Heliox increases quadriceps muscle oxygen delivery during exercise in COPD patients with and without...

Heliox increases quadriceps muscle oxygen delivery during exercise in COPDpatients with and without dynamic hyperinflation

Zafeiris Louvaris,1,2 Spyros Zakynthinos,1 Andrea Aliverti,3 Helmut Habazettl,4,5 Maroula Vasilopoulou,1,2

Vasileios Andrianopoulos,1 Harrieth Wagner,6 Peter Wagner,6 and Ioannis Vogiatzis1,2

1Department of Critical Care Medicine and Pulmonary Services, Evangelismos Hospital, M. Simou and G.P. LivanosLaboratories, National and Kapodistrian University of Athens, Athens, Greece; 2Department of Physical Education and SportSciences, National and Kapodistrian University of Athens, Athens, Greece; 3Dipartimento di Biongegneria, Politecnico diMilano, Milano, Italy; 4Institute of Physiology, Charite Campus Benjamin Franklin, Berlin, Germany; 5Institute ofAnesthesiology, German Heart Institute, Berlin, Germany; and 6Department of Medicine, University of California San Diego,La Jolla, California

Submitted 18 April 2012; accepted in final form 2 August 2012

Louvaris Z, Zakynthinos S, Aliverti A, Habazettl H, Vasilopoulou M,Andrianopoulos V, Wagner H, Wagner P, Vogiatzis I. Heliox increasesquadriceps muscle oxygen delivery during exercise in COPD patients withand without dynamic hyperinflation. J Appl Physiol 113: 1012–1023, 2012.First published August 9, 2012; doi:10.1152/japplphysiol.00481.2012.—Some reports suggest that heliox breathing during exercise mayimprove peripheral muscle oxygen availability in patients withchronic obstructive pulmonary disease (COPD). Besides COPD pa-tients who dynamically hyperinflate during exercise (hyperinflators),there are patients who do not hyperinflate (non-hyperinflators). Asheliox breathing may differently affect cardiac output in hyperinfla-tors (by increasing preload and decreasing afterload of both ventri-cles) and non-hyperinflators (by increasing venous return) duringexercise, it was reasoned that heliox administration would improveperipheral muscle oxygen delivery possibly by different mechanismsin those two COPD categories. Chest wall volume and respiratorymuscle activity were determined during constant-load exercise at 75%peak capacity to exhaustion, while breathing room air or normoxicheliox in 17 COPD patients: 9 hyperinflators (forced expiratoryvolume in 1 s � 39 � 5% predicted), and 8 non-hyperinflators (forcedexpiratory volume in 1 s � 48 � 5% predicted). Quadriceps muscleblood flow was measured by near-infrared spectroscopy using indo-cyanine green dye. Hyperinflators and non-hyperinflators demon-strated comparable improvements in endurance time during heliox(231 � 23 and 257 � 28 s, respectively). At exhaustion in room air,expiratory muscle activity (expressed by peak-expiratory gastric pres-sure) was lower in hyperinflators than in non-hyperinflators. In hy-perinflators, heliox reduced end-expiratory chest wall volume anddiaphragmatic activity, and increased arterial oxygen content (by17.8 � 2.5 ml/l), whereas, in non-hyperinflators, heliox reducedpeak-expiratory gastric pressure and increased systemic vascular con-ductance (by 11.0 � 2.8 ml·min�1·mmHg�1). Quadriceps muscleblood flow and oxygen delivery significantly improved during helioxcompared with room air by a comparable magnitude (in hyperinflatorsby 6.1 � 1.3 ml·min�1·100 g�1 and 1.3 � 0.3 ml O2·min�1·100 g�1,and in non-hyperinflators by 7.2 � 1.6 ml·min�1·100 g�1 and 1.6 �0.3 ml O2·min�1·100 g�1, respectively). Despite similar increase inlocomotor muscle oxygen delivery with heliox in both groups, themechanisms of such improvements were different: 1) in hyperinfla-tors, heliox increased arterial oxygen content and quadriceps bloodflow at similar cardiac output, whereas 2) in non-hyperinflators, helioximproved central hemodynamics and increased systemic vascularconductance and quadriceps blood flow at similar arterial oxygencontent.

exercise tolerance; near-infrared spectroscopy; cardiac output; quad-riceps muscle blood flow; COPD

EXERCISE INTOLERANCE IN PATIENTS with chronic obstructive pul-monary disease (COPD) is multifactorial, including the impair-ment of ventilatory, cardiovascular, and locomotor musclesystems in a highly variable combination (6, 15, 44). Abnor-malities in pulmonary mechanics increase the work of breath-ing; this may potentially limit the available oxygen supply tolocomotor muscles (6). In addition, heart compression andintrathoracic hypovolemia consequent to exercise-induced dy-namic hyperinflation, as well as decreased venous return due toexpiratory abdominal muscle recruitment in the face of expi-ratory flow limitation, have been postulated to induce adverseeffects on central hemodynamic responses by impeding thenormal increase in cardiac output during exercise (1, 6, 37,39–41, 46, 53).

There have been several reports demonstrating that, besidesCOPD patients who dynamically hyperinflate during exercise(hyperinflators), there are also patients who do not hyperinflateduring exercise (non-hyperinflators) (2, 3, 23, 27, 54). Non-hyperinflators recruit their expiratory abdominal muscles morestrongly than hyperinflators (3, 23, 54).

To investigate the potential mechanisms of impairment inexercise capacity, strategies of unloading the respiratory mus-cles have been extensively employed (8). Along these lines,replacement of nitrogen by heliox in the inspired gas reducesthe work of breathing, as well as the degree of dynamic lunghyperinflation and the intensity of dyspnea, thereby enhancingexercise capacity (14, 21, 34, 47, 48, 58). Furthermore, studies(14, 35, 36) have shown that lightening the work of breathingby heliox or bronchodilator administration in patients exhibit-ing substantial dynamic hyperinflation accelerated the cardio-vascular response to exercise and slowed the increase in legmuscle deoxyhemoglobin concentration (considered as an in-dex of fractional oxygen extraction). These findings weresubsequently expanded by a recent study (58), which demon-strated that heliox supplementation primarily reduced expira-tory muscle activity (as indicated by reduction in peak expira-tory gastric pressure) and improved stroke volume and cardiacoutput (possibly by increasing venous return), as well assystemic, locomotor, and respiratory muscle oxygen delivery,albeit in moderately severe COPD patients not exhibitingexercise-induced dynamic hyperinflation (i.e., non-hyperinfla-

Address for reprint requests and other correspondence: I. Vogiatzis, Tho-rax Foundation, 3str Ploutarhou 10675, Athens, Greece (e-mail: [email protected]).

J Appl Physiol 113: 1012–1023, 2012.First published August 9, 2012; doi:10.1152/japplphysiol.00481.2012.

8750-7587/12 Copyright © 2012 the American Physiological Society http://www.jappl.org1012

at Universiteit M

aastricht on Decem

ber 21, 2012http://jap.physiology.org/

Dow

nloaded from

mailto:[email protected]

mailto:[email protected]

http://jap.physiology.org/

tors) with strong expiratory abdominal muscle recruitment.Whether this also occur in hyperinflators, who usually do notstrongly recruit their expiratory muscles, is unknown. Exer-cise-induced dynamic hyperinflation has deleterious effects oncentral hemodynamics (e.g., heart compression and intratho-racic hypovolemia) (37, 40). This notion is supported byJörgensen et al. (31, 32), who found that, in COPD patientswith substantial lung hyperinflation, cardiac performance isimpaired due to lower biventricular end-diastolic volumes andincreased biventricular afterload. Therefore, mitigation of ex-ercise-induced dynamic hyperinflation during heliox breathingcould improve cardiac output by different mechanisms than innon-hyperinflators (i.e., by increasing preload and decreasingafterload of both ventricles) and increase peripheral blood flowand oxygen delivery. Accordingly, we studied COPD patientswith different degrees of dynamic hyperinflation and expira-tory abdominal muscle recruitment during exercise to differ-entiate the influence of dynamic hyperinflation from that ofexpiratory loading on locomotor muscle oxygen delivery dur-ing exercise. Since heliox breathing is effective on reducingboth lung hyperinflation (14, 36, 47) and expiratory muscleactivity (58) during exercise, the primary purpose of the presentstudy was to investigate whether heliox increases locomotormuscle oxygen delivery by different mechanisms, and by thesame extent in COPD patients who exhibit dynamic hyperinfla-tion during exercise, compared with those who avoid dynamichyperinflation, but instead recruit strongly their expiratory abdom-inal muscles.

