Helium-Hyperoxia, Exercise, and Respiratory Mechanics in Chronic Obstructive Pulmonary Disease

Helium-Hyperoxia, Exercise and Respiratory Mechanics in Chronic Obstructive Pulmonary Disease

Neil D. Eves1, Stewart R. Petersen1, Mark J. Haykowsky2, Eric Y. Wong3 and Richard L. Jones3

1Faculty of Physical Education and Recreation, University of Alberta, Edmonton, Alberta, Canada T6G 2H9

2 Faculty of Rehabilitation Medicine, University of Alberta, Edmonton, Alberta, Canada T6G

2G4

3 Faculty of Medicine and Dentistry, University of Alberta, Edmonton, Alberta, Canada T6G 2B7

Requests for reprints: Neil D. Eves, Faculty of Kinesiology, University of Calgary, 2500 University Drive NW, Calgary, Alberta, Canada, T2N 1N4 Corresponding Author: Neil D. Eves, Email: [email protected] Fax: (403) 284 3553 Phone: (403) 220 2413 Running Head: Helium-Hyperoxia, Exercise and COPD Descriptor Number: 55 Word Count (Manuscript Body): 4394

This article has an online data supplement, which is accessible from the issue’s table of content online at www.atsjournals.org

AJRCCM Articles in Press. Published on July 13, 2006 as doi:10.1164/rccm.200509-1533OC

Copyright (C) 2006 by the American Thoracic Society.

1

ABSTRACT

Rationale: Hyperoxia and normoxic-helium independently reduce dynamic hyperinflation and

improve the exercise tolerance of patients with chronic obstructive pulmonary disease (COPD).

Combining these gases could have an additive effect on dynamic hyperinflation and a greater

impact on respiratory mechanics and exercise tolerance. Objective: To investigate whether

helium-hyperoxia improves the exercise tolerance and respiratory mechanics of COPD patients.

Methods: Ten males with COPD (FEV1 = 47 ± 17%pred [mean±SD]) performed randomized

constant-load cycling at 60% maximal work rate breathing air, hyperoxia (40%O2, 60%N2),

normoxic-helium (21%O2, 79%He) or helium-hyperoxia (40%O2, 60%He). Measurements:

Exercise time, inspiratory capacity (IC), work of breathing and exertional symptoms were

measured with each gas. Results: Compared to air (9.4 ± 5.2 min), exercise time was increased

with hyperoxia (17.8 ± 5.8 min) and normoxic-helium (16.7 ± 9.1 min) but the improvement with

helium-hyperoxia (26.3 ± 10.6 min) was greater than both these gases (p=0.019 and p=0.007,

respectively). At an isotime during exercise, all three gases reduced dyspnea and both helium

mixtures increased IC and tidal volume. Only helium-hyperoxia significantly reduced the

resistive work of breathing (15.8 ± 4.2 vs. 10.1 ± 4.1 L·cm H2O-1) and the work to overcome

intrinsic positive end-expiratory pressure (7.7 ± 1.9 vs. 3.6 ± 2.1 L·cm H2O-1). At symptom-

limitation, tidal volume remained augmented with both helium mixtures but IC and the work of

breathing were unchanged compared to air. Conclusion: Combining helium and hyperoxia delays

dynamic hyperinflation and improves respiratory mechanics, which translates into added

improvements in exercise tolerance for patients with COPD.

Key Words: Dynamic Hyperinflation, Dyspnea, Work of Breathing, Pulmonary Function,

Expiratory Flow Limitation. (Word count for abstract: 250)

2

INTRODUCTION

Patients with chronic obstructive pulmonary disease (COPD) exhibit a reduced exercise capacity

(1-5) and often curtail exercise due to severe dyspnea (2, 5). In recent years, a growing body of

evidence (2, 3, 6-8) has identified a strong relationship between the degree of dynamic

hyperinflation and the intensity of breathlessness experienced by these patients. Due to the

expiratory flow limitation that is characteristic of COPD, end expiratory lung volume (EELV)

rises during exercise and end inspiratory lung volume (EILV) encroaches on total lung capacity

(TLC) in an attempt to augment or maintain tidal volume. The resulting dynamic hyperinflation

reduces lung compliance and increases the elastic work of breathing, while placing the

respiratory muscles at a mechanical disadvantage to generate pressure (9, 10). Additionally, any

increase in EELV above relaxation volume results in an intrinsic positive end-expiratory pressure

(PEEPi), which must be overcome before inspiratory flow can start (11). Unfortunately, even

with an increase in inspiratory effort, ventilatory constraint often ensues and it is the resulting

mismatch between respiratory effort and ventilatory output that has been implicated as the

primary mechanism responsible for the dyspnea experienced during exercise in this population

(3, 12).

Previous investigations have demonstrated that hyperoxia (HOX) reduces ventilatory demand,

increases expiratory time (TE) and delays dynamic hyperinflation, which leads to a reduction in

dyspnea and improved exercise tolerance in patients with COPD (6, 13, 14). However, at

symptom-limitation in both incremental exercise tests and in constant load trials where there is an

upward drift in ventilation (VE) (13-15), lung volumes, VE and dyspnea are unchanged from

breathing air, indicating that the ventilatory constraints to exercise are delayed but not removed

with hyperoxia (13). In contrast, a normoxic-helium gas (HE-OX) decreases turbulence within

medium to large airways and increases expiratory flow rate, which reduces dynamic

3

hyperinflation and dyspnea and improves exercise tolerance (16). We hypothesized that

combining helium and 40% O2 would maximize the effect of oxygen for reducing VE and

increasing expiratory time, while still maintaining the improved laminar flow and reduced airway

resistance associated with breathing helium. We further hypothesized that the resultant effect

would be a greater reduction in dynamic hyperinflation, which would decrease the work of

breathing and result in reduced dyspnea and improved exercise tolerance. Some of the data from

this study have previously been reported in abstract form (17).

METHODS (Method word count = 1023 words)

Please see the online supplement for a more detailed version of these methods.

Patients

Ten clinically stable males with moderate to severe COPD (FEV1/FVC = 56 ± 10%pred, PaO2 =

68 ± 6 mm Hg) volunteered to participate in the study. Individuals dependent on supplemental

oxygen, with cardiovascular disease and/or musculoskeletal abnormalities were excluded. All

patients signed an informed consent that had received institutional ethics review board approval.

The patient characteristics are presented in Table 1.

Study Design

The study was a randomized crossover design, which required three separate visits to the

laboratory. The first visit consisted of a pulmonary function test to confirm the severity of COPD

and a symptom-limited incremental exercise test to ensure the absence of cardiovascular

contraindications to exercise. During the other two visits, four constant-load symptom-limited

4

exercise trials were performed in a random order (two per visit) to examine the effect of each gas

on exercise tolerance, work of breathing, lung volumes and exertional symptoms. The patients

were asked to refrain from exercise in the 24 hours before a test and to avoid smoking, alcohol

and caffeine on testing days.

The four gas mixtures studied were: air (21% O2, 79% N2); hyperoxia (40% O2, 60% N2);

normoxic-helium (21% O2, 79% He) and helium-hyperoxia (40% O2, 60% He). A 40% O2

mixture was utilized to maximize the benefits of O2 while still obtaining the benefits of helium

(18). Throughout exercise, humidified gases were passed into a reservoir bag and supplied

through a low-resistance two-way breathing valve (2700 series, Hans Rudolph, Kansas City,

MO). The patients were blinded to the gas mixture used and were asked not to talk during, or for

a short period after exercise due to the change in vocal tone with helium.

Pulmonary Function Testing

Spirometry, single-breath diffusion capacity for carbon monoxide (DLCO) and lung volumes

determined by body plethysmograph (6200 Autobox; SensorMedics, Yorba Linda CA) were

measured before entry into the study. Resting arterial blood gases and pH were measured from a

radial artery taken at rest while breathing room air.

Incremental Exercise Test

Prior to inclusion in the study, an incremental cycle-exercise stress test to symptom limitation

was performed with expired gas analysis (TrueOne 2400, Parvo Medics, Salt Lake City, UT).

The results from this test are presented in Table 2.

5

Constant Load Exercise, Lung Volumes and Respiratory Mechanics

After adequate wash-in, spirometry was performed to obtain resting pulmonary function on each

gas mixture. Patients then exercised at 60% of their previously determined maximal work rate on

an electronically-braked cycle ergometer (Ergoline 800S, SensorMedics, Yorba Linda, CA) until

symptom limitation. Following exercise, subjects rested for 60-90 minutes before the second

exercise trial was performed with a different gas. Two research assistants, blinded to the gas

mixture used, consistently encouraged patients to exercise for as long as possible.

Ventilatory parameters were measured every two minutes by switching the patient from

the reservoir bag to a bag-in-box in series with a low resistance spirometer (Model 1022,

SensorMedics, Yorba Linda, CA). Calibration of the spirometer was confirmed with the

experimental gas mixture before each pulmonary function test and exercise trial. Esophageal

pressure was measured with a 10-cm latex balloon catheter (Ackrad Laboratories Inc., Cranford,

NJ) with a 1ml inflation volume, connected to a differential pressure transducer (MP45,

Validyne, Northridge, CA). Signals from the spirometer and pressure transducers were converted

to a digital signal using a data acquisition system (Powerlab ML785, ADI Instruments, Colorado

Springs, CO). The volume signal from the spirometer was differentiated to obtain flow. All data

were sampled at 100Hz and stored on a computer for analysis at a later date. A schematic of the

experimental setup is depicted in Figure E1 in the online data supplement.

