Journal of Applied Physiology AJP: Cell Physiology
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

J Appl Physiol 92: 1943-1952, 2002. First published January 11, 2002; doi:10.1152/japplphysiol.00393.2000
8750-7587/02 $5.00
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF) Free
Right arrow All Versions of this Article:
92/5/1943    most recent
00393.2000v1
Right arrow Submit a response
Right arrow Alert me when this article is cited
Right arrow Alert me when eLetters are posted
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via ISI Web of Science (21)
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Iandelli, I.
Right arrow Articles by Pedotti, A.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Iandelli, I.
Right arrow Articles by Pedotti, A.
Vol. 92, Issue 5, 1943-1952, May 2002

Determinants of exercise performance in normal men with externally imposed expiratory flow limitation

Iacopo Iandelli1, Andrea Aliverti2,3, Bengt Kayser4, Raffaele Dellacà2,3, Stephen J. Cala5, Roberto Duranti6, Susan Kelly7, Giorgio Scano1,6, Pawel Sliwinski8, Sheng Yan7, Peter T. Macklem7, and Antonio Pedotti2,3

1 Fondazione Don Gnocchi, I-50020 Pozzolatico; 6 Clinica Medica III, Università di Firenze, I-50134 Firenze; 2 Centro di Bioingegneria, Fondazione Don Gnocchi e Politecnico, I-20148 Milano; 3 Dipartimento di Bioingegneria, Politecnico di Milano, 32 I-20133 Milano, Italy; 4 University of Geneva, CH-1211 Geneva, Switzerland; 5 Westmead Hospital, NSW-2145 Sydney, Australia; 7 Meakins-Christie Laboratories, Montreal Chest Institute, McGill University Health Centre, Montreal, Quebec, Canada H2X 2P4; and 8 Department of Respiratory Medicine, Institute of Tuberculosis and Lung Diseases, 01-138 Warsaw, Poland


    ABSTRACT
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES

To understand how externally applied expiratory flow limitation (EFL) leads to impaired exercise performance and dyspnea, we studied six healthy males during control incremental exercise to exhaustion (C) and with EFL at ~1. We measured volume at the mouth (Vm), esophageal, gastric and transdiaphragmatic (Pdi) pressures, maximal exercise power (Wmax) and the difference (Delta ) in Borg scale ratings of breathlessness between C and EFL exercise. Optoelectronic plethysmography measured chest wall and lung volume (VL). From Campbell diagrams, we measured alveolar (PA) and expiratory muscle (Pmus) pressures, and from Pdi and abdominal motion, an index of diaphragmatic power (Wdi). Four subjects hyperinflated and two did not. EFL limited performance equally to 65% Wmax with Borg = 9-10 in both. At EFL Wmax, inspiratory time (TI) was 0.66s ± 0.08, expiratory time (TE) 2.12 ± 0.26 s, Pmus ~40 cmH2O and Delta VL-Delta Vm = 488.7 ± 74.1 ml. From PA and VL, we calculated compressed gas volume (VC) = 163.0 ± 4.6 ml. The difference, Delta VL-Delta Vm-VC (estimated blood volume shift) was 326 ml ± 66 or 7.2 ml/cmH2O PA. The high Pmus and long TE mimicked a Valsalva maneuver from which the short TI did not allow recovery. Multiple stepwise linear regression revealed that the difference between C and EFL Pmus accounted for 70.3% of the variance in Delta Borg. Delta Wdi added 12.5%. We conclude that high expiratory pressures cause severe dyspnea and the possibility of adverse circulatory events, both of which would impair exercise performance.

dyspnea; respiratory muscles; dynamic hyperinflation; ventilation; blood volume shifts


    INTRODUCTION
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES

IN HEALTHY HUMANS, EXERCISE PERFORMANCE is not usually limited by breathlessness. Borg scale rankings of difficulty in breathing do not approach maximal levels (16, 19, 34), although the respiratory muscles may become fatigued (7, 15, 34). On the other hand, when expiratory flow is limited by a Starling resistor during exercise, severe dyspnea develops in normal subjects, which seriously impairs exercise performance (16). The purpose of the present study was to find out why and how externally applied expiratory flow limitation (EFLe) leads to such difficulty in breathing with resulting impairment of exercise performance.

One possibility is dynamic hyperinflation (DH), which has often been implicated as a cause of dyspnea and impaired exercise performance in airway obstruction (4, 6, 9, 25, 26, 28, 38). Another is the pressure developed by respiratory muscles. This was proposed in a landmark, but frequently forgotten, paper in 1971 by Potter et al. (30) who discovered a relationship between expiratory pressures and difficulty in breathing in chronic obstructive pulmonary disease (COPD). Data supporting the hypothesis that expiratory pressures cause dyspnea during EFLe exercise have been presented by Kayser et al. (16). However, neither study measured operating lung volumes (VL) on a breath-by-breath basis and thus could not assess the role of DH in producing dyspnea and exercise limitation. Potter et al. suggested that high expiratory pressures might impair venous return (30). Thus they implicated circulatory factors in exercise impairment. Montes de Oca et al. (23) also suggested a link between breathing mechanics and exercise performance in COPD. They found that maximal O2 pulse was the best predictor of maximal O2 uptake in severe COPD and that inspiratory pleural pressures (Ppl), in turn, were good predictors of O2 pulse.

To assess the contributions of DH, respiratory pressure swings, and circulatory factors to breathing sensation and exercise performance in normal subjects with EFLe, we used pressure measurements combined with optoelectronic plethysmography to measure breath-by-breath absolute chest wall volume (Vcw) changes during incremental exercise on a cycle ergometer (1, 8). We also measured the difference between volume expired at the mouth and the volume change of the thorax1 and assumed that this difference was due to the volume of compressed gas and liquid shifts from thorax to extremities. Specifically, we tested the hypothesis of Potter et al. (30) that high expiratory pressures caused difficulties in breathing, circulatory effects, and exercise performance.