MATERIALS AND METHODS

Study subjects. Seventeen consecutive patients (3 women) withclinically stable COPD were recruited specifically for this study,according to the following inclusion criteria: 1) a postbronchodilatorforced expiratory volume in 1 s (FEV1) � 80% predicted withoutsignificant reversibility (�12% change of the initial FEV1 value or�200 ml); and 2) optimal medical therapy according to GlobalInitiative for Chronic Obstructive Lung Disease (24). Exclusion cri-teria included clinically manifest cor pulmonale, cardiovascular ill-ness, musculoskeletal abnormalities, or other diseases that couldcontribute to exercise limitation. The study was approved by AthensUniversity Hospital Ethics Committee and was conducted in accor-dance with the guidelines of the Declaration of Helsinki. Beforeparticipation in the study, all subjects were informed of any risks anddiscomforts associated with the experiments and gave written, signedinformed consent.

Experimental design. Experiments were conducted in two visits. Invisit 1, patients performed a preliminary incremental exercise test tothe limit of tolerance while breathing room air (peak work rate) toestablish the work rate corresponding to 75% of peak. In visit 2,patients underwent two constant-load exercise tests at 75% peak workrate to the limit of tolerance (i.e., exhaustion or exercise cessation),where the inspired gas mixture was initially room air and thennormoxic heliox (i.e., 79% helium and 21% oxygen). The two con-stant-load exercise protocols during visit 2 were separated by 120 minof rest. During constant-load exercise, vastus lateralis muscle bloodflow [assessed by near-infrared spectroscopy (NIRS) using the light-absorbing tracer indocyanine green (ICG) dye], as well as cardiacoutput (simultaneously assessed also by ICG dye dilution) (58), weremeasured at rest and at the final minute of the exercise protocol (i.e.,at exhaustion or exercise cessation). As endurance time was expectedto be significantly prolonged while breathing normoxic heliox (14, 47,58), measurements during these tests were also performed at the timepoint at which exercise in room air was terminated (i.e., at isotime) to

detect whether potential differences in quadriceps muscle oxygendelivery could, in part, explain differences in endurance time, whilebreathing heliox. NIRS was also used to continuously record quadri-ceps muscle tissue oxygen saturation (%StO2) (22) as an indication ofthe balance between oxygen supply and demand.

Preliminary testing. In visit 1, the incremental exercise tests wereperformed on a electromagnetically braked cycle ergometer (Ergo-line 800; Sensor Medics, Anaheim, CA) with ramp load increments of5 or 10 W/min to the limit of tolerance (the point at which the workrate could not be tolerated due to severe sensation of dyspnea and/orleg fatigue; peak exercise data are included in Table 2), with thepatients maintaining a pedaling frequency of 40–50 rpm. Tests werepreceded by a 3-min rest period, followed by 3 min of unloadedpedaling. The following pulmonary gas exchange and ventilatoryvariables were recorded breath by breath (Vmax 229; Sensor Medics):oxygen uptake, carbon dioxide elimination, minute ventilation, tidalvolume, breathing frequency, and respiratory exchange ratio. Heartrate and percent arterial oxygen saturation measured by pulse oxim-etry were determined using the R-R interval from a 12-lead onlineelectrocardiogram (Marquette Max; Marquette Hellige, Freiburg, Ger-many) and a pulse oximeter (Nonin 8600; Nonin Medical, NorthPlymouth, MN), respectively. Intensity of dyspnea and leg discomfortwere assessed using the modified Borg scale (11).

During these tests, patients performed consecutive inspiratory ma-neuvers to record the inspiratory capacity (IC) to identify thesepatients exhibiting dynamic lung hyperinflation and those who did notexhibit dynamic lung hyperinflation (42). Based on the study byO’Donnell et al. (43), indicating that, at any time point during exercisein COPD, a decrease in IC from baseline �4.5% predicted (i.e.,decrease in peak IC compared with baseline expressed as percentageof normal IC at rest) or �150 ml is necessary to indicate the incidenceof dynamic lung hyperinflation, patients exhibiting a reduction in ICfrom baseline �150 ml were classified as “hyperinflators”, whereasthose not exhibiting such reductions in IC were classified as “non-hyperinflators”. Indeed, O’Donnell et al. (43) have demonstrated thatthe 95% confidence interval of the resting IC was �0.14 l or �4.5%predicted in patients with COPD, indicating that reproducibility cri-teria of within 150 ml is appropriate for testing IC in these patients. Ofnotion, the 95% confidence interval for peak exercise IC was similar.

Subject preparation. In visit 2, subjects were prepared first withradial arterial and forearm venous catheters for blood flow measure-ments, and then with esophageal and gastric balloons for assessmentof esophageal and gastric pressures. Using local anesthesia (2%lidocaine) and sterile technique, identical catheters were introducedpercutaneously into the right radial artery and the antecubital forearmvein, both oriented in the proximal direction. The catheters were usedto collect arterial and venous blood samples and also to inject ICGinto the venous line and sample blood after each injection from thearterial line for cardiac output and muscle blood flow calculation. Thecatheters were kept patent throughout the experiment by periodicflushing with heparinized (1 U/ml) saline.

Esophageal and gastric pressures were assessed by thin-walledballoon catheters (Ackrad Laboratories, Crandford, NJ), coupled todifferential pressure transducers (MP-45, �250 cmH2O; Validyne,Northridge, CA). The balloons were inserted by nasal intubationfollowing the application of 2% lidocaine anesthetic gel into the noseand with the assistance of continuous pressure monitoring. Esopha-geal and gastric balloons were positioned in the middle third of theesophagus and stomach, respectively.

Optoelectronic plethysmography system. End-inspiratory and end-expiratory compartmental (rib cage and abdominal) chest wall vol-umes during the two constant-load exercise tests were determined byoptoelectronic plethysmography (OEP) (54, 56, 57, 58). In brief, themovement of 89 retroreflective markers placed front and back over thechest wall from clavicles to pubis was recorded. Markers were trackedby six video cameras: three in front of the subject and three behind.Dedicated software recognized in real time the markers on each

1013Regulation of Quadriceps Muscle Oxygen Delivery in COPD • Louvaris Z et al.

J Appl Physiol • doi:10.1152/japplphysiol.00481.2012 • www.jappl.org

at Universiteit M

aastricht on Decem


Dow

nloaded from


camera, reconstructed their three-dimensional coordinates by stereo-photogrammetry, and calculated volume changes. Ventilatory vari-ables (tidal volume, breathing frequency, and breathing pattern pa-rameters) reported in the paper were those recorded by OEP at rest,isotime, and exhaustion while the subjects were breathing normoxicheliox or room air.

Constant-load exercise tests. In visit 2, during exercise protocols,patients were asked to pedal for as long as possible against theconstant load to the limit of tolerance (exhaustion). Exhaustion wasdefined as the time point in which the patients signaled to stopexercising or could not maintain the required pedaling rate for 10 s,despite being encouraged by the investigators. Before imposing thetarget load on the bicycle ergometer, patients were asked to performunloaded cycling for 60 s, reaching and maintaining a cadence of �50revolutions/min. During exercise in room air, a number of IC maneu-vers were performed to confirm the magnitude of dynamic lunghyperinflation (42). Normoxic heliox breathing was achieved byhaving subjects inspire from a Douglas bag, containing normoxicheliox, that was connected to the inspiratory port of a non-rebreathingtwo-way valve by a piece of tubing. The same apparatus was utilizedduring room air breathing to ensure a blinded strategy of administra-tion of the inspired gas mixture. Recordings of pulmonary gas ex-change and ventilatory variables were performed breath by breath asduring the preliminary incremental testing. Airflow was measuredwith a hot wire pneumotachograph (Vmax 229; Sensor Medics) nearthe mouthpiece, and tidal volume changes were obtained by integrat-ing the flow signal. Before each protocol, pneumotachograph and gasanalyzers of the system (Vmax 229; Sensor Medics) were calibratedwith the experimental gas mixture. Esophageal and gastric pressuresand flow rates were displayed on a computer screen and digitized at 60Hz using an analog-to-digital converter connected to the same com-puter used for OEP (OEP system, BTS, Milan, Italy).

Esophageal and gastric pressures were averaged over 30-s breathsamples in every minute of the exercise tests. Transdiaphragmaticpressure (Pdi) was obtained by subtracting esophageal from gastricpressure. Tidal excursion in Pdi (�Pdi) was taken as peak Pdi duringinspiration minus baseline Pdi and considered as a measure of dia-phragmatic activity. Peak expiratory gastric pressure was consideredas a measure of expiratory abdominal muscle activity.

Arterial blood pressure was measured directly in all subjectsthrough the radial artery (Life Scope i BSM-2303; Nihon Kohden).Blood samples were drawn from the radial artery and antecubitalforearm vein at rest and at the final minute of the exercise protocol formeasurement of blood gases.