Measurement of Lung Volumes: Assuming that TLC does not change with exercise (19, 20),

repetitive inspiratory capacity maneuvers (IC) were performed every two minutes to track

changes in EELV (TLC - IC). This technique has previously been shown to be reliable during

exercise in this population (20, 21).

Measurement of Respiratory Mechanics: The work of breathing was estimated using Campbell

diagrams and the technique of Yan et al., (22), which allows the inspiratory elastic work of

6

breathing to be separated into the work to overcome PEEPi (WIP) and the work required to

overcome the elastic recoil of the lung or the non-PEEPi inspiratory elastic load (WINP) (11, 22).

In brief, the chest-wall compliance curve was obtained from the literature (23) and was

positioned as previously described (11, 22). Esophageal pressure-volume loops during tidal

breathing were then superimposed on the static chest wall pressure-volume compliance line. The

points of zero flow at the start and end of inspiration were joined to identify dynamic lung

compliance. Inspiratory resistive work was then calculated as the area inside the pressure-volume

curve and to the left of the lung compliance line. The total elastic work performed on inspiration

was calculated as the area enclosed by the lung compliance line and the chest-wall compliance

curve. Additional work performed by the respiratory muscles during expiration was calculated as

the portion of the pressure-volume loop positioned to the right of the chest-wall compliance

curve. This process was performed on three esophageal pressure-volume loops at rest, symptom

limitation and the two isotimes.

Data Analysis

A one-way repeated measures analysis of variance (ANOVA) was performed at symptom

limitation, and at two isotimes during the exercise tolerance trials using commercially available

software (Statistica, Statsoft, Oklahoma city, OK). Isotime 1 and 2 were defined as symptom

limitation in the air and HOX trials, respectively. When a patient exercised longer on air than on

an experimental gas (n=1) or shorter on HE-HOX than HOX (n=2) the end exercise responses

were carried forward. When the ANOVA detected a significant effect, a Tukey post hoc multiple

comparisons test was performed.

To test for associations between the change in exercise time and changes in EELV, VE,

work of breathing, and/or dyspnea, simple regression analysis using Pearson correlations were

7

performed. In addition, the strongest significant contributors to the improvement in exercise time

were selected by multiple stepwise regression analysis. For all analyses and post hoc comparisons

the alpha level was set a priori at 0.05.

RESULTS

Symptom-Limited Exercise Tolerance

All results are reported as mean ± SD. The effect of each gas mixture on exercise tolerance is

depicted in Figure 1. An increase in exercise time was observed with all three gas mixtures

compared to breathing air. Exercise time to symptom limitation was 9.4 ± 5.2 min on air, 17.8 ±

5.8 min on HOX and 16.7 ± 9.1 min on HE-OX. The combination of helium and hyperoxia had a

significantly greater effect on exercise tolerance compared to all other gases (26.3 ± 10.6 min).

During exercise, dyspnea was decreased with all three experimental gases at isotime 1

(Figure 2A, Table 3). At this time point, the reduction in dyspnea was greatest with HE-HOX as

the Borg rating was reduced from 5.8 ± 2.2 in the air trial to 1.9 ± 1.4 with HE-HOX. HE-HOX

also reduced dyspnea at isotime 2 compared to HOX (5.3 ± 2.0 vs. 3.7 ± 1.8, p=0.029).

Concomitant to the reductions in dyspnea, both hyperoxic gases reduced the sensation of leg

discomfort at isotime 1 compared to air (Figure 2B) and HE-HOX also reduced leg discomfort

compared to HE-OX (2.2 ± 1.5 vs. 3.6 ± 2.2, p=0.049).

At end exercise, dyspnea was significantly lower with HE-OX but unchanged with the

two hyperoxic gases (Table 4). No difference in leg discomfort was observed at symptom

limitation with any gas. In the air trial, dyspnea was the primary symptom limiting exercise in

eight subjects, while two stopped due to leg discomfort. In contrast, leg discomfort was reported

as the principal reason for exercise termination with HOX (n=5), HE-OX (n=4) and HE-HOX

8

(n=7), with four, three and two subjects, respectively, stopping primarily due to dyspnea. Two

subjects reported the combination of leg discomfort and dyspnea as the primary reason for

stopping in the HE-OX trial. One subject stopped due to “other” reasons with each of the

experimental gas mixtures. These reasons included mouthpiece discomfort (HOX), general

fatigue (HE-OX) and discomfort from the bicycle seat (HE-HOX).

Resting Pulmonary Function

Similar resting spirometry was observed with HOX compared to air (Table 5). Both helium-based

mixtures significantly increased peak expiratory flow rates, forced expired volume in one second

(FEV1) and forced expiratory flow at 50% of forced vital capacity (FEF50) without changing

forced vital capacity (FVC) or forced expiratory flow at 75% of FVC (FEF75). The volume of

isoflow was similar between the two helium mixtures occurring at 36 ± 13 and 37 ± 15% of FVC

with HE-OX and HE-HOX, respectively.

Ventilatory Responses to Exercise

At rest, ventilatory parameters were unaffected by any gas (Figure 3). At isotime 1, VE was

unchanged with HE-OX but reduced with both HOX and HE-HOX due to a decrease in breathing

frequency (Table 3, Figure 3). Even with the reduction in VE, tidal volume was increased with

HE-HOX compared to air (p=0.04) but not with HOX (p=0.92). Tidal volume was also increased

with HE-OX compared to both air and HOX. The reduction in breathing frequency with HE-

HOX resulted in a longer TE at isotime 1 (Figure 3E) but no significant difference in TI/TTOT

(inspiratory time/total time) was observed with any gas (Figure 3F). Compared to air at isotime 1,

inspiratory and expiratory flow rates were unchanged with HE-OX. In contrast, peak inspiratory

9

and expiratory flow rates were decreased with both hyperoxic gases (Table 3) while only HOX

significantly reduced mean inspiratory and expiratory flow rates.

At symptom limitation, there was no difference in the ventilatory responses between air and

hyperoxia. In contrast, VE was increased with HE-OX compared to all other gas mixtures due to a

larger tidal volume (Table 4). There was considerable variation in the individual ventilatory

responses observed with HE-OX as VE ranged from a decrease of 1.8 L.min-1 to an increase of

25.8 L.min-1. This finding was strongly correlated with resting FEV1 (r = 0.78, p = 0.008) such

that those with the greatest FEV1 showed the largest increases in VE with HE-OX. At end

exercise, VE was similar to air with HE-HOX (p = 0.94) while tidal volume remained increased

by 0.15L (p = 0.04).

Lung Volume Responses to Exercise

Baseline IC, EILV and inspiratory reserve volume (IRV) were unchanged with any gas (Figure

4). From quiet breathing at rest to symptom limitation, all patients dynamically hyperinflated as

demonstrated by the significant reduction in IC. The rise in EELV during exercise resulted in a

mean PEEPi at the end of the air trial of 4.4 ± 1.7 cm H2O (range 2.6 – 7.7 cm H2O). At symptom

limitation with air, the degree of dynamic hyperinflation averaged 0.43 ± 0.18 L (range = 0.17 L

to 0.69 L) and EILV reached 94% of TLC. No difference in IC, EILV, IRV or PEEPi was

observed with any gas and the increased tidal volume with both helium gas mixtures was

primarily due to a trend toward a lower EELV.

At isotime 1, IC was significantly increased with both HE-OX and HE-HOX but not HOX

(Figure 4, Table 3). Concomitant to the reduction in dynamic hyperinflation with the helium

gases, PEEPi was also reduced from 4.4 ± 1.7 cm H2O in air to 2.4 ± 2.1 and 2.0 ± 1.9 cm H2O

10

with HE-OX and HE-HOX, respectively. Interestingly, the decrease in dynamic hyperinflation

with both helium gases was not associated with a significant reduction in EILV as tidal volume

was increased at the expense of IRV. Helium-hyperoxia also had a greater effect on dynamic

hyperinflation than HOX, as IC was greater (2.45 ± 0.85 L) at isotime 2 compared to HOX (2.27

± 0.82 L, p=0.03).

Respiratory Mechanics During Exercise

The work of breathing was measured in eight subjects and is presented in Figure 5. At rest, both

helium mixtures reduced the resistive work of breathing compared to air (Figure 5A).