    METHODS
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES

Subjects. We studied six healthy normal men aged 36.2 ± 3.6 yr with anthropometric characteristics shown in Table 1. Subjects were all laboratory personnel trained in respiratory maneuvers. Chest wall kinematics were measured by optoelectronic plethysmography as previously described (8). Eighty-nine reflective markers were placed front and back over the chest wall from clavicles to pubis. Each marker was tracked in 3D by four video cameras, two in front of the subject and two behind. A parallel-processing computer integrated the motion of each marker and, by using Gauss's theorem, calculated the Vcw and its compartments. We used the three-compartment chest wall model of Ward et al. (36) to measure volumes of the lung-apposed or pulmonary rib cage (RCp), the diaphragm-apposed or abdominal rib cage (RCa) and the abdomen. The transverse markers at the level of the xiphisternum separated RCp from RCa, whereas the markers along the costal margin separated RCa from abdomen. The sum of the volume of each compartment equalled Vcw: Vcw = Vrc,p + Vrc,a + Vab, where Vrc,p, Vrc,a, and Vab are the volumes of the RCp, RCa, and abdomen, respectively. Changes in Vcw were assumed to equal changes in lung gas volume (VL), plus the volume of any blood shifts from thorax to extremities (VB): Delta Vcw = Delta VL + VB. Changes in VL were taken as the sum of the volume of gas expired at the mouth (Vm) plus the volume of gas compressed (VC): Delta VL = Vm + VC. Thus Delta Vcw = Vm + VC + VB = Delta Vrc,p + Delta Vrc,a + Delta Vab.

                              
View this table:
[in this window]
[in a new window]
  		Table 1.   Subdivisions of lung volume, FEV1, and anthropometric characteristics of the subjects

We were concerned that if EFLe exercise induced sufficient DH, the area of apposition of diaphragm to rib cage might disappear, converting a two-compartment rib cage to a single compartment. Therefore, we monitored the upper border of the area of apposition (i.e., the border between RCp and RCa) continuously during exercise by ultrasound and found that the area was well maintained even in the presence of DH.

Gastric (Pga) and esophageal pressures (Pes) were measured by standard balloon-tipped catheters connected to pressure transducers and used as indexes of abdominal (Pab) and pleural pressures (Ppl). Mouth pressure (Pm) was measured by a tube connecting the mouthpiece to a pressure transducer. Transpulmonary pressure was taken as the difference of Pm and Pes (Pm - Pes), transdiaphragmatic pressure (Pdi) as the difference of Pga and Pes (Pga - Pes), and chest wall elastic recoil pressure (Pcw) as the difference of Ppl and atmospheric pressure (Pb) during relaxation. For the three chest wall compartments, Ppl - Pb was taken as the pressure difference across RCp and Pab - Pb for RCa and abdomen (1, 36). Flow was measured by a pneumotachygraph attached to the mouthpiece, and its integral was used to measure tidal volume (VT), respiratory frequency (f), minute ventilation (VE), inspiratory time (TI), expiratory time (TE), duty cycle (TI/Ttot) and mean inspiratory flow (VT/TI). Mean flows of the chest wall compartments between zero flow points were calculated as their volume changes divided by TI or TE. Delta Vab/TI was used as an index of diaphragmatic shortening velocity and Pdi · (Delta Vab/TI) as an index of diaphragmatic power (Wdi) (1).

Protocol. Each subject was studied on two occasions. On both, lung and chest wall relaxation pressure-volume curves were measured by having the subjects breathe into total lung capacity and then relax and breathe out passively through a high resistance to functional residual capacity (FRC). These maneuvers were repeated until the relaxation curves were reproducible, Pm finished at zero, and Pdi was zero throughout. From these data, we obtained the elastic recoil pressure of the lung as transpulmonary pressure, of the chest wall as chest wall elastic recoil pressure, of RCp (Prc,p = Ppl - Pb), of RCa (Prc,a = Pab - Pb), and of abdomen (Pab - Pb) as functions of VL, Vcw, Vrc,p, Vrc,a, and Vab, respectively. After obtaining these data, we stimulated the phrenic nerves with single shocks, bilaterally, transcutaneously with the airway closed at relaxed FRC both sub- and supramaximally and measured the resulting rib cage distortions as a function of Pdi (1). This was done to measure the restoring forces resulting from rib cage distortion, if any existed, during EFLe exercise (1). The subjects then performed an incremental exercise test either to determine maximum power output (Wmax) under control conditions or to measure exercise performance when exhaling through a Starling resistor that limited expiratory flow to ~1 l/s, thereby simulating COPD. During exercise, subjects breathed through a mouthpiece and pneumotachygraph, which was attached to a Hans Rudolph valve, which separated inspiratory and expiratory flow. Flow limitation was achieved by putting the Starling resistor on the expiratory port of the valve. A 2-l jar was placed parallel with the Starling resistor, which acted as a capacitance so that, at the beginning of expiration, flow was somewhat greater than 1 l/s, whereas at the end it was somewhat less (34a). Flow-volume loops during maximal EFLe exercise and during control exercise at the same workload are shown in Fig. 1.


View larger version (13K):
[in this window]
[in a new window]
  		Fig. 1.   Example of flow volume loops during maximal expiratory flow limitation (EFL) exercise (B) and at the same workload during control exercise (A) in a hyperinflator. Also shown are loops during quiet breathing at rest. Volumes were measured by optoelectronic plethysmography so that the decrease in volume before end inspiration (insp) and the increase in volume before end expiration (exp) are due to gas compressibility and liquid shifts. FRC, functional residual capacity.

After 3 min of spontaneous breathing, subjects began pedalling on a cycle ergometer. After starting at zero load, workload was increased in 25-W steps every 4 min until exhaustion. Data were gathered during the last 40 s of each workload, including the subject's subjective assessment of breathing difficulty using a 10-point Borg scale. We randomized the order in which these two exercise tests were performed. Wmax for both control and EFLe exercise was defined as the highest level of exercise that could be sustained for 4 min.