Cardiac output. Cardiac output was determined by the dye dilutionmethod (19), using known volumes of ICG (�1 ml at 5 mg/ml)injected into the right antecubital forearm vein followed by a rapid10-ml flush of isotonic saline. Blood was withdrawn from the radialartery using an automated pump (Harvard Apparatus) at 20 ml/minthrough a linear photodensitometer (Pulsion ICG; ViCare Medical)connected to a cardiac output computer (Waters CO-10; Waters,Rochester, MN) through a closed-loop, sterile tubing system, aspreviously described (55, 56, 57).

Quadriceps muscle blood flow and oxygenation by NIRS. Tomeasure quadriceps muscle blood flow, one set of NIRS optodes wasplaced over the left vastus lateralis muscle 10–12 cm above the knee,secured using double-sided adhesive tape. The optode separation dis-tance was 4 cm, corresponding to a penetration depth of �2 cm. Optodeswere connected to a NIRO 200 spectrophotometer (Hamamatsu Photo-nics KK, Hamamatsu, Japan), which was used to measure ICG concen-tration following the same 5-mg bolus injection of ICG in the antecubitalforearm vein used for cardiac output assessment (13). Tissue microcir-culatory ICG was detected transcutaneously by measuring light atten-uation with NIRS at 775-, 813-, and 850-nm wavelengths and ana-lyzed using an algorithm incorporating the modified Beer-LambertLaw (13, 20, 33, 52), as previously described (26, 56, 57). Blood flowwas calculated from the rate of tissue ICG accumulation over time,

measured by NIRS according to the Sapirstein principle (49), asdescribed previously (13, 28).

Quadriceps muscle oxygenation was assessed continuouslythroughout the whole testing period by the same NIRO 200 spectro-photometer as used for the measurement quadriceps muscle bloodflow. High ICG tissue concentrations during the passage of the dyebolus through the quadriceps muscle may interfere with hemoglobin(Hb) results. Therefore, to avoid any interference between ICG andHb wavelengths, tissue oxygenation data were averaged over 10 simmediately before the appearance of ICG. The ratio of oxygenatedHb to total Hb was used as an index of changes in quadriceps muscleStO2, which reflects the balance between oxygen supply and demand(16, 17, 18, 25, 33).

Systemic and quadriceps muscle oxygen delivery were calculatedas the product of the cardiac output and the quadriceps muscle bloodflow, respectively, and arterial oxygen content (CaO2). Systemic vas-cular conductance was calculated by dividing the cardiac output bythe mean arterial blood pressure.

Blood analysis and calculations. Radial arterial tensions of O2

(PaO2) and CO2 (PaCO2), Hb concentration, and arterial oxygen satu-ration (%SaO2) were measured from 2-ml blood samples using ablood-gas analyzer combined with a co-oximeter (ABL 625; Radiom-eter, Copenhagen, Denmark) within a few seconds of collection. CaO2

was computed [CaO2 � (1.34 � Hb � SaO2) (0.003 � PaO2)]. Theblood gas analyzer was auto-calibrated every 4 h throughout the dayand before each set of measurements. Blood-gas measurements werecorrected for patient’s tympanic temperature taken during withdrawalof each arterial blood-gas sample.

Quadriceps muscle strength and 6-min walking test. Quadricepsmuscle strength was assessed using the maximal isometric voluntarycontraction technique of the knee extensors following a standardizedprotocol (51). The 6-min walking test was performed according toAmerican Thoracic Society guidelines (9).

Statistical analysis. Data are reported as means � SE. The mini-mum sample size was calculated on the basis of 80% power and atwo-sided 0.05 significance level. The sample size capable of detect-ing between-condition (i.e., air, heliox) difference of 30% was esti-mated for the change in quadriceps muscle oxygen delivery using thestandard deviations from our laboratory’s previous study (58). Thecritical sample size was estimated to be eight patients per group.During exercise in air, a paired t-test was employed within each groupto test whether the magnitude of change from rest to exhaustion wassignificant. During exercise in heliox, one-way ANOVA with re-peated measures was used to identify statistically significant differ-ences across the three time points (i.e., rest, isotime, exhaustion).When ANOVA detected statistical significance, pairwise differenceswere identified using Tukey’s honestly significant difference post hocprocedure. Paired or independent t-tests were employed to compareresponses at comparable exercise time point (i.e., isotime; exhaustion)between the two conditions within each group or between the twogroups, respectively. The level of significance for all analyses was setat P � 0.05.

RESULTS

Subject characteristics and exercise capacity. Patients inboth groups exhibited moderate to severe airflow obstructionwith increased static lung volumes, substantial reductions incarbon monoxide diffusion capacity, and mildly reduced arte-rial oxygen tension (Table 1). Hyperinflators were more se-verely impaired in terms of baseline lung function character-istics. Accordingly, for the group of hyperinflators, one patientwas Global Initiative for COPD (GOLD) stage II, six patientswere GOLD stage III, and the remaining two were GOLDstage IV, whereas for the group of non-hyperinflators, fivepatients were GOLD stage II, two patients were GOLD stage

1014 Regulation of Quadriceps Muscle Oxygen Delivery in COPD • Louvaris Z et al.


at Universiteit M

aastricht on Decem


Dow

nloaded from


III, and one was GOLD stage IV. Hyperinflators also displayedgreater airway obstruction, greater static lung volumes, andthus greater resting lung hyperinflation and lower diffusingcapacity compared with non-hyperinflators (Table 1). Hyper-inflators and non-hyperinflators had comparable body compo-sition characteristics (Table 1), as well as functional capacityassessed by the 6-min walking test and quadriceps musclestrength (Table 2). At the limit of tolerance during the prelim-inary incremental exercise test, the decrease from baseline inIC was 0.45 � 0.07 liter in hyperinflators, but only 0.11 � 0.01liter in non-hyperinflators (Table 2). The magnitude of dy-namic lung hyperinflation was confirmed by the data obtainedduring the constant-load test in room air, as, at the limit oftolerance, the decrease from baseline in IC was 0.44 � 0.05liter in hyperinflators and 0.10 � 0.02 liter in non-hyperinfla-tors. Moreover, hyperinflators demonstrated worse peak exer-cise tolerance than non-hyperinflators in the incremental exer-cise test (Table 2). Interestingly, both hyperinflators and non-hyperinflators demonstrated comparable improvement inendurance time during heliox compared with room air breath-ing during constant-load exercise (by 231 � 23 and 257 � 28s, corresponding to 62.9 � 4.1 and 61.3 � 5.0%, respectively).

Respiratory muscle kinematics and loading. In hyperinfla-tors, the increase in end-expiratory chest wall volume from restto exhaustion was 0.47 � 0.12 liter (Fig. 1C). Notably, duringheliox breathing compared with room air at isotime and ex-

haustion, there was a significant decrease in end-expiratorychest wall volume (by 0.44 � 0.09 and 0.41 � 0.11 liter,respectively; P � 0.001 for both) (Fig. 1C), which was attrib-uted to the significant decrease in both end-expiratory rib cagevolume (by 0.22 � 0.01 and 0.19 � 0.11 liter, respectively;P � 0.05 for both; Fig. 1A) and end-expiratory abdominalvolume (by 0.21 � 0.08 and 0.22 � 0.10 liter, respectively;P � 0.05 for both; Fig. 1B). In these patients, peak expiratorygastric pressure did not change from rest during exercise inroom air; however, following heliox administration, there wasa significant increase in peak expiratory gastric pressure at bothisotime and exhaustion (or exercise cessation) from rest (P �0.05) (Fig. 2B). This increase in peak expiratory gastric pres-sure during heliox breathing is in line with the significantdecrease in end-expiratory abdominal volume (Fig. 2B). Hy-perinflators also demonstrated a significant decrease in dyspneasensations during exercise, at both isotime and exhaustion (orexercise cessation), while breathing heliox compared withroom air (P � 0.05) (Table 3).