Additionally, the total elastic work of inspiration was reduced with HE-HOX (p=0.011)

predominantly due to a reduction in WINP, as the work to overcome PEEPi was not different

between conditions (Figure 5B-C). When the total work of breathing per minute was calculated,

only HE-HOX was lower than with air (291 vs. 165 L·cm H2O-1.min-1 in air and HE-HOX,

respectively; p=0.008)

During the first five minutes of exercise, all three experimental gas mixtures reduced the

resistive work of breathing and the work to overcome PEEPi. By isotime 1, only the 33.3 ± 17.4

% reduction in the resistive work of breathing and the 51.9 ± 29.8% decrease in WIP with HE-

HOX remained lower than air. At isotime 1 there was no significant difference in WINP with any

gas even though tidal volume was increased with the helium mixtures. When the total resistive

and elastic components of the work of breathing were averaged over one minute, both parameters

were reduced with the two hyperoxic gas mixtures, due to the significant reductions in breathing

frequency. However, the total work of breathing was only reduced at isotime 1 with HE-HOX

11

(940 ± 317 L·cm H2O-1·min-1) compared to air (1938 ± 1457 L·.cm H2O-1·min-1). At isotime 2 and

symptom limitation, there were no significant differences in the work of breathing with any gas.

Correlates of Improved Exercise Tolerance

The increase in exercise time to symptom limitation with HOX significantly correlated with the

decrease in VE (r = -0.82) and breathing frequency (r = -0.67) at isotime 1. The increase in

exercise time with HE-OX correlated best with the change in peak inspiratory flow (r = 0.93),

EELV expressed as a %TLC (r = -0.93), inspiratory capacity (r = 0.89), breathing frequency (r =

-0.88) and total work of breathing per minute (r = -0.85). Stepwise multiple regression analysis

demonstrated that the combination of improved peak inspiratory flow, EELV as a %TLC and the

reduction in the total work of breathing explained 99% of the variance associated with increased

endurance time with HE-OX (r2 = 0.99, p<0.0001).

The improved exercise time with HE-HOX was associated with a reduction in the work of

breathing and improved inspiratory flow rates. The strongest correlates of improved exercise time

with HE-HOX were the change in total elastic work of breathing per minute (r = -0.87), total

inspiratory resistive work of breathing per minute (r = -0.83), peak inspiratory (r = 0.82) and

expiratory flow rates (r = -0.81) and mean expiratory flow rate (r = -0.79). In addition,

significant correlations were also observed between the improvement in exercise time and the

decrease in VE (r = -0.71) and breathing frequency (r = 0.68) with HE-HOX. Stepwise multiple

regression analysis of these variables, demonstrated that the decrease in the total elastic work of

breathing with HE-HOX explained 75% of the variance in exercise time. The favorable decrease

in mean expiratory flow rate then added approximately 8% to the explained variance (r2 =0 .83, p

<0.01).

12

DISCUSSION

The principal finding of this study supported our hypotheses that helium-hyperoxia would

improve exercise tolerance to a greater degree than hyperoxia or normoxic-helium. As postulated,

these improvements in exercise tolerance were related to the greater effect of this gas for delaying

dynamic hyperinflation, alleviating dyspnea and reducing the work of breathing.

Symptom Limited Exercise Tolerance

Hyperoxia has consistently been reported to improve the exercise capacity of COPD patients with

(6, 14, 24) and without hypoxemia (13, 25). More recently, Palange et al., (16) reported that

normoxic-helium improved the exercise tolerance of COPD patients performing high intensity

cycle exercise by 114%. The improvements in exercise time with HOX (118 ± 74%) and HE-OX

(91 ± 103%) found in the present study are similar to those previously reported. However, the

primary finding of this investigation was that HE-HOX improved the exercise tolerance of COPD

patients by 245 ± 208% compared to air and by 54 ± 56 and 92 ± 116% compared to HOX and

HE-OX, respectively. These findings complement those of Laude et al., (26) who demonstrated

that a 28:72 O2-He mixture improved walking distance in a shuttle walk test by 64% in patients

with COPD. However, our results demonstrate how a combination of helium and hyperoxia has a

greater effect on reducing dynamic hyperinflation and work of breathing than either normoxic-

helium or hyperoxia alone, which explains the underlying physiological mechanisms responsible

for the improved exercise tolerance.

In the present study, the majority of subjects terminated exercise because of dyspnea

when breathing air. However, leg discomfort was the principal reason for stopping exercise with

HE-HOX. This change in the symptom responsible for limiting exercise suggests that HE-HOX

13

decreases the ventilatory constraints associated with COPD to an extent that skeletal muscle

function becomes more of a limiting factor.

Ventilation and Lung Volume Responses to Exercise

The significant reduction in VE observed with HOX is consistent with others (6, 13, 14, 24) and

has been attributed to reduced chemoreceptor drive (6, 24). As a result, inspiratory and

expiratory flow rates were decreased, EELV was maintained closer to resting lung volumes and

the progression of dynamic hyperinflation was delayed, which improved exercise tolerance,

independent of the level of hypoxemia (r = 0.28, p = 0.48). At symptom limitation, the magnitude

of dynamic hyperinflation, VE and tidal volume were all unchanged from air, which supports the

previous findings of O’Donnell et al., (13) and indicates that the ventilatory constraints to

exercise are delayed but not removed with hyperoxia.

A number of previous studies have reported that normoxic-helium increases VE during

exercise in patients with COPD (5, 16, 27). In the present study VE was increased 16 ± 21% with

HE-OX, which is comparable to that reported by others (28, 29). Improvements in IC with HE-

OX were primarily due to an increase in the maximal flow volume envelope as inspiratory and

expiratory flow rates were similar to that observed with air but EELV was reduced. A significant

correlation was observed between the change in EELV as a %TLC and the change in dyspnea

with HE-OX (r = 0.71, p = 0.02), which is not surprising as EELV was one of the primary

predictors of improved exercise performance with this gas. Interestingly, no relationship was

observed between the increase in maximal VE with HE-OX and exercise time (r = 0.10, p = 0.79),

which suggests that the ability of the HE-OX to reduce dynamic hyperinflation is more important

than its effect on VE for improving exercise tolerance.

14

The ventilatory response to exercise with HE-HOX was similar to HOX, with VE and

breathing frequency being reduced. Lower expiratory flow rates combined with a significantly

lengthened TE reduced EELV to a greater extent than HOX and increased tidal volume. At

symptom limitation, IC was not significantly increased with HE-HOX but tidal volume was still

augmented due again to the slight but non-significant increase in TE (r = 0.65, p = 0.04). These

positive effects of HE-HOX on lung volumes demonstrate the additive benefit of combining

helium and 40% O2, as the increase in FIO2 reduced expiratory flow rates and increased

expiratory time while helium increased the maximal flow-volume envelope allowing for a greater

reduction in dynamic hyperinflation.

Respiratory Mechanics During Exercise

The effect of each gas on the resistive work of breathing was evident early in the constant

load trials and was related to the physical properties of both helium and oxygen. The lower

density of helium reduces the pressure needed to overcome airway resistance at higher flow rates

by maintaining laminar flow, whereas, oxygen reduces ventilatory drive and expiratory flow rate

which decreases airway resistance. The further reduction in resistive work only with helium-

hyperoxia at isotime 1 demonstrates that there is a synergistic effect of combining these two

gases on airway resistance that is related to the reduced density of helium and the ability of

oxygen to decrease VE.

To interpret the different elastic work of breathing responses observed in this study it is

also important to consider the effects of helium and oxygen (both separately and in combination)

on lung volumes and VE during exercise. At isotime 1, HOX decreased the total inspiratory

elastic work per minute primarily due to a reduction in ventilatory drive as EELV and tidal

volume were unchanged compared to air. In contrast, EELV was decreased with HE-OX but the

15

total inspiratory elastic work was unaffected as tidal volume was increased at the expense of

maintaining the inspiratory work needed to overcome the elastic recoil of the lung. HE-HOX

combined these effects as the total elastic work of breathing was reduced even with an increase in

tidal volume. This finding can be attributed to both the greater reduction in dynamic

hyperinflation, which decreased the work to overcome PEEPi, and the hyperoxia induced

reduction in VE.

The significant relationship observed between the reduced work of breathing and the

improved exercise time with HE-HOX is likely due to alleviation of dyspnea, since a significant

relationship was observed between the change in the total work of breathing and dyspnea

reduction (r=0.63, p=0.048). However, another benefit of the reduced work of breathing may be

a better distribution of cardiac output to the exercising muscles. In healthy individuals the

respiratory muscles can demand 13-15% of total body oxygen consumption during heavy

exercise (30) and 14-16% of total cardiac output (31). Reducing the work of breathing decreases

respiratory muscle demand and has been shown to increase leg blood flow during exercise (32).

The increased work of breathing in patients with COPD is estimated to govern as much as 35-

40% of total body oxygen consumption (33), which would demand a much greater proportion of

cardiac output. Therefore, it is possible that unloading the respiratory system with HE-HOX

allowed a greater portion of total cardiac output to be distributed to the working muscles.

Improved leg blood flow combined with a higher oxyhemoglobin saturation would increase O2

supply and decrease metabolic acidosis (13), which would conceivably reduce leg fatigue and

improve exercise tolerance.