Data analysis. During quiet breathing and control and flow-limited exercise, we measured end-expiratory and end-inspiratory Vrc,p, Vrc,a, Vab, and Vcw at each workload. All volumes reported, except when otherwise specified, are the ensemble averages over the last 40 s of each run.

Campbell diagrams containing the static deflation pressure-volume curve of the lung, the relaxation pressure-volume curve of the chest wall, and dynamic pressure-volume loops during exercise were used to calculate alveolar pressure (PA) and the pressures developed by the expiratory muscles (Pmus). Pressures developed by rib cage and abdominal muscles were calculated as the distance between dynamic and static relaxation pressure-volume loops for RCp and abdomen, respectively, as previously reported (1, 36). We estimated velocity of shortening of rib cage muscles and abdominal muscles as Delta Vrc,p/TI and Delta Vab/TE, respectively, and calculated power as the product of pressure and flow, as we had done previously and reported in detail in the companion paper (1, 2).

To measure rib cage distortion, we first plotted Vrc,p vs. Vrc,a during relaxation at different VL to obtain the undistorted configuration of the rib cage. Distortion was then measured as the perpendicular distance between a given configuration after phrenic stimulation and the relaxation line and expressed as a percent of Vrc,p (10, 18).

Differences between volume expired at the mouth and optoelectronic plethysmographic volumes were evaluated by plotting the integral of flow at the mouth against the Vcw measured optoelectronically. From measured values of PA, the operating VL measured plethysmographically, and separately measured subdivisions of VL, we calculated VC from Boyle's law. We subtracted this volume from the difference between plethysmographic and integrated flow volumes to obtain the VB during exercise: VB = Vcw - Vm - VC.

Statistical analysis. To obtain determinants of EFLe exercise performance compared with normal exercise, we used the difference between Borg scale ratings of breathing effort at the same exercise workload with and without EFLe (Delta Borg) as the dependent variable. We tested a wide variety of tentative, predictive, independent variables comprising various inspiratory and expiratory pressures, breathing pattern indexes, the chest wall and its compartmental volumes, and power output of the diaphragm. Pressures and powers developed by muscles were obtained from the companion paper (2). Linear correlation coefficients were obtained for all of the tentative independent variables. Those with P <=  0.0002 (r2 >=  42.0) were retained for further analysis. We then developed a model and performed a multiple stepwise linear regression analysis to obtain the parameters that best predicted flow-limited exercise performance.

Significance of differences between control and Starling runs was performed by the nonparametric Wilcoxon matched-pairs test.


    RESULTS
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES

Volume displacements and exercise workload. During the flow-limited run, four subjects hyperinflated dynamically at the highest EFLe workload, whereas the remaining two subjects did not. We have analyzed these two subjects, called euvolumics, separately from the other four. Exercise workload appeared to be equally impaired in both groups, as shown in Table 2. Mean volume displacements ± SE, as a function of exercise workload in these two groups, are shown in Fig. 2. Figure 2A is the results in the hyperinflators, and Fig. 2B is the euvolumics. In both exercise runs and in both groups, the progressive increase in end-inspiratory volume with exercise workload was due almost entirely to an increase in Vrc,p and Vrc,a. There was little or no increase in end-inspiratory Vab. The hyperinflators (but not the euvolumics) increased end-inspiratory volume and approached total lung capacity at the highest EFLe workload compared with control. Again, this was due to an increase in rib cage volume, not abdominal displacement. The contribution of RCp to this increase was greater than that of RCa. Flow limitation in the euvolumics produced very little change in exercise volume displacements. Euvolumics' end-expiratory Vcw were, if anything, surprisingly less during flow-limited exercise. In contrast, the hyperinflators had a normal decrease in end-expiratory Vcw until they reached an exercise workload of 37.5% Wmax. Then at 50% Wmax, end-expiratory Vcw increased, but not above FRC (0.0 on the ordinate). Finally at the maximal EFLe workload (65% Wmax), hyperinflators were able to achieve an end-expiratory volume that exceeded FRC by ~850 ml. All chest wall compartments contributed to this increase, but only end-expiratory Vrc,p and Vrc,a were greater than their volumes at FRC. It is notable that, despite the lack of an effect of severe flow limitation on volume displacements in the euvolumics, maximal exercise performance was impaired to the same extent as the hyperinflators at 65% Wmax. In all subjects, exercise was terminated by intense difficulty in breathing when Borg scale ratings reached 9 or 10. 

                              
View this table:
[in this window]
[in a new window]
  		Table 2.   Wmax during control and flow-limited run



View larger version (32K):
[in this window]
[in a new window]
  		Fig. 2.   Mean ± SE volumes of the chest wall and its compartments during control (circles) and flow-limited (triangles) exercise. Closed symbols are end expiration, and open symbols are end inspiration. A: hyperinflators. B: euvolumics. CW, chest wall; RCp, lung-apposed rib cage; RCa, diaphragm-apposed rib cage; AB, abdominal volume; QB, quiet breathing. Abscissa is exercise workload as a percentage of maximum control. Far right triangles in each diagram represent maximal EFL exercise.

Breathing pattern. In Fig. 3, the breathing pattern is shown as measured by integration of flow at the mouth. The four plots are of VE, VT, f, and TI/Ttot. During control exercise, the data for all six subjects were averaged. During EFLe exercise, the data for the euvolumics were analyzed separately from those for the hyperinflators. VE tended to be less during flow-limited exercise, and this difference became greater as exercise workload increased. At zero workload, EFLe VE was 97% of control zero workload VE but was only 88% at the maximal EFLe workload. VT was increased during flow-limited exercise until the last exercise workload, when it became equal to the control VT in the euvolumics and less than control in the hyperinflators. The f was less than control during all flow-limited exercise workloads in the euvolumics. This was also true for the hyperinflators at the workloads less than the maximum achieved with flow limitation, but during the last workload there was a brisk increase in f in this group. Because VT was higher and f less during EFLe exercise, alveolar ventilation might have been preserved despite the reduced VE. Therefore, we calculated alveolar ventilation on the basis of an assumed anatomical dead space equal in milliliters to body weight in pounds. At workloads higher than 0 W, alveolar ventilation was decreased on average by 9%. For a given CO2 production, this would result in an 11% increase in arterial PCO2 over that during control exercise (~4.5 Torr).