In non-hyperinflators, end-expiratory chest wall volumefrom rest to exhaustion in room air decreased by 0.33 � 0.15liter (P � 0.001 vs. hyperinflators; Fig. 1F). Heliox breathingcompared with room air significantly reduced the degree ofend-expiratory rib cage hyperinflation at isotime and exhaus-tion (P � 0.05; Fig. 1D), but simultaneously caused a signif-icant increase in end-expiratory abdominal volume (P � 0.05;Fig. 1E), reflecting lower expiratory abdominal muscle recruit-ment. In fact, peak expiratory gastric pressure at both isotimeand exhaustion was lower during heliox compared with roomair breathing (P � 0.001; Fig. 2D). Due to different directionsof change in end-expiratory rib cage and abdominal volumesfollowing heliox breathing (Fig. 1, D and E), end-expiratorytotal chest wall at isotime and exhaustion was not differentbetween heliox and air breathing (Fig. 1F). This response ofnon-hyperinflators to heliox breathing was different from thatof hyperinflators who reduced end-expiratory chest wall vol-ume compared with air breathing (by 0.440 � 0.090 vs.�0.023 � 0.011 ml in non-hyperinflators at isotime; P �

Table 2. Functional capacity and peak exercise data

Hyperinflators Non-hyperinflators

6-min walking test, m 382.0 � 14.0 404.0 � 27.0Quadriceps muscle strength, kg 29.6 � 2.8 31.2 � 4.0WRpeak, W 54.0 � 9.0 72.0 � 10.0*V̇O2peak, ml �kg�1 �min�1 14.9 � 1.0 17.2 � 1.4*HRpeak, beats/min 115 � 4 122 � 6*V̇Epeak, l/min 37.0 � 3.0 53.0 � 7.0*VTpeak, liters 1.18 � 0.11 1.62 � 0.22*fpeak, breaths/min 32.0 � 3.0 33.0 � 3.0�IC, liters 0.45 � 0.07 0.11 � 0.01*�IC, %predicted 15.2 � 5.3 3.1 � 0.8*SpO2, % 89 � 2 90 � 1Borg dyspnea scores 7.2 � 0.8 6.3 � 0.9*Borg leg effort scores 5.2 � 0.7 5.4 � 0.6

Values are means � SE for 9 and 8 subjects in hyperinflators and nonhy-perinflators, respectively. WRpeak, peak work rate; V̇O2peak, peak oxygenuptake; HRpeak, peak heart rate; V̇Epeak, peak minute ventilation; VTpeak, peaktidal volume; fpeak, peak breathing frequency; �IC, decrease in peak inspira-tory capacity compared with baseline; �IC (% predicted), decrease in peakinspiratory capacity compared with baseline expressed as percentage of normalinspiratory capacity at rest; SpO2, arterial oxygen saturation measured bypulse oximetry. *Significant difference compared with hyperinflators (P �0.05).

Table 1. Demographic and lung function characteristics


Demographic/Anthropometric

n 9 8Age,yr 64.0 � 3.0 62.0 � 3.0Height,cm 173.0 � 2.0 168.0 � 4.0Weight,kg 76.0 � 7.0 78.0 � 6.0Body mass index, kg/m2 25.2 � 2.0 27.4 � 1.0Fat free mass index, kg/m2 18.1 � 0.9 19.4 � 1.2

Pulmonary function

FEV1, liters 1.19 � 0.21 1.41 � 0.21*FEV1, %predicted 39.1 � 5.3 48.4 � 5.4*FVC, liters 2.88 � 0.22 2.87 � 0.44FVC, %predicted 79.8 � 6.8 77.1 � 5.8FEV1/FVC, %predicted 49.4 � 4.3 62.4 � 6.6*IC, liters 2.24 � 0.15 2.66 � 0.22*IC, %predicted 76.1 � 6.5 84.3 � 7.5*RV, liters 4.74 � 0.57 4.01 � 0.48*RV, %predicted 215.0 � 22.0 182.0 � 19.0*FRC, liters 5.74 � 0.46 5.11 � 0.56*FRC, %predicted 182.0 � 11.0 164.0 � 9.0*TLC, liters 8.00 � 0.43 7.81 � 0.56TLC, %predicted 124.0 � 9.0 120.0 � 5.0IC/TLC, %predicted 59.0 � 4.1 72.3 � 5.9*RV/TLC, %predicted 172.0 � 10.0 153.0 � 9.0*TLCO, %predicted 37.8 � 5.3 53.1 � 8.2*Resting room air PaO2, Torr 81.8 � 9.3 86.5 � 7.9Resting room air PaCO2, Torr 43.3 � 3.4 41.6 � 1.9Resting room air SaO2, % 95.0 � 1.5 96.2 � 1.9

Values are means � SE; n, no. of subjects. FEV1, forced expiratory volumein 1 s; FVC, forced vital capacity; TLC, total lung capacity; IC, inspiratorycapacity; RV, residual volume; FRC, functional residual capacity; TLCO,diffusing capacity of the lung for carbon monoxide; PaO2, partial pressure ofarterial oxygen; PaCO2, partial pressure of arterial carbon dioxide; SaO2, arterialoxygen saturation. *Significant difference compared with hyperinflators (P �0.05).



at Universiteit M

aastricht on Decem


Dow

nloaded from


0.001). The response of expiratory muscle activity in non-hyperinflators to heliox breathing was also different from thatin hyperinflators, as non-hyperinflators reduced peak expira-tory gastric pressure compared with air breathing (by 8.2 � 3.1vs. �2.5 � 2.3 cmH2O in hyperinflators at isotime; P �0.001).

Central hemodynamic responses. During air breathing, themagnitude of increase in heart rate, stroke volume, and cardiacoutput from rest to exhaustion was not significantly differentbetween groups. At isotime during exercise while breathingheliox, stroke volume (Fig. 3E) was significantly greater (P �0.02) compared with room air breathing in non-hyperinflators.In these patients, there was also a tendency (P � 0.08) forcardiac output to be greater with heliox breathing (Fig. 3D). Incontrast, hyperinflators did not exhibit any improvement instroke volume or cardiac output while breathing heliox com-pared with room air. In hyperinflators, CaO2

was greater (P �0.03) at isotime during exercise while breathing heliox com-pared with room air breathing (Fig. 4A). This led to a signifi-cant increase in systemic oxygen delivery with heliox atisotime compared with room air breathing (Fig. 4B), despite

the absence of a significant increase in cardiac output. Innon-hyperinflators, heliox breathing did not increase CaO2

(Fig.4D). However, in this group, systemic vascular conductanceand systemic oxygen delivery were greater at isotime (P �0.03 and P � 0.02, respectively) while breathing heliox com-pared with room air (Fig. 4, F and E, respectively). Accord-ingly, systemic oxygen delivery significantly improved duringheliox compared with air breathing at isotime by a comparablemagnitude in both hyperinflators and non-hyperinflators as theimprovement in systemic oxygen delivery did not differ be-tween the two groups (0.26 � 0.05 and 0.29 � 0.06 l/min,respectively) (Fig. 4, B and E, respectively).

Quadriceps muscle hemodynamic and oxygenation responses.Vastus lateralis muscle blood flow and oxygen delivery, aswell as vastus lateralis muscle StO2

, were significantly greaterat isotime during exercise while breathing normoxic helioxcompared with room air in both hyperinflators and non-hyper-inflators (P � 0.05–0.001; Fig. 5). Moreover, quadricepsmuscle blood flow, oxygen delivery, and StO2

at isotime im-proved during heliox breathing compared with room air bya comparable magnitude (in hyperinflators by 6.1 � 1.3

Rest Isotime Exhaustion

-0.3

0.0

0.3

0.6

0.9

1.2

-0.3

0.0

0.3

0.6

0.9

1.2

End- inspiration

End -expiration

Tota

l Che

st w

all v

olum

e(L

)

CTo

tal C

hest

wal

l vol

ume

(L)

F

End- inspiration

End- expiration

-0.5

0.0

0.5

1.0

1.5

2.0

-0.5

0.0

0.5

1.0

1.5

2.0

Rest Isotime

*†

*

*

*

*

*

*

*

*

End- inspiration

End- expirationAbdo

min

al v

olum

e (L

)

End- inspiration

End- expiration

B

Abd

omin

al v

olum

e (L

)

E

-0.6

-0.3

0.0

0.3

0.6

0.9

-0.6

-0.3

0.0

0.3

0.6

0.9

*† ‡

*‡

**

*†

*

‡

*†*‡

Exhaustion

Hyperinflators Non-hyperinflatorsΑ

End- inspiration

End- expiration

Rib

cag

e vo

lum

e (L

)

End- inspiration

End- expiration

D

Rib

cag

e vo

lum

e (L

)

*†

*

*†*‡

*‡

‡

*

*†

* *

*

Fig. 1. Chest wall volume regulation. Rib cage volume (A and D), abdominal volume (B and E), and total chest wall volume (C and F) regulation at the endof inspiration and expiration recorded at rest, at the time of exhaustion on room air (isotime), and at the limit of tolerance on normoxic heliox (exhaustion) whilebreathing normoxic heliox (o) or room air (Œ) are shown. Values are means � SE for 9 and 8 subjects in hyperinflators (A–C) and non-hyperinflators (D–F),respectively. Isotime data are those obtained on normoxic heliox at the same time as at exhaustion on room air. *Significant differences (P � 0.05) from rest.†Significant differences (P � 0.05) between air and heliox breathing at the same time point of exercise. ‡Significant differences (P � 0.05) between heliox andair breathing at exhaustion.



at Universiteit M

aastricht on Decem


Dow

nloaded from


ml·min�1·100 g�1, 1.3 � 0.3 ml O2·min�1·100 g�1, and 3.4 �0.8%, and in non-hyperinflators by 7.2 � 1.6 ml·min�1·100g�1, 1.6 � 0.3 ml O2·min�1·100 g�1, and 3.6 � 0.9%, respec-tively).