A potential limitation of the present methodology could be an inaccurate positioning of

the static chest wall compliance curve in our calculation of the elastic work of breathing. The

curve is positioned assuming that end-expiratory lung volume during quiet breathing is an

16

accurate representation of the relaxation volume and pressure of the respiratory system. However,

this may not be accurate as dynamic hyperinflation is often present in COPD patients at rest and

EELV may actually be above relaxation volume even during quiet breathing. Continued

expiratory muscle recruitment at end expiration can also artificially increase PEEPi above that

due to dynamic hyperinflation, which would increase end-expiratory pressure independent of the

chest wall elasticity. To correct for this, previous studies have subtracted the expiratory rise in

gastric pressures from the esophageal pressure at EELV (11, 22). This could not be done in this

study as gastric pressures were not measured. As a result the work needed to overcome PEEPi

may be slightly overestimated. However, only four subjects demonstrated evidence of PEEPi at

rest and no significant difference was observed in EELV or PEEPi with the four gas mixtures at

baseline. Therefore, any error in WIP would be consistent across the four conditions and

considering the magnitude of the reduction in WIP observed with HE-HOX (52 ± 30%), we

believe that a possible overestimation in WIP did not affect our conclusions.

In summary, combining 40% O2 with helium greatly increased exercise tolerance in

patients with moderate to severe COPD due to the additional benefits of this gas on lung

mechanics, dynamic hyperinflation and dyspnea alleviation. The ability of helium-hyperoxia to

acutely reduce dyspnea and improve exercise tolerance are important clinical findings and could

be useful within rehabilitation practice. It is now well recognized that exercise training as a part

of pulmonary rehabilitation can lead to important improvements in exercise capacity, dyspnea

reduction and health related quality of life (34, 35). However, patients who suffer from

debilitating dyspnea often cannot perform a sufficient volume of exercise to gain these benefits.

Breathing helium-hyperoxia during exercise should reduce dyspnea and increase the ability of

17

patients with COPD to exercise, which could translate into improved outcomes from a pulmonary

rehabilitation program.

18

Acknowledgements The authors would like to thank The Caritas Centre for Lung Health for their support with patient

recruitment and Chris Sellar and Tim Hartley for providing assistance with data collection

throughout this study.

19

References

1. Jones, N. L., G. Jones, and R. H. Edwards. 1971. Exercise tolerance in chronic airway

obstruction. Am Rev Respir Dis 103(4):477-91.

2. O'Donnell, D. E., S. M. Revill, and K. A. Webb. 2001. Dynamic hyperinflation and exercise

intolerance in chronic obstructive pulmonary disease. Am J Respir Crit Care Med 164(5):770-7.

3. O'Donnell, D. E., and K. A. Webb. 1993. Exertional breathlessness in patients with chronic

airflow limitation. The role of lung hyperinflation. Am Rev Respir Dis 148(5):1351-7.

4. Mitlehner, W., and W. Kerb. 1994. Exercise hypoxemia and the effects of increased

inspiratory oxygen concentration in severe chronic obstructive pulmonary disease. Respiration

61(5):255-62.

5. Oelberg, D. A., R. M. Kacmarek, P. P. Pappagianopoulos, L. C. Ginns, and D. M. Systrom.

1998. Ventilatory and cardiovascular responses to inspired He-O2 during exercise in chronic

obstructive pulmonary disease. Am J Respir Crit Care Med 158(6):1876-82.

6. O'Donnell, D. E., D. J. Bain, and K. A. Webb. 1997. Factors contributing to relief of

exertional breathlessness during hyperoxia in chronic airflow limitation. Am J Respir Crit Care

Med 155(2):530-5.

7. Diaz, O., C. Villafranca, H. Ghezzo, G. Borzone, A. Leiva, J. Milic-Emil, and C. Lisboa.

2000. Role of inspiratory capacity on exercise tolerance in COPD patients with and without tidal

expiratory flow limitation at rest. Eur Respir J 16(2):269-75.

8. Marin, J. M., S. J. Carrizo, M. Gascon, A. Sanchez, B. Gallego, and B. R. Celli. 2001.

Inspiratory capacity, dynamic hyperinflation, breathlessness, and exercise performance during the

6-minute-walk test in chronic obstructive pulmonary disease. Am J Respir Crit Care Med

163(6):1395-9.

20

9. Yan, S., and B. Kayser. 1997. Differential inspiratory muscle pressure contributions to

breathing during dynamic hyperinflation. Am J Respir Crit Care Med 156(2 Pt 1):497-503.

10. Sharp, J. T. 1983. The respiratory muscles in emphysema. Clin Chest Med 4(3):421-32.

11. Sliwinski, P., D. Kaminski, J. Zielinski, and S. Yan. 1998. Partitioning of the elastic work of

inspiration in patients with COPD during exercise. Eur Respir J 11(2):416-21.

12. Leblanc, P., E. Summers, M. D. Inman, N. L. Jones, E. J. Campbell, and K. J. Killian. 1988.

Inspiratory muscles during exercise: a problem of supply and demand. J Appl Physiol

64(6):2482-9.

13. O'Donnell, D. E., C. D'Arsigny, and K. A. Webb. 2001. Effects of hyperoxia on ventilatory

limitation during exercise in advanced chronic obstructive pulmonary disease. Am J Respir Crit

Care Med 163(4):892-8.

14. Somfay, A., J. Porszasz, S. M. Lee, and R. Casaburi. 2001. Dose-response effect of oxygen

on hyperinflation and exercise endurance in nonhypoxaemic COPD patients. Eur Respir J

18(1):77-84.

15. Emtner, M., J. Porszasz, M. Burns, A. Somfay, and R. Casaburi. 2003. Benefits of

Supplemental Oxygen in Exercise Training in Nonhypoxemic Chronic Obstructive Pulmonary

Disease Patients. Am. J. Respir. Crit. Care Med. 168(9):1034-1042.

16. Palange, P., G. Valli, P. Onorati, R. Antonucci, P. Paoletti, A. Rosato, F. Manfredi, and P.

Serra. 2004. Effect of heliox on lung dynamic hyperinflation, dyspnea, and exercise endurance

capacity in COPD patients. J Appl Physiol 97(5):1637-42.

17. Eves, N. D., Petersen SR, Haykowsky MJ, Wong EY, Jones RL. 2004. The effects of helium

and hyperoxia on exercise tolerance and ventilatory mechanics in chronic obstructive pulmonary

disease. Canadian Journal of Applied Physiology 29((Suppl)):S45.

21

18. Jaber, S., R. Fodil, A. Carlucci, M. Boussarsar, J. Pigeot, F. Lemaire, A. Harf, F. Lofaso, D.

Isabey, and L. Brochard. 2000. Noninvasive ventilation with helium-oxygen in acute

exacerbations of chronic obstructive pulmonary disease. Am J Respir Crit Care Med 161(4 Pt

1):1191-200.

19. Stubbing, D. G., L. D. Pengelly, J. L. Morse, and N. L. Jones. 1980. Pulmonary mechanics

during exercise in subjects with chronic airflow obstruction. J Appl Physiol 49(3):511-5.

20. Yan, S., D. Kaminski, and P. Sliwinski. 1997. Reliability of inspiratory capacity for

estimating end-expiratory lung volume changes during exercise in patients with chronic

obstructive pulmonary disease. Am J Respir Crit Care Med 156(1):55-9.

21. Dolmage, T. E., and R. S. Goldstein. 2002. Repeatability of inspiratory capacity during

incremental exercise in patients with severe COPD. Chest 121(3):708-14.

22. Yan, S., D. Kaminski, and P. Sliwinski. 1997. Inspiratory muscle mechanics of patients with

chronic obstructive pulmonary disease during incremental exercise. Am J Respir Crit Care Med

156(3 Pt 1):807-13.

23. Estenne, M., A. Heilporn, L. Delhez, J. C. Yernault, and A. De Troyer. 1983. Chest wall

stiffness in patients with chronic respiratory muscle weakness. Am Rev Respir Dis 128(6):1002-7.

24. Dean, N. C., J. K. Brown, R. B. Himelman, J. J. Doherty, W. M. Gold, and M. S. Stulbarg.

1992. Oxygen may improve dyspnea and endurance in patients with chronic obstructive

pulmonary disease and only mild hypoxemia. Am Rev Respir Dis 146(4):941-5.

25. Fujimoto, K., Y. Matsuzawa, S. Yamaguchi, T. Koizumi, and K. Kubo. 2002. Benefits of

oxygen on exercise performance and pulmonary hemodynamics in patients with COPD with mild

hypoxemia. Chest 122(2):457-63.

26. Laude, E. A., N. C. Duffy, C. Baveystock, B. Dougill, M. J. Campbell, R. Lawson, P. W.

Jones, and P. M. Calverley. 2006. The Effect of Helium and Oxygen on Exercise Performance in

22

Chronic Obstructive Pulmonary Disease: A Randomized Crossover Trial. Am. J. Respir. Crit.

Care Med. 173(8):865-870.

27. Richardson, R. S., J. Sheldon, D. C. Poole, S. R. Hopkins, A. L. Ries, and P. D. Wagner.

1999. Evidence of skeletal muscle metabolic reserve during whole body exercise in patients with

chronic obstructive pulmonary disease. Am J Respir Crit Care Med 159(3):881-5.

28. Babb, T. G. 1997. Ventilatory response to exercise in subjects breathing CO2 or HeO2. J

Appl Physiol 82(3):746-54.