View larger version (27K):
[in this window]
[in a new window]
  		Fig. 3.   Mean ± SE values of ventilatory parameters during control () exercise (6 subjects) and during flow-limited exercise. open circle , Hyperinflators; black-triangle, euvolumics. Ordinates are as in Fig. 1.

During control exercise, TI/Ttot increased gradually with exercise workload from ~40 to ~48%. In contrast, flow limitation resulted in a progressive drop in TI/Ttot to a value of ~23%, which was similar for both groups.

Table 3 gives the values for VT/TI, TI, TE, and f at the maximal workload achievable during the flow-limited run and at the same workload during control exercise. Although the effects of flow limitation on f were variable and the mean values not significantly different, the effects on mean inspiratory flow and TI were systematic. Mean inspiratory flow increased by an average of 65%, TI was reduced by nearly 50%, and TE increased by 30%.

                              
View this table:
[in this window]
[in a new window]
  		Table 3.   Ventilatory parameters

Pressure-volume relationships. Figure 4 shows Campbell diagrams with examples of dynamic pressure-volume curves at the maximal levels of EFLe exercise in a euvolumic (Fig. 4A) and a hyperinflator (Fig. 4B). The left heavy line in Fig. 4, A and B, is the static pressure-volume curve of the lung, and the right heavy line is the relaxation curve of the chest wall. A loop during quiet breathing is also shown encircling the static pressure-volume curve of the lung with end-expiration at FRC. Note the large transpulmonary pressure swings during exercise, amounting to 70-80 cmH2O. As pointed out above, the euvolumic was able to decrease VL normally well below FRC during EFLe exercise, while end-expiratory volume remained stable. The hyperinflator's end-expiratory VL was above FRC and showed a progressive increase over the four breaths illustrated.


View larger version (20K):
[in this window]
[in a new window]
  		Fig. 4.   Campbell diagrams at the highest EFL exercise workload in a euvolumic (A) and a hyperinflator (B). Vcw, absolute chest wall volume; Pes, esophageal pressure. The heavy line with a negative slope is the quasi-static deflation pressure-volume curve of the lung from total lung capacity to FRC. The heavy line with the positive slope is the quasi-static relaxation pressure-volume curve of the chest wall. The large loops are the pressure-volume relationships during exercise. The narrow ellipse is the loop during QB. The minimum volume on this loop is at FRC. Alveolar pressure at any point in time is the horizontal distance from a dynamic loop to the static pressure-volume curve of the lung. Pressure developed by the respiratory muscles at any point in time is the horizontal distance from a dynamic loop to the chest wall relaxation curve.

A prominent feature of both traces is the markedly increased expiratory PA. Although inspiratory PA was remarkably low between -20 and -30 cmH2O, expiratory PA exceeded +50 cmH2O in both individuals. Evidently, when expiratory flow is markedly limited, normal subjects develop greater pressures in a futile attempt to increase flow. This is demonstrated by the values of Pmus given by the horizontal distance between the chest wall relaxation curve and the dynamic expiratory loops that reached ~50 cmH2O in both subjects. The small increase in volume toward the end of expiration accompanying the sudden decrease in PA is largely due to gas decompression. These two examples are the highest values of expiratory pressure that we measured. There was considerable between-individual variation with maximum values of PA ranging from 27 to 59 cmH2O.

Gas compression and blood volume shifts. The differences between volumes measured by optoelectronic plethysmography and those measured by integration of flow at the mouth during maximal EFLe exercise are shown in Fig. 5. Each panel is the ensemble average of the last few breaths during the run in a given subject. During the control run at the same level of exercise, the two volume traces moved along the line of identity and no differences were detectable. By contrast, in EFLe exercise, expiratory plethysmographic volumes led the volume of gas expired at the mouth systematically, producing a clockwise loop so that the volume change of the chest wall was greater than the volume expired at the mouth in all subjects. During inspiration, the differences were small. From the operating VL measured optoelectronically and from PA, we estimated VC and subtracted this from the total volume difference to obtain VB. By dividing VB by PA, we estimated the amount of blood shifted per unit change in PA. These data are shown in Table 4.


View larger version (19K):
[in this window]
[in a new window]
  		Fig. 5.   Instantaneous changes in volume of gas measured at the mouth (Delta Vm) and that measured plethysmographically (Delta Vcw) during control and flow-limited exercise in each subject. The data are ensemble averages of several breaths during the last 20 s of the highest level of exercise achieved with EFL. Control breaths follow the line of identity; loops during flow-limited exercise are clockwise; Delta Vcw leads Delta Vm, and expiratory Delta Vcw are systematically greater than expiratory Delta Vm. Letters in top left of each panel are subjects' initials.


                              
View this table:
[in this window]
[in a new window]
  		Table 4.   Gas compression and blood volume shifts

Rib cage distortion. During both control and EFLe exercise, rib cage distortions were small and generally within ±1%. Because the pressure cost of such distortions is small (10, 18), we ignored them.

Determinants of breathing sensation. Figure 6 shows the Borg scale rankings of breathing effort. The closed circles are the means ± SE for all six subjects as a function of workload during the control runs. The open squares give the results for EFLe exercise in the hyperinflators and the closed triangles the results in the euvolumics. Breathing sensation did not limit control exercise; the Borg scale rating reached a value of only 4 (range: 2-6) at Wmax. In contrast, as we previously observed (16) during flow-limited exercise, performance was impaired because of intense difficulty in breathing; the mean Borg scale ranking at the end of exercise was 9.3 (range: 9-10), and the subjects on average were only able to exercise to 65% of control Wmax.