DISCUSSSION

Some COPD patients given heliox to breathe during exerciserespond by reduction in dynamic hyperinflation, while othersby reducing expiratory muscle activity. This study investigatedwhether, and if so how, these two strategies are equallyeffective in enhancing locomotor muscle oxygen availability

by comparing patients who exhibit dynamic hyperinflationduring exercise and patients who avoid dynamic hyperinflationbut strongly contract their expiratory abdominal muscles. Themain findings were the following. 1) Alleviation of abnormalpulmonary mechanics by means of heliox breathing increasedexercise tolerance and locomotor muscle blood flow and oxy-gen delivery by a comparable magnitude in patients with orwithout dynamic hyperinflation. 2) In hyperinflators, helioxreduced end-expiratory chest wall volume and diaphragmaticactivity, whereas, in non-hyperinflators, heliox reduced peakexpiratory gastric pressure. 3) Despite similar increase in

∆Pd

i (c

m H

20)

∆Pd

i (c

m H

20)

6

12

18

24

30

36

Pea

k ex

pira

tory

gas

tric

pres

sure

(cm

H20

)

Pea

k ex

pira

tory

gas

tric

pres

sure

(cm

H20

)Hyperinflators Non-hyperinflators

Rest IsotimeRest Isotime

A

B

C

D

*‡

*†

*

6

12

18

24

30

36

* *

Exhaustion Exhaustion

5

10

15

20

25

30

5

10

15

20

25

30*†

*

*‡

*

* *

Fig. 2. Respiratory muscle load. Tidal excursion of transdiaphragmatic pressure (�Pdi; A and C) and peak expiratory gastric pressure (B and D) recorded at rest,at the time of exhaustion on room air (isotime), and at the limit of tolerance on normoxic heliox (exhaustion) while breathing normoxic heliox (o) or room air(Œ) are shown. Values are means � SE for 9 and 8 subjects in hyperinflators (A and B) and non-hyperinflators (C and D), respectively. Isotime data are thoseobtained on normoxic heliox at the same time as at exhaustion on room air. *Significant differences (P � 0.05) from rest. †Significant differences (P � 0.05)between air and heliox breathing at the same time point of exercise. ‡Significant differences (P � 0.05) between heliox and air breathing at exhaustion.

Table 3. Exercise and ventilatory responses during constant-load exercise in air and heliox


Heliox Heliox

Air at exercise cessation Isotime At exercise cessation Air at exercise cessation Isotime At exercise cessation

Endurance time, s 367 � 32 598 � 56† 419 � 40* 676 � 75*†V̇E, l/min 33.1 � 4.1 37.6 � 4.8† 37.3 � 4.8† 46.5 � 6.7* 48.4 � 6.1* 49.6 � 6.3*VT, liters 1.15 � 0.10 1.25 � 0.15† 1.17 � 0.13 1.54 � 0.19* 1.57 � 0.11* 1.51 � 0.12*f, breaths/min 30.0 � 2.0 30.0 � 2.0 32.0 � 2.0 30.0 � 2.0 31.0 � 2.0 33.0 � 2.0TI, s 0.80 � 0.06 0.81 � 0.07 0.73 � 0.07 0.83 � 0.06 0.81 � 0.07 0.74 � 0.06TE, s 1.20 � 0.23 1.22 � 0.16 1.12 � 0.16 1.21 � 0.08 1.18 � 0.09 1.11 � 0.09TI/TT, % 39.1 � 1.6 39.9 � 1.3 39.4 � 1.3 40.6 � 1.7 41.3 � 1.2 40.7 � 1.2SaO2, % 90.2 � 1.5 91.3 � 1.2 91.0 � 1.2 91.7 � 1.3 92.4 � 1.2 92.1 � 1.2Borg dyspnea scores 7.1 � 1.3 5.1 � 0.9† 5.7 � 0.9† 5.3 � 0.8* 5.5 � 0.7 5.9 � 0.7Borg leg effort scores 5.8 � 1.1 5.4 � 1.0 6.3 � 1.0 6.1 � 0.9 5.9 � 1.2 5.8 � 1.2

Values are means � SE for 9 and 8 subjects in hyperinflators and nonhyperinflators, respectively. V̇E, minute ventilation; VT, tidal volume; f, breathingfrequency; TI, time of inspiration; TE, time of expiration; TI/TT, duty cycle of inspiration. *Significant difference compared with hyperinflators (P � 0.05) atthe same time-point. †Significant difference compared with air breathing (P � 0.05) in the same group.



at Universiteit M

aastricht on Decem


Dow

nloaded from


locomotor muscle oxygen delivery with heliox in both groups,the mechanisms of such improvements were different: in hy-perinflators, heliox increased CaO2

and quadriceps blood flowat similar cardiac output, whereas, in non-hyperinflators, helioximproved central hemodynamics and increased systemic vas-cular conductance and quadriceps blood flow at similar CaO2

.Central hemodynamic responses. At isotime during heliox

compared with air breathing, only non-hyperinflators increasedstroke volume (Fig. 3E), thereby allowing an �12% increase incardiac output (Fig. 3D), albeit not reaching the level ofstatistical significance (P � 0.08). Nevertheless, at isotimeduring exercise on heliox, systemic vascular conductance wasgreater in non-hyperinflators (Fig. 4F). The increase in strokevolume following heliox administration in non-hyperinflatorsmight be due to the substantial reduction of expiratory abdom-inal muscle activity reflected by the decrease in peak expiratorygastric pressure (by 8.2 � 3.1 cmH2O) (Fig. 2D). Indeed,several studies have shown that the development of highpositive expiratory pressures decrease venous return and car-

diac output, as high abdominal pressures compress the inferiorvena cava and interfere with venous return from the exercisingleg muscles to the right ventricle, thereby impeding the normalincrease in cardiac output (37, 39, 40, 46). Moreover, the highalveolar pressures during excessive expiratory muscle recruit-ment during expiration would compress pulmonary blood ves-sels and thus decrease left ventricular preload (6). Previousstudies in healthy humans have documented that alleviation ofartificially imposed expiratory muscle loading allowed a sig-nificant increase in cardiac output during submaximal exercise(4, 50).

Patients experiencing dynamic hyperinflation during exer-cise on room air demonstrated a substantial reduction in end-expiratory chest wall volume (Fig. 1C) following heliox ad-ministration. Dynamic hyperinflation per se has known nega-tive cardiovascular effects, including heart compression thatimpairs diastolic function (the so-called pneumonic cardiactamponade) and intrathoracic hypovolemia (1, 31, 32, 39, 41,53), that may impede the normal increase in cardiac output

Hea

rt ra

te(b

eats

/min

)

Hea

rt ra

te(b

eats

/min

)C

Car

diac

Out

put

(l /m

in)

Stro

ke v

olum

e(m

l /be

at)

A

B

Non-hyperinflators

2

4

6

8

10

12

60

75

90

105

120

135

60

75

90

105

120

135

55

65

75

85

95

105

55

65

75

85

95

105

F

D

E

*

Stro

ke v

olum

e(m

l /be

at)

Car

diac

Out

put

(l /m

in)

2

4

6

8

10

12

Hyperinflators

Rest Isotime Rest Isotime

*†

**

*

* *

* *

* *

*

* *

*

*

*

*


Fig. 3. Cardiac output responses. Cardiac output (A and D), stroke volume (B and E), and heart rate (C and F) recorded at rest, at the time of exhaustion on roomair (isotime), and at the limit of tolerance on normoxic heliox (exhaustion) while breathing normoxic heliox (o) or room air (Œ) are shown. Values are means �SE for 9 and 8 subjects in hyperinflators (A–C) and non-hyperinflators (D–F), respectively. Isotime data are those obtained on normoxic heliox at the same timeas at exhaustion on room air. *Significant differences (P � 0.05) from rest. †Significant differences (P � 0.05) between air and heliox breathing at the sametime point of exercise.



at Universiteit M

aastricht on Decem


Dow

nloaded from


during exercise. Due to these deleterious effects of dynamichyperinflation on central hemodynamics, we expected thatalleviation of exercise-induced dynamic hyperinflation duringheliox breathing would improve cardiac output. However, thiswas not the case in the present study, and heliox breathing hadno effect on stroke volume or cardiac output in hyperinflators(Fig. 3, A and B), probably because the effects of dynamichyperinflation per se on cardiac performance are not so prom-inent (50). Indeed, our findings are compatible with those byOelberg et al. (45) and Chiappa et al. (14), reporting no effectson hemodynamic responses following heliox administration inpatients exhibiting similar degrees of dynamic lung hyperin-flation on room air during constant-work rate exercise. On theother hand, hyperinflators demonstrated improved CaO2

duringheliox breathing compared with room air (Fig. 4A). Such animprovement can be attributed to alleviated ventilation-perfu-sion mismatch due to increased alveolar ventilation, better lungmechanics (34, 58), and alveolar hyperventilation consequentlyto hyperventilation induced by heliox breathing (47). In non-hyperinflators there was not any change in CaO2

(Fig. 4 D).