29. Esposito, F., and G. Ferretti. 1997. The effects of breathing He-O2 mixtures on maximal

oxygen consumption in normoxic and hypoxic men. J Physiol 503 ( Pt 1):215-22.

30. Aaron, E. A., K. C. Seow, B. D. Johnson, and J. A. Dempsey. 1992. Oxygen cost of exercise

hyperpnea: implications for performance. J Appl Physiol 72(5):1818-25.

31. Harms, C. A., T. J. Wetter, S. R. McClaran, D. F. Pegelow, G. A. Nickele, W. B. Nelson, P.

Hanson, and J. A. Dempsey. 1998. Effects of respiratory muscle work on cardiac output and its

distribution during maximal exercise. J Appl Physiol 85(2):609-18.

32. Harms, C. A., M. A. Babcock, S. R. McClaran, D. F. Pegelow, G. A. Nickele, W. B. Nelson,

and J. A. Dempsey. 1997. Respiratory muscle work compromises leg blood flow during maximal

exercise. J Appl Physiol 82(5):1573-83.

33. Evison, H., and R. M. Cherniack. 1968. Ventilatory cost of exercise in chronic obstructive

pulmonary disease. J Appl Physiol 25(1):21-7.

34. Troosters, T., R. Casaburi, R. Gosselink, and M. Decramer. 2005. Pulmonary Rehabilitation

in Chronic Obstructive Pulmonary Disease. Am. J. Respir. Crit. Care Med. 172(1):19-38.

35. Wedzicha, J. A. 2006. Heliox in Chronic Obstructive Pulmonary Disease: Lightening the

Airflow. Am. J. Respir. Crit. Care Med. 173(8):825-826.

23

36. 2003. ATS/ACCP Statement on Cardiopulmonary Exercise Testing. Am. J. Respir. Crit. Care

Med. 167(2):211-277.

24

Figure Legends Figure 1. The effect of helium and oxygen gas mixtures on exercise tolerance. = mean values, = 95%

confidence intervals. p values for each comparison are also presented. Helium-hyperoxia significantly enhanced the

exercise tolerance of COPD patients compared to the other gas mixtures.

Figure 2. Borg ratings of dyspnea and leg discomfort during the exercise tolerance tests. Significant differences

between each gas at the two isotimes and symptom limitation are depicted on the graph. During exercise, dyspnea

was reduced with all experimental gas mixtures but the greatest effect was with HE-HOX. At end exercise, only HE-

OX significantly reduced the rating of dyspnea compared to air (A). Leg discomfort was reduced with both

hyperoxic gases at isotime one but no differences existed at end exercise for any gas. Values represent means ± SD,

p<0.05. * = AIR vs. HOX; † = AIR vs. HE-OX; ‡ = AIR vs. HE-HOX; § = HOX vs. HE-OX; ll = HOX vs. HE-

HOX; ** = HE-OX vs. HE-HOX.

Figure 3. Ventilatory responses to exercise are depicted in panels A to F. Ventilation was significantly reduced with

both the hyperoxic mixtures during exercise and was increased with HE-OX at symptom limitation (A). Tidal

volume was greater with the helium-based gases but not HOX (B). Breathing frequency was reduced with HOX and

HE-HOX (C). No significant changes in inspiratory time were observed at either isotime or end exercise with any

gas (D). HE-HOX lengthened expiratory time at isotime one (E) and duty cycle was reduced with HE-OX at

symptom limitation (F). Values represent means ± SD, p<0.05. * = AIR vs. HOX; † = AIR vs. HE-OX; ‡ = AIR vs.

HE-HOX; § = HOX vs. HE-OX; ll = HOX vs. HE-HOX; ** = HE-OX vs. HE-HOX.

Figure 4. Lung volume responses to constant-load exercise. Open shapes = end inspiratory lung volumes, filled

shapes = end expiratory lung volumes. = air, = hyperoxia, = normoxic-helium and = helium-hyperoxia.

End expiratory lung volume was reduced at isotime one with both helium mixtures but not HOX. HE-HOX also

decreased EELV at isotime two compared to HOX. End-inspiratory lung volume was unchanged with any gas at

either isotime or end exercise. Values represent means ± SD, p<0.05. * = AIR vs. HOX; † = AIR vs. HE-OX; ‡ =

AIR vs. HE-HOX; § = HOX vs. HE-OX; ll = HOX vs. HE-HOX; ** = HE-OX vs. HE-HOX.

25

Figure 5. Work of breathing responses to constant-load exercise. Inspiratory resistive work (WIR) was only reduced

with HE-HOX at isotime 1 (A). The work to overcome the elastic recoil of the lung (WINP) was not reduced with the

helium mixtures due to an enhanced tidal volume (B). The work to overcome PEEPi (WIP) was only significantly

reduced at isotime one with HE-HOX due to a greater reduction in dynamic hyperinflation (C). Additional expiratory

work (WE) performed by the respiratory muscles was not affected by any gas at either isotime or symptom limitation

(D). Significant reductions in the total inspiratory elastic work per minute (WIE) were observed at isotime one with

both hyperoxic gases (E). The total work of breathing was only reduced at isotime 1 with HE-HOX (F). Values

represent means ± SD, n=8, p<0.05. * = AIR vs. HOX; † = AIR vs. HE-OX; ‡ = AIR vs. HE-HOX; § = HOX vs.

HE-OX; ll = HOX vs. HE-HOX; ** = HE-OX vs. HE-HOX.

26

TABLE 1: PATIENT CHARACTERISTICS Characteristic Value %Pred Age, yr 65 ± 11 Height, cm 179 ± 5 Mass, kg 82 ± 15 Body Mass Index, kg/m2 26 ± 5 FEV1, L 1.66 ± 0.59 47 ± 17 FVC, L 3.81 ± 0.99 83 ± 21 FEV1/FVC, % 42.8 ± 8.0 56 ± 10 TLC, L 9.04 ± 1.44 136 ± 20 RV, L 4.82 ± 1.60 192 ± 63 FRC, L 5.87 ± 1.63 157 ± 44 IC, L 3.18 ± 0.97 DLCO, ml/min/mm Hg 19.0 ± 4.4 71 ± 16 SRaw, cm H2O/L/s 4.36 ± 0.22 PaO2, mm Hg 68.3 ± 6.4 PaCO2, mm Hg 36.7 ± 3.3 PH 7.43 ± 0.02 HCO3, mmol/L 23.7 ± 1.8 Hb, g/dL 14.7 ± 1.2 Hct, % 45.0 ± 3.6

Definition of abbreviations: FEV1 = forced expired volume in one second, FVC = forced vital capacity, TLC – total lung capacity, RV = residual volume, FRC = functional residual capacity, IC = inspiratory capacity, DLCO =diffusion capacity of the lung for carbon monoxide, SRaw = specific airway resistance, PaO2 = partial pressure of arterial oxygen, PaCO2 = partial pressure of arterial carbon dioxide, HCO3 = bicarbonate, Hb = hemoglobin concentration, Hct = hematocrit. Values are means ± SD, n=10.

27

TABLE 2: INCREMENTAL EXERCISE RESULTS Value VO2, ml.kg-1.min-1 18.5 ± 4.9 VO2, l.min-1 1.52 ± 0.4 VO2, %pred 59 ± 17 PO, W 118 ± 37 PO, %pred 61 ± 17 HR, 130 ± 17 HR, %pred 84 ± 13 SpO2, % 88 ± 4 ∆SpO2, % 6 ± 4 RER 1.04 ± 0.09 VE, l.min-1 57.6 ± 21.9 VE, %pred 98 ± 44 VT, l 1.80 ± 0.55 F 32 ± 9 Dyspnea 6.7 ± 1.8 Leg Discomfort 5.3 ± 2.1 Reason for stopping Dyspnea 5 Leg Discomfort 1 Both 4

Definition of abbreviations: VO2 = peak oxygen consumption, PO = power output, HR = heart rate, SPO2 = oxyhemoglobin saturation, ∆SpO2 = change in SpO2 from resting values, RER = respiratory exchange ratio, VE = ventilation, VT = tidal volume, f = breathing frequency, n=10. Predicted values from reference (36).