View larger version (19K):
[in this window]
[in a new window]
  		Fig. 6.   Mean ± SE Borg scale ratings as a function of exercise workload (abscissa) expressed as percent control of maximal work rate. , Control exercise; open circle , flow-limited exercise, hyperinflators; black-triangle, flow-limited exercise, euvolumics. The difference between the two curves (control and flow-limited exercise) was taken as the dependent variable for regression analysis of determinants of exercise performance.

There was no difference in breathing sensation between the euvolumics and the hyperinflators at higher workloads, although the hyperinflators tended to have higher rankings at lower exercise levels before they hyperinflated. The degree of exercise limitation in both averaged 65% of control Wmax (range: 57-83). Thus both groups had the same degree of exercise impairment. Hyperinflation did not impair exercise performance in the euvolumics.

The parameters that correlated best with Delta Borg were the changes in Pmus developed by the rib cage and abdominal muscles (2) and the global differences in Pmus (Delta Pmus) measured from Campbell diagrams (P < 0.00001, r2 = 0.703). Next were ventilatory parameters reflecting the prolonged TE and shortened TI (Delta TI/Ttot: P < 0.00001, r2 = 0.519; VT/TI: P < 0.00001, r2 = 0.513). Finally, parameters reflecting central inspiratory drive were important. These were the change in Pdi (Delta Pdi; P < 0.00001, r2 = 0.503) and fold changes in Wdi (Delta Wdi; P < 0.0002, r2 = 0.420). Indexes of hyperinflation, such as changes in end-expiratory and end-inspiratory Vcw, were either nonsignificant or had r2 values of <0.370.

From these preliminary linear regressions, we constructed a model using Delta Pmus, Delta VT/TI, and Delta Wdi as independent variables. Multiple stepwise linear regression analysis revealed that this model accounted for 82.8% of the variance in Delta Borg. Of that value, 70.3% was accounted for by Delta Pmus, and Delta Wdi added an additional 12.5%. Delta VT/TI was rejected. No other model tested accounted for as much of the variance in Delta Borg as these two variables. Wdi turned out to be more important than other variables that were better correlated with Delta Borg because it was the least well correlated with Delta Pmus.


    DISCUSSION
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES

Critique of methods. The introduction of optoelectronic plethysmography solves the difficult problem of tracking absolute VL changes on a breath-by-breath basis. In addition, we used the difference between this volume measurement and the volume expired at the mouth during a single breath to calculate VC and the volume of liquid shifted from the thorax to the extremities. This assumes that this volume difference can only be due to gas compression and/or liquid shifts to the extremities. That the liquid displaced is blood seems reasonable as shifts of extravascular fluid of the magnitude we measured are unlikely. The blood could come from the lung, rib cage, or abdomen, and the shifts must be to the extremities because liquid displacements from one thoracic compartment to another (e.g., from lung to rib cage or abdomen) would not be detected by measurement of Vcw. Shifts from rib cage to abdomen could not be distinguished from isovolume maneuvers produced by respiratory muscle contraction.

EFLe mimics EFL due to dynamic compression of intrathoracic airways in that flow becomes independent of driving pressure. There was significant volume dependence of expiratory flow during each expiration because the jar in parallel with the resistor acted as a capacitance. Nevertheless, EFLe produced by the Starling resistor does not lead to dynamic compression of intrathoracic airways as occurs in physiological EFL. Presumably, if VL had decreased to the point where the maximal expiratory flow was <1 l/s, dynamic compression of intrathoracic airways would have occurred and Pm would have fallen to zero. This was not observed. Thus any feedback evoked by dynamic compression was absent in our subjects. However, our data show that tracheobronchial compression is not required to impair exercise performance, limitation of exercise by breathlessness, and CO2 retention (2, 16). Similar results have been reported by Wood and Bryan (37) in normal subjects exercising at depth and by others in the absence of any flow limitation at all; adding high external inspiratory and expiratory resistances is sufficient (11, 29).

In our experiments, in contrast to pathophysiological flow limitation, no benefit in terms of increased flow was obtained by DH. This may be a reason why two subjects remained euvolumic and the other four subjects only hyperinflated toward the end of flow-limited exercise.

Factors limiting exercise performance. In our experiments, exercise performance was limited by severe dyspnea. In every case, Borg scale ratings reached 9-10, confirming our earlier report (16). In the presence of EFL, DH has been strongly implicated as a cause of breathlessness and impairment of exercise performance (4, 9, 25, 26, 28, 38). A number of factors may be involved, including shortening of diaphragmatic fiber length, threshold load (9, 21, 22), progressive decrease of inspiratory capacity to a value lower than the desired VT (4, 5, 6, 10, 26, 28, 38), a decrease in VE (4, 5), and a decrease in lung compliance, which increases the elastic work of breathing.

There is not uniform agreement that DH is the principle factor causing exercise limitation. Noseda et al. (24) found that an index of central inspiratory drive, peak inspiratory flow, was the best predictor of dyspnea during exercise in COPD. Because we only studied two euvolumics and four hyperinflators, generalizations and statistical comparisons between these two groups are not meaningful. However, our data in euvolumics show that EFLe does not necessarily lead to DH, although exercise was terminated by breathlessness equal in degree to that experienced by hyperinflators with the same impairment of performance. In the other four subjects, DH reduced inspiratory reserve volume almost to zero, but it was a relatively late occurrence, developing only near the end of EFLe exercise (Fig. 2) after Borg scale ratings had significantly increased (Fig. 5). Thus, in our experiments, DH did not lead to impaired exercise performance in two subjects and did not contribute to the dyspnea experienced by the other four until they reached the highest EFLe exercise level. This makes us suspect that its role may be overemphasized.

