Figure 5 show clearly that elevating EELV increased the perceived inspiratory difficulty significantly and independently, consistent with previous observations in normal subjects who reported a progressively greater magnitude of dyspnea when EELV was increased by lowering the body surface pressure without accompanied PEEPi (27). However, because of a strong association between EELV and EILV (r = 0.9, p < 0.0001), entry of EELV as a significant independent variable was rejected by the model. The current results are in agreement with previous studies (4, 22, 28) performed in patients with dynamic hyperinflation, which suggest that EILV is a better predictor of breathing-associated sensation than EELV. This is probably because variation of EILV takes into consideration changes in both EELV and VT.

In the present study, we chose only three parameters, the imposed ITL, EELV, and EILV, as the primary independent variables. These variables were chosen because they are important parameters of dynamic hyperinflation. All other respiratory measurements were treated, like the perceived inspiratory difficulty, as outputs of the experimental intervention. The advantage of this arrangement is that it avoids any direct interference of these other breathing parameters with the isolated evaluation of the effect of our three independent variables on inspiratory difficulty. However, the perceived inspiratory difficulty may also have been influenced by changes in breathing pattern or respiratory pressures. Hence, the residuals of each breathing parameter were used as additional independent variables to evaluate the perceived inspiratory difficulty. Our results show that only an additional 10% of the variability in inspiratory difficulty was accounted for by including the residuals of these breathing parameters into the analysis (Table 4), suggesting that the perceived inspiratory difficulty was essentially a direct result of our experimental interventions, not secondary to alterations in breathing.

It should be noted that although the imposed ITL was a stronger predictor for inspiratory difficulty than operating lung volume, it does not necessarily mean that inspiratory difficulty is more sensitive to changes in inspiratory threshold than operating lung volume. It is hard to directly compare the contributions of these two factors to the magnitude of inspiratory difficulty as their units are different. During dynamic hyperinflation, the increase in PEEPi and EELV should be proportional. We are therefore able to use our results to estimate the relative magnitude of effects of threshold load and volume on inspiratory difficulty. The mean IC of our subjects was 2.60 L. The coefficients (slopes) of the imposed ITL and EILV were 0.21 cm H₂O¹ and 0.051 %IC¹, respectively. Assuming that during dynamic hyperinflation, EELV increases by 1 L with a constant VT, then EILV also increases by 1 L or 38.5% IC. This will induce an inspiratory difficulty of 1.96 units (38.5% IC multiplied by 0.051 %IC¹). Assuming that respiratory system compliance is 0.1 L/cm H₂O, increase in EELV by 1 L will result in a PEEPi of 10 cm H₂O, which will induce an inspiratory difficulty of 2.10 units (10 cm H₂O multiplied by 0.21 cm H₂O¹). These results imply that the relative magnitudes of the inspiratory difficulty in response to the imposed ITL and operating lung volume would probably be very similar.

The current experimental design permits us not only to examine the independent effect of the imposed ITL and operating lung volume, but also to assess the possible interaction between these two factors in affecting inspiratory difficulty. We assume that if their effects are perfectly independent, the sum of the inspiratory difficulty scores from external ITL and CPAP should be equal to the scores obtained during PEEP with both factors being presented. In order to make the calculation comparable, the inspiratory difficulty scores during external ITL and CPAP (at a comparable imposed ITL or EELV actually determined during PEEP) were calculated from the individual relationships as shown in Figure 5. They were then compared with the inspiratory difficulty obtained during PEEP. As shown in Figure 6, the results were not significantly different from the identity line except at the highest load. In addition, we showed that the imposed ITL alone explained 40% of variability of inspiratory difficulty. Adding EILV into the model explained an additional 24%, which was very close to the 26% of variability explained by EILV alone. The above results seem to suggest that the effect of the imposed ITL and increasing operating lung volume on inspiratory difficulty may be largely additive not multiplicative under our experimental conditions.

View larger version (19K): [in this window] [in a new window]   Figure 6.   The sum of the inspiratory difficulty scores from external ITL and CPAP (external ITL + CPAP) plotted against those obtained from PEEP. The dashed line is the line of identity. *The sum of inspiratory difficulty during external ITL and CPAP significantly different from that during PEEP.

Finally, it needs to be pointed out that the actual increases in EELV during PEEP and CPAP were greater than predicted from the applied positive pressures. This was unlikely due to overestimation of the increase in EELV by IC maneuvers. IC efforts have been proved reliable in estimating the changes in EELV not only in healthy subjects (29) but also in patients with COPD (4, 20, 28). Specifically, the Pes values and its coefficient of variation at peak IC in the present study were comparable to those reported by Younes and Kivinen (29), and did not change either among different loads or at the end of each given loading run (Figure 3). This suggests that the IC maneuvers and therefore the determination of EELV were reliable. The greater than expected EELV during positive pressure applications may have been the result of the tonic inspiratory muscle activity. In the present study, in order to reduce expiratory muscle activity and ensure the desired increase in EELV during CPAP and PEEP, we advised the subjects not to push too much during expiration. It was possible that they may have overadjusted by tonically contracting inspiratory muscles as "not pushing" was very hard to control precisely. In fact, tonic inspiratory muscle activity has been observed during dynamic hyperinflation (10, 30, 31). How it may affect breathing sensation is unknown. However, in the present study, the effect of the tonic inspiratory muscle activity on inspiration has been included in the measurement of the imposed inspiratory threshold load, as the latter was measured as the Pes value at the beginning of inspiratory flow relative to that on PVstat,w for a given EELV (10). Although the part of the tonic inspiratory muscle activity during expiration as well as expiratory muscle activity may also affect breathing sensation, we assume that the inspiratory difficulty as our subjects reported only reflects the sensation of inspiration. This is an important assumption of the present study.

In summary, we partitioned the effect of the imposed ITL and the increase in operating lung volume on breathing effort sensation under an experimental condition simulating the pathophysiological consequences of dynamic hyperinflation. We chose inspiratory difficulty as the dependent variable, as it had been selected as the qualitatively unique descriptor for exertional dyspnea during dynamic hyperinflation (15). We conclude that under our experimental conditions, both the imposed ITL and the increase in operating lung volume are responsible for generating the sensation of inspiratory difficulty. However, the inspiratory threshold is a stronger predictor of the perceived inspiratory difficulty than increasing operating lung volume, while the relative sensitivity of the inspiratory difficulty in response to changes in these two factors may be very close.