28

TABLE 3: ISOTIME RESPONSES TO EXERCISES AIR HOX HE-OX HE-HOX VE, L⋅min-1 52.7 ± 22.4†§ 43.5 ± 12.6*‡ 54.0 ± 20.7†§ 44.4 ± 13.7*‡ VT, L 1.69 ± 0.55‡§ 1.72 ± 0.55‡ 1.90 ± 0.44*† 1.85 ± 0.49* F, 32.3 ± 12.6†§ 25.9 ± 5.3* 27.8 ± 6.9 23.9 ± 3.8* VCO2, L⋅min-1 1.30 ± 0.40 1.22 ± 0.33 1.32 ± 0.40 1.24 ± 0.35 PETCO2, mm Hg 34.9 ± 7.2†‡ 38.9 ± 5.1*‡§ 31.2 ± 4.3*†§ 35.2 ± 3.7†‡ TI 0.85 ± 0.42 0.91 ± 0.16 0.84 ± 0.20 1.02 ± 0.15 TE 1.27 ± 0.44§ 1.51 ± 0.41 1.45 ± 0.43 1.55 ± 0.31* TI/TTOT 0.39 ± 0.04 0.38 ± 0.03 0.37 ± 0.02 0.40 ± 0.04 IC, L 2.24 ± 0.91‡§ 2.39 ± 0.85 2.51 ± 0.85* 2.53 ± 0.90* IRV, L 0.56 ± 0.39 0.67 ± 0.40 0.60 ± 0.48 0.68 ± 0.49 EILV/TLC, % 93.9 ± 3.8 92.7 ± 3.8 93.4 ± 4.7 92.5 ± 4.9 PEEPi, cm H2O 4.4 ± 1.7‡§ 3.5 ± 3.0 2.4 ± 2.1* 2.0 ± 1.9* Peak Pes Exp, cm H2O 18.0 ± 18.6 10.7 ± 7.0 7.7 ± 5.5 8.9 ± 4.1 Peak Pes Ins, cm H2O -19.9 ± 5.8†‡§ -16.0 ± 4.5* -15.8 ± 4.4* -14.7 ± 5.2* PEF, L⋅s-1 3.00 ± 1.40†§ 2.10 ± 0.63*‡ 3.00 ± 1.04†§ 2.28 ± 0.63*‡ MEF, L⋅s-1 1.53 ± 0.61† 1.22 ± 0.36*‡ 1.62 ± 0.59†§ 1.34 ± 0.41‡ PIF, L⋅s-1 -3.16 ± 0.94†§ -2.09 ± 1.10*‡ -3.17 ± 0.91†§ -2.65 ± 0.62*‡ MIF, L⋅s-1 -2.32 ± 0.79† -1.83 ± 0.37*‡ -2.05 ± 1.45† -2.00 ± 0.54 SpO2, % 89.3 ± 4.4†§ 97.7 ± 1.3*‡ 91.4 ± 2.2†§ 98.4 ± 1.1*‡ Dyspnea 5.8 ± 2.2†‡§ 2.7 ± 0.9* 3.1 ± 1.1* 1.9 ± 1.4* Leg Discomfort 5.0 ± 1.6†§ 2.8 ± 1.6* 3.6 ± 2.2§ 2.2 ± 1.5*‡ Definition of abbreviations: VE = Ventilation, VT = tidal volume, f = breathing frequency, VCO2 = carbon dioxide produced, PETCO2 = partial pressure of end tidal carbon dioxide, TI = inspiratory time, TE = expiratory time, TI/TTOT = ratio of inspiratory time to total time, IC = Inspiratory capacity, IRV = inspiratory reserve volume, EILV = end inspiratory lung volume as a percentage of total lung capacity, PEEPi = intrinsic positive end expiratory pressure, Peak Pes Exp = peak expiratory esophageal pressure during tidal breathing, Peak Pes Ins = Peak inspiratory esophageal pressure during tidal breathing, PEF = peak expiratory flow during tidal breathing, MEF = mean expiratory flow during tidal breathing, PIF = peak inspiratory flow during tidal breathing, MIF = mean inspiratory flow during tidal breathing, SpO2 = oxyhemoglobin saturation.. * = p<0.05 vs. AIR; † = p<0.05 vs. HOX; ‡ = p<0.05 vs. HE-OX; § = p<0.05 vs. HE-HOX.

29

TABLE 4: RESPONSES TO EXERCISES AT SYMPTOM LIMITATION AIR HOX HE-OX HE-HOX VE, L⋅min-1 52.7 ± 22.4‡ 50.5 ± 19.1‡ 61.1 ± 27.4*†§ 54.2 ± 22.5‡ VT, L 1.69 ± 0.55‡§ 1.74 ± 0.56‡ 1.90 ± 0.62*† 1.84 ± 0.64* F, 32.3 ± 12.6 29.8 ± 9.8 32.1 ± 10.6 29.7 ± 7.8 VCO2, L⋅min-1 1.30 ± 040 1.34 ± 0.38 1.39 ± 0.43 1.36 ± 0.41 PETCO2, mm Hg 34.9 ± 7.2†‡ 39.0 ± 5.9*‡§ 30.2 ± 5.4*†§ 34.1 ± 5.1†‡ TI 0.85 ± 0.42 0.83 ± 0.25 0.68 ± 0.22 0.76 ± 0.18 TE 1.27 ± 0.44 1.36 ± 0.39 1.35 ± 0.41 1.40 ± 0.46 TI/TTOT 0.39 ± 0.04‡ 0.38 ± 0.02 0.34 ± 0.06* 0.36 ± 0.03 IC, L 2.24 ± 0.91 2.27 ± 0.82 2.40 ± 0.80 2.35 ± 0.85 IRV, L 0.56 ± 0.39 0.53 ± 0.29 0.51 ± 0.24 0.51 ± 0.26 EILV/TLC, % 93.9 ± 3.8 94.1 ± 2.9 94.4 ± 2.5 94.4 ± 2.7 PEEPi, cm H2O 4.4 ± 1.7‡§ 4.6 ± 2.3 3.4 ± 2.1* 4.1 ± 3.6* Peak Pes Exp, cm H2O 18.0 ± 18.6 17.5 ± 14.6 10.8 ± 7.4 12.8 ± 8.1 Peak Pes Ins, cm H2O -19.9 ± 5.8‡ -18.7 ± 4.2 -16.4 ± 3.8* -18.1 ± 5.01 PEF, L⋅s-1 3.00 ± 1.40 2.69 ± 1.15‡ 3.46 ± 1.50†§ 2.61 ± 1.05‡ MEF, L⋅s-1 1.53 ± 0.61 1.41 ± 0.49‡ 1.68 ± 0.75† 1.48 ± 0.51 PIF, L⋅s-1 -3.16 ± 0.94 -2.69 ± 1.44‡ -3.83 ± 1.36†§ -2.97 ± 0.68‡ MIF, L⋅s-1 -2.32 ± 0.79 -2.22 ± 0.56‡ -2.73 ± 1.05†§ -2.28 ± 0.55‡ SpO2, % 89.2 ± 4.4†§ 97.4 ± 1.2*‡ 91.2 ± 2.8†§ 97.8 ± 1.1*‡ Dyspnea 5.8 ± 2.2‡ 5.4 ± 2.6 4.9 ± 2.7* 5.2 ± 2.1 Leg Discomfort 5.0 ± 1.6 5.4 ± 2.1 5.6 ± 1.7 5.5 ± 2.1

Definition of abbreviations: see Table 3. * = p<0.05 vs. AIR; † = p<0.05 vs. HOX; ‡ = p<0.05 vs. HE-OX; § = p<0.05 vs. HE-HOX.

30

TABLE 5: EFFECT OF GAS MIXTURES ON PULMONARY FUNCTION AIR HOX HE-OX HE-HOX FEV1, L 1.54 ± 0.73 1.58 ± 0.70 1.89 ± 0.82* 1.82 ± 0.77* FVC, L 3.76 ± 1.13 3.73 ± 1.15 3.86 ± 1.18 3.83 ± 1.13 PEF, L.s-1 4.64 ± 1.88 4.88 ± 1.89 5.98 ± 2.29* 5.72 ± 2.06* FEF50, L.s-1 0.78 ± 0.45 0.78 ± 0.46 1.02 ± 0.55* 1.07 ± 0.75* FEF75, L.s-1 0.32 ± 0.17 0.32 ± 0.15 0.37 ± 0.16 0.35 ± 0.14 VisoV, L N/A N/A 1.35 ± 0.56 1.37 ± 0.62 VisoV, %FVC N/A N/A 36 ± 13 37 ± 15 Definition of abbreviations: FEV1 = forced expired volume in one second, FVC = forced vital capacity, PEF =peak expiratory flow, FEF50 = forced expiratory flow at 50% FVC, FEF75 = forced expiratory flow at 75% FVC, VisoV = volume of isoflow expressed as an absolute value and as a %FVC from residual volume. Values are means ± SD, n=10. * = p>0.05 vs. AIR.