Montes de Oca et al. (23) and Potter et al. (30) specifically linked the mechanics of breathing to circulatory factors during exercise in COPD with the implication that the link between abnormal mechanics of breathing and impaired exercise performance may be via the circulation rather than a malfunctioning ventilatory pump per se.

There is considerable evidence that energy supplies to locomotor muscles are inadequate at maximal exercise workloads in both healthy subjects and patients with COPD (3, 31, 33) when there is competition between respiratory and locomotor muscles for the available energy supplies. At maximal exercise in trained normal subjects, respiratory muscles consume about 15% of total body O2 consumption and take a similar percentage of the cardiac output (13, 14). If one applies this concept to COPD where the oxygen cost of breathing is about an order of magnitude greater than normal (20), the competition becomes much more severe so that as much as 50% of the O2 consumption and cardiac output may go to the respiratory muscles (3). Richardson et al. (31) reported that the mass specific O2 consumption of the quadriceps muscle in COPD was much greater when it was the only muscle exercising than it was at maximal whole body exercise. Thus, if energy supplies are adequate, locomotor muscles can perform more work. It appears that there are some patients with COPD in whom exercise may be limited by energy supplies that are insufficient to meet demands.

Blood volume shifts and respiratory pressure swings. The high expiratory pressures we measured during EFLe exercise caused gas compression and what we assume to be VB averaging 325.7 ml from the thorax to the extremities (Table 4). Although we are unable to determine how much was displaced from the abdomen and how much from the lung, only a small fraction displaced from the pulmonary capillaries could effect distribution of ventilation-to-perfusion ratios because pulmonary capillary VB is only ~220 ml during exercise (35). This might increase alveolar dead space and could be a factor in the CO2 retention that occurs (34a).

These blood shifts show that expiratory pressure swings have circulatory effects in addition to their effects on respiratory sensation (16). They did not occur during normal exercise (Fig. 5) and are therefore a specific and significant effect of high expiratory pressure.The reasons for the increased pressures are twofold (2). The enforced slowing of expiratory flow decreased expiratory muscle velocity of shortening that, for a given degree of activation, increases force development according to the muscles' force-velocity relationship (30). In addition, we found that both inspiratory and expiratory muscle power were increased during EFL exercise compared with control (2), suggesting that there was an increase in central drive probably secondary to CO2 retention (2, 16).

Potter et al. (30) thought that the high PA during exhaustive exercise in COPD might impede venous return and could impose a limitation on the cardiovascular response to exercise in patients producing a situation similar to a Valsalva maneuver. Our data show that, during EFL exercise, TE averaged ~2 s whereas TI was only ~0.7 s. (Table 3). These data combined with the pressures confirm the presence of a Valsalva-like maneuver. However, the effects of Valsalva's maneuver during exercise are unknown.

Determinants of EFLe exercise performance. Because linear regression analysis was performed on the difference in Borg scale rankings as the dependent variable and the differences in a variety of physiological parameters as independent variables, between control and EFLe exercise, our analysis is different from previous studies attempting to define determinants of exercise performance in COPD where control experiments cannot be performed in the same subject. In developing a model for the multiple stepwise linear regression, we chose an index of expiratory muscle recruitment, Delta Pmus (e.g., Fig. 4), Delta VT/TI as a reflection of increased shortening velocity of of inspiratory muscles (2) and thus their loss of force-generating ability and the increase in Wdi as an index of central drive (2). Analysis showed that Delta Pmus accounted for 70.3% of the variability in Delta Borg and Delta Wdi for an additional 12.5%. Delta VT/TI was rejected. We conclude that Pmus was an important contributor to dyspnea during EFLe exercise. The contribution of central inspiratory drive is in agreement with the importance of peak inspiratory flow found by Noseda et al. in COPD (24).

Relevance to COPD. It is uncertain how relevant our studies are to COPD. Although we have produced three major pathophysiological features of COPD in normal subjects, namely dyspnea, exercise limitation, and CO2 retention, the fact that the sensation induced by dynamic compression of intrathoracic airways (38), which may inhibit expiratory muscle recruitment (27), does not occur with the Starling resistor may have contributed to the high expiratory pressures we measured. Such pressures have rarely been reported in COPD, yet both Potter et al. (30) and Dodd et al. (12) reported a mean peak expiratory pressure at maximal exercise in COPD of 22 cmH2O with large between-individual variations. In 6 of 12 patients studied by Potter et al. (30), EEVL increased by <50% of resting VT, and there was no decrease in inspiratory capacity. Similarly, Calverley (personal communication) found no DH during incremental exercise in 8 of 20 COPD patients. These findings suggest that the individual response to EFL exercise in COPD is heterogenous but similar to what we found in normal subjects. Some patients hyperinflate whereas others do not. There is a large between-individual difference in the degree of expiratory pressure produced, which in some patients is excessive. Perhaps the relevance of our model to COPD may be to generate testable hypotheses. Our laboratory's studies (16) suggest that high expiratory presures might be a factor limiting exercise in COPD. If so, it might be possible by appropriate physiotherapy to train patients to derecruit expiratory muscles during exercise and improve performance.


    ACKNOWLEDGEMENTS

Supported by the Breath Project (Biomed 2 Programme) and The Respiratory Health Network Centres of Excellence, funded by The Medical Research Council of Canada.


    FOOTNOTES

Address for reprint requests and other correspondence: A. Aliverti, Dipartimento di Bioingegneria, Politecnico di Milano, Piazza L. da Vinci, 32 I-20133 Milano, Italy (E-mail: aliverti{at}mail.cbi.polimi.it).

1  We define thorax as did Roussos (32) as the volume of the chest wall and thus the sum of rib cage and abdominal volumes. This is in keeping with the ancient Greek usage of the word thorax.

The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

First published January 11, 2002;10.1152/japplphysiol.00393.2000

Received 17 May 2000; accepted in final form 11 December 2001.