This difference between hyperinflators and non-hyperinflatorscould be attributed to the more beneficial effects of helioxbreathing on reducing airway resistance and improving lungmechanics and alveolar and minute ventilation in patients withmore severe airway obstruction that is in hyperinflators com-pared with non-hyperinflators (Tables 1 and 3). Another reasoncould be that hyperinflators were more hypoxemic comparedwith non-hyperinflators at room air (Tables 1 and 3), and thusthe positive effects of heliox breathing on CaO2

were more inpatients with a larger room to improve.

Quadriceps muscle hemodynamic and oxygenation responses.At isotime during exercise with heliox breathing, we observedimproved quadriceps muscle oxygen delivery in both groups(Fig. 5, B and E). In hyperinflators, this was due to theincreases in both CaO2

(Fig. 4A) and quadriceps muscle bloodflow (Fig. 5A), while for non-hyperinflators the increase inquadriceps muscle oxygen delivery was due solely to anincrease in quadriceps muscle blood flow (Fig. 5D) as CaO2

remained unaffected (Fig. 4D). The finding that unloading therespiratory muscles (by using proportional assist ventilation,

CaO

2(m

l /l)

D

160

170

180

190

200

210

CaO

2(m

l /l)

A Non-hyperinflators

Syst

emic

O2

deliv

ery

(l / m

in)

B

0.5

1.0

1.5

2.0

2.5

3.0

*†

Syst

emic

O2

deliv

ery

(l / m

in)

E

0.5

1.0

1.5

2.0

2.5

3.0

*†

Hyperinflators

40

55

70

85

100

115

Syst

emic

vas

cula

r co

nduc

tanc

e

(m

l / m

in /

mm

Hg)

C F

40

55

70

85

100

115 *†Sy

stem

ic v

ascu

lar

cond

ucta

nce

(ml /

min

/ m

mH

g)


*

*

*

*

*

*

**

*

*

140

150

160

170

180

190

*

*

††


Fig. 4. Central hemodynamic responses. Arterial oxygen content (CaO2; A and D), systemic oxygen delivery (B and E), and systemic vascular conductance (Cand F) recorded at rest, at the time of exhaustion on room air (isotime), and at the limit of tolerance on normoxic heliox (exhaustion) while breathing normoxicheliox (o) or room air (Œ) are shown. Values are means � SE for 9 and 8 subjects in hyperinflators (A–C) and non-hyperinflators (D–F), respectively. Isotimedata are those obtained on normoxic heliox at the same time as at exhaustion on room air. *Significant differences (P � 0.05) from rest. †Significant differences(P � 0.05) between air and heliox breathing at the same time point of exercise.



at Universiteit M

aastricht on Decem


Dow

nloaded from


heliox, or bronchodilators) enhances locomotor muscle oxygenavailability has been illustrated or postulated in several previ-ous studies by showing that reductions in the work of breathingare associated with an increase in locomotor muscle oxygen-ation and performance in patients with COPD (7, 12, 14). Sucheffects have been ascribed to redistribution of blood flow fromthe respiratory to locomotor muscles (7, 14, 29). In the case ofhyperinflators, the increase in quadriceps muscle blood flowmight be due to blood flow redistribution from inspiratory (i.e.,diaphragm, intercostals, and accessory inspiratory rib cagemuscles) to locomotor muscles, as heliox breathing did notimprove cardiac output. Accordingly, it is possible that themechanism of redistribution can be mainly ascribed to lowerblood flow requirements of the inspiratory muscles, asduring heliox breathing we observed reduced diaphragmaticactivity (Fig. 2A) and decreased end-inspiratory rib cagevolume (Fig. 1A), which may signify reduced activity ofintercostals and accessory inspiratory rib cage muscles. Onthe contrary, in non-hyperinflators, the increase in quadri-

ceps muscle blood flow can partly be attributed to theincrease in systemic vascular conductance (Fig. 4F). Inaddition, reduced activity of expiratory abdominal musclesduring exercise with heliox breathing (Fig. 2D) may havealso contributed to the increase in locomotor muscle bloodflow, if redistribution of blood flow from the expiratorymuscles has taken place.

The improvement in peripheral muscle oxygen availabilityat isotime during heliox breathing in both groups was associ-ated with improvement in quadriceps muscle oxygen saturation(by 3.4 � 0.8 and 3.6 � 0.9% in hyperinflators and non-hyperinflators, respectively; Fig. 5, C and F). These results areconsistent with previous findings showing that unloading re-spiratory muscles with heliox or bronchodilators or usingproportional assisted ventilation increases peripheral musclesoxygenation by de-accelerating the dynamics of muscle deoxy-hemoglobin concentration (an index of oxygen extraction) andaccelerating whole body oxygen uptake kinetics (10, 12, 14).In addition, these findings support the value of measurement of

Non-hyperinflatorsHyperinflators


Qua

dric

eps

mus

cle

O2

deliv

ery

(m

l O2 /m

in /1

00g)

B

0.0

1.5

3.0

4.5

6.0

7.5

*

**†

Qua

dric

eps

mus

cle

O2

deliv

ery

(m

l O2 /m

in/1

00g)

E

0.0

1.5

3.0

4.5

6.0

7.5

*

*

*†

Qua

dric

eps

mus

cle

bloo

d flo

w(m

l /m

in/1

00g)

A

0

8

16

24

32

40 *†*

*

Qua

dric

eps

mus

cle

bloo

d flo

w(m

l /m

in/1

00g)

D

0

8

16

24

32

40 *†

*

*


C

Mus

cle

tissu

e ox

ygen

sat

urat

ion

(%)

50

52

54

56

58

60

*

†

*

F

50

52

54

56

58

60

†

*

*

Mus

cle

tissu

e ox

ygen

sat

urat

ion

(%)

Fig. 5. Quadriceps muscle oxygenation and hemodynamic responses. Quadriceps muscle blood flow (A and D), quadriceps muscle oxygen delivery (B and E),and quadriceps muscle tissue oxygen saturation (C and F) recorded at rest, at the time of exhaustion on room air (isotime), and at the limit of tolerance onnormoxic heliox (exhaustion) while breathing normoxic heliox (o) or room air (Œ) are shown. Values are means � SE for 9 and 8 subjects in hyperinflators (A–C)and non-hyperinflators (D–F), respectively. Isotime data are those obtained on normoxic heliox at the same time as at exhaustion on room air. *Significantdifferences (P � 0.05) from rest. †Significant differences (P � 0.05) between air and heliox breathing at the same time point of exercise.



at Universiteit M

aastricht on Decem


Dow

nloaded from


fractional StO2as an index for assessing local muscle oxygen-

ation responses during exercise.Technical considerations. In the present study, we assessed

changes in both lung and chest wall volumes during exercise tomeasure dynamic hyperinflation, as assessing decreases in ICby integrating flow at the mouth (spirometry) as an index ofhyperinflation measures the volume of gas entering the lungs atthe mouth, whereas OEP measures the volumes of the trunk,which includes volume changes at the mouth, but also twoother variables, gas compression and decompression in thelungs and blood shifts between the trunk and extremities (38).Therefore, there is a discrepancy between the two methodsused to measure dynamic hyperinflation, and this discrepancyis particularly important in patients who forcefully recruitexpiratory abdominal muscles during expiration, thus exhibit-ing gas compression in the lungs and blood shifts away fromthe trunk (5, 38), as was the case in non-hyperinflators of thepresent study. Indeed, non-hyperinflators during constant-loadexercise in room air exhibited a small increase from baseline inend-expiratory lung volume (by 0.10 � 0.02 liter) when lungvolume changes were measured by spirometry, whereas, at thesame time, a significant decrease from baseline in end-expira-tory chest wall volume (by 0.33 � 0.15 liter) was measured byOEP. It has been calculated (38) that, for peak expiratorygastric pressures of �22 cmH2O at maximal exercise workload[our peak expiratory gastric pressures at exhaustion in non-hyperinflators were even higher, reaching �30 cmH2O duringair breathing (Fig. 2D)], OEP would measure an end-expiratorychest wall volume that would be 0.33 liter less than thatmeasured by spirometry. Therefore, strong expiratory abdom-inal muscle recruitment in non-hyperinflators of the presentstudy adequately explains the observed discrepancy betweenthe two methods used to measure dynamic hyperinflation. Incontrast, hyperinflators who did not increase their peak expi-ratory gastric pressure at exhaustion compared with rest duringconstant-load exercise in room air (Fig. 2B) exhibited a de-crease in IC from baseline, equivalent to 0.44 � 0.05 liter,which is in line with the increase in end-expiratory total chestwall volume (by 0.47 � 0.12 liter) recorded by OEP.