31

Gas Mixture

AIR HOX HE-OX HE-HOX

Exe

rcis

e Ti

me

(Min

)

0

10

20

30

40

50# 1# 2# 3# 4# 5# 6# 7# 8# 9# 10

p = 0.019

p = 0.976

p = 0.007p = 0.048

p = 0.019

p = 0.0002Patient

Figure 1

32

0 10 20 30

Dys

pnea

(Bor

g U

nits

)

0

2

4

6

8

IT1 IT2

*†‡

*†‡

Exercise Time (Min)

0 10 20 30

Leg

Dis

com

fort

(Bor

g U

nits

)

0

2

4

6

8

AIRHOXHE-OXHE-HOX

IT1 IT2

‡

†

ll

A

*‡**B

Figure 2

33

Exercise Time (Min)

0 10 20 30

T I/T

TOT

0.25

0.30

0.35

0.40

0.45

0.50

0.55AIRHOXHE-OXHE-HOX

Exercise Time (Min)

0 10 20 30

Exp

irato

ry T

ime

(s)

0.5

1.0

1.5

2.0

2.5

3.0

3.5

0 10 20 30

Tida

l Vol

ume

(L)

0.5

1.0

1.5

2.0

2.5

IT1 IT2

0 10 20 30

Ven

tilat

ion

(L. m

in-1

)

0

15

30

45

60

75

90

0 10 20 30

F (b

reat

hs. m

in- 1

)

0

10

20

30

40

50

IT1 IT2

Exercise Time (Min)

0 10 20 30

Insp

irato

ry T

ime

(s)

0.0

0.5

1.0

1.5

2.0

2.5

3.0

IT1 IT2

IT1 IT2IT1 IT2 IT1 IT2

*‡§***‡§**

†§**†§ †‡§

ࠤ

*‡

*‡

‡

†

‡

‡*

A B C

D E F

Figure 3

34

Exercise Time (Min)0 10 20 30

Lung

Vol

umes

(%TL

C)

60

70

80

90

100IRV

VT

TLC

EILV

EELV

‡

*†‡ †‡ ll

Figure 4

35

0 10 20 30

WIR

(L. c

m H

2O-1

)

0

5

10

15

20

25

0 10 20 30

WIN

P (L

. cm

H2O

-1)

0

10

20

30

40

50

AIRHOXHE-OXHE-HOX

0 10 20 30

WIP

(L. c

m H

2O-1

)

0

2

4

6

8

10

12

Exercise Time (Min)

0 10 20 30

Tota

l Wor

k (L

.. cm

H2O

-1. m

in-1

)

0

1000

2000

3000

4000

Exercise Time (Min)

0 10 20 30

WIE

(L.. c

m H

2O-1

. min

-1)

0

500

1000

1500

2000

IT1 IT2 IT1 IT2 IT1 IT2

IT1 IT2

Exercise Time (Min)0 10 20 30

WE

( L.c

. cm

H2O

-1)

0

5

10

15

20

IT1 IT2 IT1 IT2

†‡

*†‡‡

‡

*†‡

‡

†

†

‡

‡

*†‡

*†‡‡

*‡

A B C

D E F

Figure 5

E1

ONLINE DATA SUPPLEMENT

Helium-Hyperoxia, Exercise Tolerance and Respiratory Mechanics in Chronic Obstructive Pulmonary Disease

Neil D. Eves, Stewart R. Petersen, Mark J. Haykowsky, Eric Y. Wong and Richard L. Jones

E2

METHODS

Patients

Ten clinically stable males with moderate to severe COPD (FEV1/FVC= 56 ± 10% predicted)

volunteered to participate in the study after completing a pulmonary rehabilitation program.

Individuals dependent on supplemental oxygen, with cardiovascular disease and/or

musculoskeletal abnormalities were excluded. Nine of the ten had a smoking history (56 ± 36

mean pack years). All aspects of the research protocol were explained and patients signed an

informed consent that had received institutional ethics review board approval.

Study Design

The study was a randomized crossover design, which required three separate visits to the

laboratory at the University of Alberta. The first visit consisted of a pulmonary function test with

blood gas analysis to confirm the severity of COPD and a symptom-limited incremental exercise

test to ensure the absence of cardiovascular contraindications to exercise. In the other two visits,

four constant-load symptom-limited exercise studies were performed in a randomized order (two

per visit) to examine the effect of each gas on exercise tolerance, work of breathing, lung

volumes and exertional symptoms. The patients were asked to refrain from exercise in the 24

hours before a test and to avoid smoking, alcohol and caffeine on testing days. Patients were

allowed to continue their normal bronchodilator therapy throughout the study and tests were

performed at consistent time of day. Routine spirometry was also made each day on arrival at the

laboratory to ensure pulmonary function remained consistent across testing days.

The four gas mixtures were: air (21% O2, 79% N2); hyperoxia (40% O2, 60% N2);

normoxic-helium (21% O2, 79% He) and helium-hyperoxia (40% O2, 60% He). This design

E3

allowed us to investigate the individual and combined effects of helium and 40% O2.

Throughout exercise, humidified gases were passed into a reservoir bag and supplied through a

low-resistance two-way breathing valve (2700 series, Hans Rudolph, Kansas City, MO). The

patients were blinded to the gas mixture used and were asked not to talk during, or for a short

period after exercise due to the change in vocal tone with helium.

Pulmonary Function Testing

Routine pulmonary function variables and single-breath diffusion capacity for carbon monoxide

(DLCO) were measured in the sitting position according to American Thoracic Society guidelines

(E1). Lung volumes were determined using a constant-volume body plethysmograph (6200

Autobox; SensorMedics, Yorba Linda CA). Total lung capacity was then calculated as the sum

of functional residual capacity and inspiratory capacity. Spirometry values were compared to the

reported norms of Crapo et al., (E2), DLCO was compared to the norms of Miller et al., (E3), and

lung volume values was compared to Goldman and Becklake (E4). Additionally, resting arterial

blood gases (PaO2 and PaCO2), and pH were measured from a radial artery at rest while breathing

room air.

Incremental Exercise Test

Prior to inclusion in the study, a physician-supervised incremental exercise test to symptom

limitation was performed with 12-lead ECG monitoring to determine peak oxygen consumption

(VO2peak) and to confirm that participants were free from any cardiovascular contraindications to

exercise. Participants performed the test on an electrically braked cycle ergometer (Ergoline

800S, SensorMedics, Yorba Linda, CA) and expired gases were analyzed by a calibrated

E4

metabolic measurement system (TrueOne 2400, Parvo Medics, Salt Lake City, UT). After stable

resting metabolic values were achieved, subjects cycled at 10W and the load was increased by 5-

10 W.min-1 until symptom-limitation. This protocol has been demonstrated to be appropriate for

measuring VO2peak in this population (E5). During exercise, oxyhemoglobin saturation and heart

rate were monitored continuously, while blood pressure was measured every two minutes. At

the end of each workload, exertional dyspnea and leg discomfort were evaluated using the Borg

scale (E6). On termination of the exercise test, participants were asked what symptom was

responsible for stopping exercise.

Constant Load Exercise, Lung Volumes and Respiratory Mechanics

On arrival at the laboratory, a balloon tipped catheter (Ackrad Laboratories Inc., Cranford, NJ)

was inserted for the measurement of esophageal pressure. After applying a topical anaesthetic to

the patient’s nares and nasal conchae (Xilocaine, Lidocaine Hydrochloride), the catheter was

advanced into the stomach and then withdrawn approximately 10 cm and positioned in the lower

third of the esophagus. Subjects were asked to perform a brief Valsalva maneuver while the

catheter was open to the atmosphere to empty the balloon and then the balloon was inflated to

1.0 ml as per manufacturer’s recommendation. The validity of the balloon position was assessed

using the occlusion test of Baydur et al., (E7) and then secured in place.

Subjects were seated on the cycle ergometer and after breathing the test gas for an

adequate wash-in period spirometry was performed to obtain resting pulmonary function on each

gas mixture. Resting ventilatory parameters and inspiratory capacity were also measured to

obtain baseline values. Constant-load exercise was then performed at 60% maximal workload

until symptom limitation. The rationale behind using 60% of maximal workload was that we

E5

wanted patients to achieve a symptom limited maximum within approximately 4-10 minutes of

exercise with room air. Cycling cadence was self selected between 50 and 70 rpm and patients

were then asked to maintain a constant pedaling rate during each trial. Throughout exercise,

oxyhemoglobin saturation and heart rate were monitored continuously using pulse oximetry

(Biox 3700, Ohmeda Medical Boulder, CO) and a ECG (Model 43200A, Hewlett-Packard, Palo

Alto, CA), respectively. Every two minutes, blood pressure was measured by auscultatory

sphygmomanometry and exertional dyspnea and leg discomfort were evaluated using the Borg

Scale (E6). The Borg scale was explained by anchoring the ends of the scale to descriptors such

that “0” represented no dyspnea or leg discomfort at all while “10” was indicative of the

maximal dyspnea or leg fatigue that had ever been experienced by the patient.

Two research assistants, blinded to the gas mixture used, encouraged the patients to

exercise for as long as possible in a consistent fashion. Following exercise termination, subjects

rested for 60-90 minutes before the second exercise trial was performed with a different gas.

Previous work by O’Donnell et al., (E8, E9) have demonstrated this to be adequate rest to allow

for a repeatable effort in this population. The final two constant-load trials were then completed

at a similar time on a different day. Test days were separated by at least 48 hours and were

generally repeated within a one-week period.

Ventilatory parameters [VE, inspiratory and expiratory tidal volume, breathing frequency,

inspiratory time (TI) and TE] were measured every two minutes by switching the patient from the

reservoir bag to a bag-in-box in series with a low resistance spirometer (Model 1022,

SensorMedics, Yorba Linda, CA). To confirm the spirometer calibration before each pulmonary

function test and exercise trial, the bag-in-box system was evacuated using a vacuum pump and

then filled with the experimental gas mixture. The system was flushed with the test gas and the

E6

spirometer calibration was checked using a 3L syringe. The duration of data collection varied

during each test depending on the ventilation rate as there was only a limited volume of gas in

the bag-in-box system. At all but the highest ventilation rates we were able to collect 60s worth

of data at every time point. Patients were asked to indicate if they were within approximately 30s

of symptom limitation if they were between IC measurements. This allowed adequate time for

patients to be switched onto the bag-in-box system to collect IC data at symptom limitation.