    REFERENCES
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES

1.   Aliverti, A, Cala SJ, Duranti R, Ferrigno G, Kenyon CM, Pedotti A, Scano G, Sliwinski P, Macklem PT, and Yan S. Human respiratory muscle actions and control during exercise. J Appl Physiol 83: 1256-1269, 1997.

2.  Aliverti A, Iandelli I, Duranti R, Cala S, Kayser B, Kelly S, Misuri G, Pedotti A, Scano G, Sliwinski P, Yan S, and Macklem PT. Respiratory muscle dynamics and control during exercise with externally imposed expiratory flow limitation. J Appl Physiol 92: 1953-1963.

3.   Aliverti, A, and Macklem PT. How and why exercise is limited in COPD. Respiration 68: 229-239, 2001.

4.   Bauerle, O, Chrusch CA, and Younes M. Mechanisms by which COPD affects exercise tolerance. Am J Respir Crit Care Med 157: 57-68, 1998.

5.   Bauerle, O, and Younes M. Role of ventilatory response to exercise in determining exercise capacity in COPD. J Appl Physiol 79: 180-187, 1995.

6.   Belman, MJ, Botnick WC, and Shin JW. Inhaled bronchodilators reduce dynamic hyperinflation during exercise in patients with chronic obstructive pulmonary disease. Am J Respir Crit Care Med 153: 967-975, 1996.

7.   Bye, PT, Esau SA, Walley KR, Macklem PT, and Pardy RL. Ventilatory muscles during exercise in air and oxygen in normal man. J Appl Physiol 56: 464-471, 1984.

8.   Cala, S, Kenyon CM, Ferrigno G, Carnevali P, Aliverti A, Pedotti A, Macklem PT, and Rochester DF. Chest wall and lung volume estimation by optical reflectance motion analysis. J Appl Physiol 81: 2680-2689, 1996.

9.   Chen, RC, and Yan S. Perceived inspiratory difficulty during inspiratory threshold and hyperinflationary loadings. Am J Respir Crit Care Med 159: 720-729, 1999.

10.   Chihara, K, Kenyon CM, and Macklem PT. Human rib cage distortability. J Appl Physiol 81: 437-447, 1996.

11.   Demedts, M, and Anthonisen NR. Effects of increased external airway resistance during steady-state exercise. J Appl Physiol 35: 361-366, 1973.

12.   Dodd, DS, Brancatisano T, and Engel LA. Chest wall mechanics during exercise in patients with severe chronic air-flow obstruction. Am Rev Respir Dis 129: 33-38, 1984.

13.   Harms, CA, Babcock MA, McLaren SR, Pegelow DF, Nickele GA, Nelson WB, and Dempsey JA. Respiratory muscle work compromises leg blood flow during maximal exercise. J Appl Physiol 62: 1573-1583, 1997.

14.   Harms, CA, Wetter TJ, McLaren SR, Pegelow DF, Nickele GH, Nelson WB, Hanson P, and Dempsey JA. Effect of respiratory muscle work on cardiac output and its distribution during maximal exercise. J Appl Physiol 85: 609-618, 1998.

15.   Johnson, BP, Babcock MH, Surnan OE, and Dempsey JA. Exercise induced diaphragmatic fatigue in healthy humans. J Physiol (Lond) 460: 385-405, 1993.

16.   Kayser, B, Sliwinski P, Yan S, Tobaisz M, and Macklem PT. Respiratory effort sensation during exercise with induced expiratory flow-limitation in healthy humans. J Appl Physiol 83: 936-947, 1997.

18.   Kenyon, CM, Cala SJ, Yan S, Aliverti A, Scano G, Duranti R, Pedotti PT, and Macklem A. Rib cage mechanics during quiet breathing and exercise in humans. J Appl Physiol 83: 1242-1255, 1997.

19.   Killian, KJ, Leblanc P, Martin DH, Summers E, Jones NL, and Campbell EJM Exercise capacity and ventilatory, circulatory and symptom limitation in patients with chronic airflow limitation. Am Rev Respir Dis 146: 935-940, 1992.

20.   Levinson, H, and Cherniack RM. Ventilatory cost of exercise in chronic obstructive pulmonary disease. J Appl Physiol 25: 21-27, 1968.

21.   Martin, JG, Habib M, and Engel LA. Inspiratory muscle activity during induced hyperinflation. Respir Physiol 39: 302-313, 1980.

22.   Martin, JG, Shore SA, and Engel LA. Mechanical load and inspiratory muscle action during induced asthma. Am Rev Respir Dis 128: 455-460, 1983.

23.   Montes de Oca, M, Rassulo J, and Celli BR. Respiratory muscle and cardiopulmonary function during exercise in very severe COPD. Am J Respir Crit Care Med 154: 1284-1289, 1996.

24.   Noseda, A, Carpiaux JP, Schmerber J, Valente F, and Yernault JC. Dyspnoea and flow-volume curve during exercise in COPD patients. Eur Respir J 7: 279-285, 1994.

25.   O'Donnell, DE, Bertley JC, Chau LKL, and Webb KA. Qualitative aspects of exertional breathlessness in chronic airflow limitation. Am J Respir Crit Care Med 155: 109-115, 1997.

26.   O'Donnell, DE, Lam M, and Webb KA. Measurement of symptoms, lung hyperinflation and endurance during exercise in chronic obstructive pulmonary disease. Am J Respir Crit Care Med 158: 1557-1565, 1998.

27.   O'Donnell, DE, Sanil R, Anthonisen NR, and Younes M. Effect of dynamic airway compression on breathing pattern and respiratory sensation in severe chronic obstructive pulmonary disease. Am Rev Respir Dis 135: 912-918, 1987.

28.   O'Donnell, ED, and Webb KA. Exertional breathlessness in patients with chronic airflow limitation: the role of lung hyperinflation. Am Rev Respir Dis 148: 1351-1357, 1993.