For the tracer method used, we did not correct the inputfunction for dispersion, because the time constant of dispersionfor the setting used in our experiments is not known. Iida et al.(30) measured the tracer (O15 H2O) concentration in the leftventricle to obtain the “true” input function, and, by compar-ison with the measured input function, they estimated timeconstant to be 5 s. Application of this time constant to ourmethod does not seem reasonable, because the techniques ofsampling arterial blood for measurement of the input functionare very different. Indeed, Iida et al. (30), using the sametechnique, sample arterial blood into tubes at 1- or 2-s inter-vals, while we continuously draw the blood through the mea-suring photodensitometer cuvette and measure ICG concentra-tion at a sampling rate of 20 Hz. Thus external dispersion isprobably different with the two techniques.

Furthermore, blood flow measurements of main respiratorymuscles, i.e., diaphragm or abdominal muscles, to provideevidence of redistribution of respiratory muscle perfusion,were not performed. NIRS optodes cannot be used for mea-suring blood flow to muscles that are deeply inside the humanbody (i.e., diaphragm), while placing optodes on the abdominalmuscles may result in underestimation of blood flow, owing to

potential contribution of subcutaneous tissue and skin to thelight-absorption signal.

Conclusions. The present study shows that alleviating ab-normal pulmonary mechanics by heliox breathing enhancesexercise tolerance and quadriceps muscle oxygen deliverysimilarly in COPD patients, with and without exercise-induceddynamic hyperinflation (hyperinflators and non-hyperinflators,respectively). However, the mechanisms of improvement inlocomotor muscle oxygen delivery were different between thetwo groups. In hyperinflators, heliox reduced end-expiratorychest wall volume and diaphragmatic activity and increasedCaO2

and quadriceps muscle blood flow at similar cardiacoutput (likely by redistributing blood flow from the inspiratorymuscles), whereas, in non-hyperinflators, heliox reduced expi-ratory muscle activity and improved systemic hemodynamicsand quadriceps muscle blood flow at similar CaO2

. Therefore,strategies aiming at minimizing the negative effects of cardio-pulmonary interactions on blood flow distribution and O2

transport during exercise are useful in COPD patients, regard-less of their respiratory system mechanical responses.

ACKNOWLEDGMENTS

We thank our subject for considerable time and patience during thisintensive, exhausting, and long investigation.

GRANTS

This work was supported by Thorax Foundation (directed by ProfessorCharis Roussos) and by the A. Perotti and K&V Kostantakopoulos VisitingProfessorship Funds of Thorax Foundation.

DISCLOSURES

No conflicts of interest, financial or otherwise, are declared by the author(s).

AUTHOR CONTRIBUTIONS

Author contributions: Z.L., S.G.Z., A.A., H.H., P.D.W., and I.V. conceptionand design of research; Z.L., S.G.Z., A.A., H.H., V.A., H.W., P.D.W., and I.V.performed experiments; Z.L., S.G.Z., A.A., H.H., M.V., V.A., H.W., P.D.W.,and I.V. analyzed data; Z.L., S.G.Z., A.A., H.H., P.D.W., and I.V. interpretedresults of experiments; Z.L., S.G.Z., M.V., and I.V. prepared figures; Z.L.,S.G.Z., and I.V. drafted manuscript; Z.L., S.G.Z., A.A., H.H., P.D.W., and I.V.edited and revised manuscript; Z.L., S.G.Z., A.A., H.H., M.V., V.A., H.W.,P.D.W., and I.V. approved final version of manuscript.

REFERENCES

1. Agusti AG, Barbera JA, Roca J, Wagner PD, Guitart R, Rodriguez-Roisin R. Hypoxic pulmonary vasoconstriction and gas exchange duringexercise in chronic obstructive pulmonary disease. Chest 97: 268–275,1990.

2. Aliverti A, Stevenson N, Dellaca RL, Lo Mauro A, Pedotti A, CaverleyPM. Regional chest wall volumes during exercise in chronic obstructivepulmonary disease. Thorax 59: 210–216, 2004.

3. Aliverti A, Rodger K, Dellaca RL, Stevenson N, Lo Mauro A, PedottiA, Caverley PM. Effect of salbutamol on lung function and chest wallvolumes at rest and during exercise in COPD. Thorax 60: 916–924, 2005a.

4. Aliverti A, Dellaca RL, Lotti P, Bertini S, Duranti R, Scaco G,Heyman J, Lo Mauro A, Pedotti A, Macklem PT. Influence of expira-tory flow-limitation during exercise on systemic oxygen delivery inhumans. Eur J Appl Physiol 95: 229–242, 2005b.

5. Aliverti A, Kayser B, Macklem PT. A human model of the pathophys-iology of chronic obstructive pulmonary disease. Respirology 12: 478–485, 2007.

6. Aliverti A, Macklem PT. The major limitation to exercise performance inCOPD is inadequate energy supply to the respiratory and locomotormuscles. J Appl Physiol 105: 749–751, 2008.

7. Amann M, Regan MS, Kobitary M, Eldrigde MW, Boutellier U,Pegelow DF, Dempsey JA. Impact of pulmonary system limitations on



at Universiteit M

aastricht on Decem


Dow

nloaded from


locomotor muscle fatigue in patients with COPD. Am J Physiol RegulIntegr Comp Physiol 299: R314–R324, 2010.

8. Ambrossino N, Strambi S. New strategies to improve exercise tolerancein chronic obstructive pulmonary disease. Eur Respir J 24: 313–322, 2004.

9. ATS. ATS statement: guidelines for the six-minute walk test. Am J RespirCrit Care Med 166: 111–117, 2002.

10. Berton DC, Barbosa PB, Takara LS, Chiappa GR, Siqueira AC,Bravo DM, Ferreira LF, Neder JA. Bronchodilators accelerate thedynamics of muscle O2 delivery and utilisation during exercise in COPD.Thorax 65: 588–593, 2010.

11. Borg GA. Psychophysical bases of perceived exertion. Med Sci SportsExerc 14: 377–381, 1982.

12. Borghi-Silva A, Carneiro Oliveira C, Carrascosa C, Maia J, BertonDC, Queiroga F Jr, Ferreira EMV, Ribeiro D, Nery LE, Neder JA.Respiratory muscle unloading improves leg muscle oxygenation duringexercise in patients with COPD. Thorax 63: 910–915, 2008.

13. Boushel R, Langberg H, Olesen J, Nowak M, Simonsen L, Bulow,Kjaer M J. Regional blood flow during exercise in humans measured bynear-infrared spectroscopy and indocyanine green. J Appl Physiol 89:1868–1878, 2000.

14. Chiappa GR, Queiroga F Jr, Meda E, Ferreira LF, Diefenthaeler F,Nunes M, Vaz MA, Machado MC, Nery LE, Neder JA. Helioximproves oxygen delivery and utilization during dynamic exercise inpatients with chronic obstructive pulmonary disease. Am J Respir CritCare Med 179: 1004–1010, 2009.

15. Debigaré R, Maltais F. The major limitation to exercise performance inCOPD is lower limp muscle dysfunction. J Appl Physiol 105: 751–753,2008.

16. DeLorey DS, Kowalchuk JM, Paterson DH. Relationship betweenpulmonary O2 uptake kinetics and muscle deoxygenation during moder-ate-intensity exercise. J Appl Physiol 95: 113–120, 2003.

17. DeLorey DS, Kowalchuk JM, Paterson DH. Effects of prior heavyintensity exercise on pulmonary O2 uptake and muscle deoxygenationkinetics in young and older adult humans. J Appl Physiol 97: 998–1005,2004.

18. DeLorey DS, Kowalchuk JM, Paterson DH. Adaptation of pulmonaryO2 uptake kinetics and muscle deoxygenation at the onset of heavyintensity exercise in young and older adults. J Appl Physiol 98: 1697–1704, 2005.

19. Dow P. Estimations of cardiac output and central blood volume by dyedilution. Physiol Rev: 36: 77–102, 1956.

20. Duncan A, Meek JH, Clemence M, Elwell CE, Tyszczuk L, Cope M,Delpy DT. Optical pathlength measurements on adult head calf andforearm and the head of the newborn infant using phase resolved opticalspectroscopy. Phys Med Biol 40: 295–304, 1995.

21. Eves ND, Petersen SR, Haykowsky MJ, Wong EY, Jones RL. Heliox-hyperoxia, exercise, and respiratory mechanics in chronic obstructivepulmonary disease. Am J Respir Crit Care Med 174: 763–771, 2006.

22. Ferrari M, Mottola L, Quaresima V. Principles, techniques, and limi-tations of near infrared spectroscopy. Can J Appl Physiol 29: 463–487,2004.