Once the patient was switched back to the reservoir bag to breathe, expired gases were passed

through the metabolic measurement system (MMC). The gas analyzers of the MMC were

calibrated before each test and verified immediately after using primary standard gases

containing appropriate concentrations of O2 and CO2 in either N2 or He. The pneumotachometer

(Hans Rudolph, Kansas City, MO) was also calibrated with the appropriate inspired gas. The

coefficient of variation for VE measured with the bag-in-box compared to the MMC was <1% for

all gases. A schematic of the experimental setup is depicted in Figure E1.

For measurement of esophageal pressure the balloon catheter was connected to a

differential pressure transducer (MP45, Validyne, Northridge, CA) that was calibrated before

each test using a water filled manometer. Signals from the spirometer and pressure transducers

were converted to a digital signal using a data acquisition system (Powerlab ML785, ADI

Instruments, Colorado Springs, CO). The volume signal from the spirometer was differentiated

to obtain flow. All data was sampled at 100Hz and stored on a computer for analysis at a later

date.

Measurement of Lung Volumes: Assuming that total lung capacity (TLC) does not change

with exercise (E10, E11), repetitive inspiratory capacity maneuvers (IC) were performed to track

changes in EELV (TLC - IC). This technique has previously been shown to be reliable for

E7

measuring EELV during exercise in this population (E11, E12). Subjects performed a minimum

of three IC maneuvers at rest and every two minutes during each of the constant-load trials. To

ensure IC maneuvers were performed accurately during exercise tidal breathing was

continuously displayed on a computer monitor while the subject breathed from the bag-in-box.

At the end of a normal expiration the patient was asked to breathe-in without warning and to give

an additional effort on top of a maximal inspiration (E11). To practice this technique and to

make sure patients could reproduce satisfactory IC maneuvers, a practice session was included

before the first constant load trial. Additionally, to guarantee that the IC maneuvers were

performed to TLC during exercise, the esophageal pressure achieved at the peak inspired volume

plateau (zero flow) was compared during exercise to resting values (E11).

Measurement of Ventilatory Mechanics: The work of breathing was estimated using

Campbell diagrams and the technique of Yan et al., (E13), which allows the inspiratory elastic

work of breathing to be separated into the work to overcome PEEPi (WIP) and the work required

to overcome the elastic recoil of the lung or the non-PEEPi inspiratory elastic load (WINP) (E13,

E14). Assuming a normal chest wall compliance in COPD (E15, E16), the static chest-wall

compliance curve was obtained from the literature taking the age and sex of our subjects into

consideration (E17). Positioning of the curve was achieved using the esophageal pressure at end

expiration during quiet breathing (E13, E14). Esophageal pressure-volume loops during tidal

breathing were then superimposed onto the chest-wall compliance curve

The points of zero flow at the start and end of inspiration were joined to identify dynamic

lung compliance and inspiratory resistive work was calculated as the area inside the pressure-

volume curve and to the left of the lung compliance line. The total elastic work performed on

inspiration was then calculated as the area enclosed by the lung compliance line and the chest-

E8

wall compliance curve. In the presence of dynamic hyperinflation, PEEPi was calculated as

pressure difference between the initiation of inspiratory effort and the start of inspiratory flow.

Inspiratory elastic work was then separated into the two aforementioned components by drawing

a line parallel to the chest-wall compliance curve passing through the point on the dynamic

pressure-volume loop where inspiratory flow started (E14). As a result, the area left of this line

represents WINP and the remaining elastic work is considered WIP. Additional work performed

by the respiratory muscles during expiration was then calculated as the portion of the pressure-

volume loop positioned to the right of the chest-wall compliance curve. This process was

performed on three esophageal pressure-volume loops at rest, symptom limitation and two

isotimes. The plots were imported into a software package to accurately determine areas (Image-

Pro Plus, Media Cybernetics, Silver Spring, MD). These results were averaged and reported both

as the work performed per breath and as a minute average by taking breathing frequency into

consideration.

Data Analysis

A one-way repeated measures analysis of variance (ANOVA) was performed at symptom

limitation, and at two isotimes during the exercise tolerance trials using commercially available

software (Statistica, Statsoft, Oklahoma city, OK). Isotime 1 and 2 were defined as symptom

limitation in the air and HOX trials, respectively. When a patient went longer on air than on an

experimental gas (n=1) or shorter on HE-HOX than HOX (n=2) the end exercise responses were

carried forward. When the ANOVA detected a significant effect, a Tukey post hoc multiple

comparisons test was performed.

E9

To ascertain any association between the improvements in exercise duration and changes

in EELV, VE, work of breathing, and/or dyspnea, simple regression analysis using Pearson

correlations were performed. In addition, the strongest significant contributors to the

improvement in exercise time were selected by multiple stepwise regression analysis. For all

analyses and post hoc comparisons the alpha level was set a priori at 0.05.

E10

E1. 1995. Standardization of Spirometry, 1994 Update. American Thoracic Society. Am J

Respir Crit Care Med 152(3):1107-36.

E2. Crapo, R. O., A. H. Morris, and R. M. Gardner. 1981. Reference spirometric values using

techniques and equipment that meet ATS recommendations. Am Rev Respir Dis 123(6):659-64.

E3. Miller, A., J. C. Thornton, R. Warshaw, H. Anderson, A. S. Teirstein, and I. J. Selikoff.

1983. Single breath diffusing capacity in a representative sample of the population of Michigan,

a large industrial state. Predicted values, lower limits of normal, and frequencies of abnormality

by smoking history. Am Rev Respir Dis 127(3):270-7.

E4. Goldman, H. I., and M. R. Becklake. 1959. Respiratory function tests; normal values at

median altitudes and the prediction of normal results. Am Rev Tuberc 79(4):457-67.

E5. O'Donnell, D. E., S. M. Revill, and K. A. Webb. 2001. Dynamic hyperinflation and

exercise intolerance in chronic obstructive pulmonary disease. Am J Respir Crit Care Med

164(5):770-7.

E6. Borg, G. A. 1982. Psychophysical bases of perceived exertion. Med Sci Sports Exerc

14(5):377-81.

E7. Baydur, A., P. K. Behrakis, W. A. Zin, M. Jaeger, and J. Milic-Emili. 1982. A simple

method for assessing the validity of the esophageal balloon technique. Am Rev Respir Dis

126(5):788-91.

E8. O'Donnell, D. E., D. J. Bain, and K. A. Webb. 1997. Factors contributing to relief of

exertional breathlessness during hyperoxia in chronic airflow limitation. Am J Respir Crit Care

Med 155(2):530-5.

E11

E9. O'Donnell, D. E., C. D'Arsigny, and K. A. Webb. 2001. Effects of hyperoxia on ventilatory

limitation during exercise in advanced chronic obstructive pulmonary disease. Am J Respir Crit

Care Med 163(4):892-8.

E10. Stubbing, D. G., L. D. Pengelly, J. L. Morse, and N. L. Jones. 1980. Pulmonary mechanics

during exercise in subjects with chronic airflow obstruction. J Appl Physiol 49(3):511-5.

E11. Yan, S., D. Kaminski, and P. Sliwinski. 1997. Reliability of inspiratory capacity for

estimating end-expiratory lung volume changes during exercise in patients with chronic

obstructive pulmonary disease. Am J Respir Crit Care Med 156(1):55-9.

E12. Dolmage, T. E., and R. S. Goldstein. 2002. Repeatability of inspiratory capacity during

incremental exercise in patients with severe COPD. Chest 121(3):708-14.

E13. Yan, S., D. Kaminski, and P. Sliwinski. 1997. Inspiratory muscle mechanics of patients

with chronic obstructive pulmonary disease during incremental exercise. Am J Respir Crit Care

Med 156(3 Pt 1):807-13.

E14. Sliwinski, P., D. Kaminski, J. Zielinski, and S. Yan. 1998. Partitioning of the elastic work

of inspiration in patients with COPD during exercise. Eur Respir J 11(2):416-21.

E15. Sharp, J. T., P. van Lith, C. v. Nuchprayoon, R. Briney, and F. N. Johnson. 1968. The

thorax in chronic obstructive lung disease. The American Journal of Medicine 44(1):39-46.

E16. Fleury, B., D. Murciano, C. Talamo, M. Aubier, R. Pariente, and J. Milic-Emili. 1985.

Work of breathing in patients with chronic obstructive pulmonary disease in acute respiratory

failure. Am Rev Respir Dis 131(6):822-7.

E17. Estenne, M., A. Heilporn, L. Delhez, J. C. Yernault, and A. De Troyer. 1983. Chest wall

stiffness in patients with chronic respiratory muscle weakness. Am Rev Respir Dis 128(6):1002-

7.

E12

Figure E1: Breathing circuit schematic of the bag in box and gas delivery system used during the

constant load trials. Thick black arrows designate path of airflow. MMC = metabolic

measurement system. Thin dashed lines indicate outputs from pressure transducers, CO2 analyzer

and the spirometer.

E13

Figure E1

Spirometer

Patient

Reservoir Bag

Bag-in-box Gas

Mixture

Pes

A/D Converter

MMC

Computer

CO2 analyzer 3-way valve