29.   Olgiati, R, Atchou G, and Cerretelli P. Hemodynamic effects of resistive breathing. J Appl Physiol 60: 846-853, 1986.

30.   Potter, WA, Olafsson S, and Hyatt RE. Ventilatory mechanics and expiratory flow limitation during exercise in patients with obstructive lung disease. J Clin Invest 50: 910-919, 1971.

31.   Richardson, RS, Sheldon J, Poole DC, Hopkins SR, Ries AL, and Wagner PD. Evidence of skeletal muscle reserve during whole body exercise in patients with chronic obstructive pulmonary disease. Am J Respir Crit Care Med 159: 881-885, 1999.

32.   Roussos, C. Prologue. In: The Thorax. Lung Biology in Health and Disease (2nd ed.), edited by Lenfant C, and Roussos C.. New York: Dekker, 1985, vol. 29, p. xv-xix.

33.   Roussos, C, and Macklem PT. The respiratory muscles. N Engl J Med 307: 786-797, 1982.

34.   Sliwinski, P, Yan S, Gauthier AP, and Macklem PT. Influence of global inspiratory muscle fatigue on breathing during exercise. J Appl Physiol 80: 1270-1278, 1996.

34a.   Suzuki, J, Kayser B, Yan S, and Macklem PT. CO2 retention and hypoxemia during exercise in normal subjects with induced expiratory flow limitation (Abstract). Am J Respir Crit Care Med 157: A45, 1998.

35.   Vaughan, TR, Jr, DeMarino E, and Staub NC. Indication dilution lung water and capillary blood volume in prolonged heavy exercise in normal men. Am Rev Respir Dis 113: 757-772, 1976.

36.   Ward, ME, Ward J, and Macklem PT. Analysis of human chest wall motion using a two-compartment chest wall model. J Appl Physiol 72: 1338-1347, 1992.

37.   Wood, LDH, and Bryan AC. Exercise ventilatory mechanics at increased ambient pressure. J Appl Physiol 44: 231-237, 1978.

38.   Yan, S. Sensation of inspiratory difficulty during inspiratory threshold and hyperinflationary loadings. Am J Respir Crit Care Med 160: 1544-1549, 1999.

J APPL PHYSIOL 92(5):1943-1952
8750-7587/02 $5.00 Copyright © 2002 the American Physiological Society


This article has been cited by other articles:


Home page
 J. Appl. Physiol.Home page
J. L. Smith, J. E. Butler, P. G. Martin, R. A. McBain, and J. L. Taylor
Increased ventilation does not impair maximal voluntary contractions of the elbow flexors
J Appl Physiol, June 1, 2008; 104(6): 1674 - 1682.
[Abstract] [Full Text] [PDF]

Home page
 ERRHome page
P. M. A. Calverley
Exercise and dyspnoea in COPD
Eur. Respir. Rev., December 1, 2006; 15(100): 72 - 79.
[Abstract] [Full Text] [PDF]

Home page
 ERRHome page
P. T. Macklem
Circulatory effects of expiratory flow-limited exercise, dynamic hyperinflation and expiratory muscle pressure
Eur. Respir. Rev., December 1, 2006; 15(100): 80 - 84.
[Abstract] [Full Text] [PDF]

Home page
 J. Appl. Physiol.Home page
J. D. Miller, S. J. Hemauer, C. A. Smith, M. K. Stickland, and J. A. Dempsey
Expiratory threshold loading impairs cardiovascular function in health and chronic heart failure during submaximal exercise
J Appl Physiol, July 1, 2006; 101(1): 213 - 227.
[Abstract] [Full Text] [PDF]

Home page
 ThoraxHome page
P T Macklem
Exercise in COPD: damned if you do and damned if you don't
Thorax, November 1, 2005; 60(11): 887 - 888.
[Full Text] [PDF]

Home page
 ThoraxHome page
A Aliverti, K Rodger, R L Dellaca, N Stevenson, A Lo Mauro, A Pedotti, and P M A Calverley
Effect of salbutamol on lung function and chest wall volumes at rest and during exercise in COPD
Thorax, November 1, 2005; 60(11): 916 - 924.
[Abstract] [Full Text] [PDF]

Home page
 ThoraxHome page
I Vogiatzis, O Georgiadou, S Golemati, A Aliverti, E Kosmas, E Kastanakis, N Geladas, A Koutsoukou, S Nanas, S Zakynthinos, et al.
Patterns of dynamic hyperinflation during exercise and recovery in patients with severe chronic obstructive pulmonary disease
Thorax, September 1, 2005; 60(9): 723 - 729.
[Abstract] [Full Text] [PDF]

Home page
 Eur Respir JHome page
P. M. A. Calverley and N. G. Koulouris
Flow limitation and dynamic hyperinflation: key concepts in modern respiratory physiology
Eur. Respir. J., January 1, 2005; 25(1): 186 - 199.
[Abstract] [Full Text] [PDF]

Home page
 J. Appl. Physiol.Home page
A. Aliverti, B. Kayser, and P. T. Macklem
Breath-by-breath assessment of alveolar gas stores and exchange
J Appl Physiol, April 1, 2004; 96(4): 1464 - 1469.
[Abstract] [Full Text] [PDF]

Home page
 ThoraxHome page
A Aliverti, N Stevenson, R L Dellaca, A Lo Mauro, A Pedotti, and P M A Calverley
Regional chest wall volumes during exercise in chronic obstructive pulmonary disease
Thorax, March 1, 2004; 59(3): 210 - 216.
[Abstract] [Full Text] [PDF]

Home page
 ChestHome page
R. Bianchi, F. Gigliotti, I. Romagnoli, B. Lanini, C. Castellani, M. Grazzini, and G. Scano
Chest Wall Kinematics and Breathlessness During Pursed-Lip Breathing in Patients With COPD
Chest, February 1, 2004; 125(2): 459 - 465.
[Abstract] [Full Text] [PDF]

Home page