23. Georgiadou O, Vogiatzis I, Stratakos G, Koutsoukou A, Golemati S,Aliverti A, Roussos C, Zakynthinos S. Effects of rehabilitation on chestwall volume regulation during exercise in COPD patients. Eur Respir J 29:284–291, 2007.

24. Global Initiative for Chronic Obstructive Lung Disease. 2005 Update:Executive Summary, Global Strategy for the Diagnosis, Management, andPrevention of COPD (Online). http://www.goldcopd.org/Guidelines.

25. Grassi B, Pogliaghi S, Rampichini S, Quaresima V, Ferrari M, Mar-coni C, Cerretelli P. Muscle oxygenation and pulmonary gas exchangekinetics during cycling exercise on-transitions in humans. J Appl Physiol95: 149–158, 2003.

26. Guenette JA, Vogiatzis I, Zakynthinos S, Athanasopoulos D, Kos-kolou M, Golemati S, Vasilopoulou M, Wagner HE, Roussos C,Wagner PD, Boushel R. Human respiratory muscle blood flow measuredby near-infrared spectroscopy and indocyanine green. J Appl Physiol 104:1202–1210, 2008.

27. Guenette JA, Webb KA, O’Donnell DE. Does dynamic hyperinflationcontribute to dyspnoea sensation during exercise in patients with COPD?Eur Respir J 40: 322–329, 2012.

28. Habazettl H, Athanasopoulos D, Kuebler WM, Wagner H, Roussos C,Wagner PD, Ungruhe J, Zakynthinos S, Vogiatzis I. Near-infraredspectroscopy and indocyanine green derived blood flow index for nonin-

vasive measurement of muscle perfusion during exercise. J Appl Physiol108: 962–967, 2010.

29. Harms CA, Babcock MA, McClaran SR, Pegelow DF, Nickele GA,Nelson WB, Dempsey JA. Respiratory muscle work compromises legblood flow during maximal exercise. J Appl Physiol 82: 1573–1583, 1997.

30. Iida H, Kanno I, Miura S, Murakami M, Takahashi K, Uemura K.Error analysis of a quantitative cerebral blood flow measurement usingH2(15)O autoradiography and positron emission tomography, with respectto the dispersion of the input function. J Cereb Blood Flow Metab 6:536–545, 1986.

31. Jörgensen K, Houltz E, Westfelt U, Nilsson F, Schersten H, RickstenSE. Effects of lung volume reduction surgery on left ventricular diastolicfilling and dimensions in patients with severe emphysema. Chest 124:1863–1870, 2003.

32. Jörgensen K, Muller MF, Nel J, Upton RN, Houltz E, Ricksten SE.Reduced intrathoracic blood volume and left and right ventricular dimen-sions in patients with severe emphysema: an MRI study. Chest 131:1050–1057, 2007.

33. Kalliokoski KK, Scheede-Bergdahl C, Kjaer M, Boushel R. Muscleperfusion and metabolic heterogeneity: insights from non-invasive imag-ing techniques. Exerc Sport Sci Rev 34: 164–170, 2006.

34. Laude EA, Duffy NC, Baveystock C, Dougill B, Campbell MJ, LawsonR, Jones PW, Calverley PM. The effect of heliox and oxygen on exerciseperformance in chronic obstructive pulmonary disease: a randomizedcrossover trial. Am J Respir Crit Care Med 173: 865–870, 2006.

35. Laveneziana P, Palange P, Ora J, Martolini D, O’Donnell DE. Bron-chodilator effect on ventilatory, pulmonary gas exchange, and heart ratekinetics during high-intensity exercise in COPD. Eur J Appl Physiol 107:633–643, 2009.

36. Laveneziana P, Valli G, Onorati P, Paoletti P, Ferrazza AM, PalangeP. Effect of heliox on heart rate kinetics and dynamic hyperinflationduring high-intensity exercise in COPD. Eur J Appl Physiol 111: 225–234,2011.

37. Light RW, Mintz HM, Linden GS, Brown SE. Hemodynamics ofpatients with severe chronic obstructive pulmonary disease during pro-gressive upright exercise. Am Rev Respir Dis 130: 391–395, 1984.

38. Macklem PT. Exercise in COPD: damned if you do and damned if youdon’t. Thorax 60: 887–888, 2005.

39. Magee F, Wright JL, Wiggs BR, Pare PD, Hogg JC. Pulmonaryvascular structure and function in chronic obstructive pulmonary disease.Thorax 43: 183–189, 1988.

40. Mahler DA, Brent BN, Loke J, Zaret BL, Matthay RA. Right ventric-ular performance and central circulatory hemodynamics during uprightexercise in patients with chronic obstructive pulmonary disease. Am RevRespir Dis 130: 722–729, 1984.

41. Matthay RA, Arroliga AC, Wiedemann HP, Schulman DS, MahlerDA. Right ventricular function at rest and during exercise in chronicobstructive pulmonary disease. Chest 101: 255–262, 1992.

42. O’Donnell DE, Lam M, Webb KA. Measurement of symptoms, lunghyperinflation, and endurance during exercise in chronic obstructive pul-monary disease. Am J Respir Crit Care Med 158: 1557–1565, 1998.

43. O’Donnell DE, Revill SM, Webb KA. Dynamic hyperinflation andexercise intolerance in chronic obstructive pulmonary disease. Am JRespir Crit Care Med 164: 770–777, 2001.

44. O’Donnell DE, Webb KA. The major limitation to exercise performancein COPD is dynamic hyperinflation. J Appl Physiol 105: 753–755, 2008.

45. Oelberg DA, Kacmarek M, Papagianopoulos PP, Ginns LC, SystromDM. Ventilatory and cardiovascular responses to inspired He-O2 duringexercise in chronic obstructive pulmonary disease. Am J Respir Crit CareMed 179: 1876–1882, 1998.

46. Oswald-Mammosser M, Apprill M, Bachez P, Ehrhart M, Weitzen-blum E. Pulmonary hemodynamics in chronic obstructive pulmonarydisease of the emphysematous type. Respiration 58: 304–310, 1991.

47. Palange P, Valli G, Onorati P, Antonucci R, Paoletti P, Rosato A,Manfredi F, Serra P. Effect of heliox on lung dynamic hyperinflation,dyspnea, and exercise endurance capacity in COPD patients. J ApplPhysiol 85: 1637–1642, 2004.

48. Palange P, Crimi E, Pellegrino R. Supplemental oxygen and heliox:“new” tools for exercise training in chronic obstructive pulmonary disease.Curr Opin Pulm Med 11: 145–8, 2005.

49. Sapirstein LA. Fractionation of the cardiac output of rats with isotopicpotassium. Circ Res 4: 689–692, 1956.



at Universiteit M

aastricht on Decem


Dow

nloaded from

http://www.goldcopd.org/Guidelines


50. Stark-Leyva KN, Beck CK, Jonhson BD. Influence of expiratory load-ing and hyperinflation on cardiac output during exercise. J Appl Physiol96: 1920–1927, 2004.

51. Swallow EB, Reyes D, Hopkinson NS, Man WD, Porcher R, Cetti EJ,Moore AJ, Moxham J, Polkey MI. Quadriceps strength predicts mortal-ity in patients with moderate to severe chronic obstructive pulmonarydisease. Thorax 62: 115–120, 2007.

52. Van der Zee P, Cope M, Arridge SR, Essenpreis M, Potter LA,Edwards AD, Wyatt JS, McCormick DC, Roth SC, Reynolds EO,Delpy DT. Experimentally measured optical pathlengths for the adulthead, calf and forearm and the head of the newborn infants as a functionof inter optode spacing. Adv Exp Med Biol 316: 143–153, 1992.

53. Vizza CD, Lynch JP, Ochoa LL, Richardson G, Trulock EP. Right andleft ventricular dysfunction in patients with severe pulmonary disease.Chest 113: 576–583, 1998.

54. Vogiatzis I, Stratakos G, Athanasopoulos D, Georgiadou O, GolematiS, Koutsoukou A, Weisman I, Roussos C, Zakynthinos S. Chest wallvolume regulation during exercise in COPD patients with GOLD stages IIto IV. Eur Respir J 32: 42–52, 2008a.

55. Vogiatzis I, Athanasopoulos D, Boushel R, Guenette JA, Koskolou M,Vasilopoulou M, Wagner H, Roussos C, Wagner PD, Zakynthinos S.Contribution of respiratory muscle blood flow to exercise-induced dia-phragmatic fatigue in trained cyclists. J Physiol 586: 5575–5587, 2008b.

56. Vogiatzis I, Athanasopoulos D, Habazettl H, Kuebler W, Wagner H,Roussos C, Wagner PD, Zakynthinos S. Intercostal muscle blood flowlimitation during exercise in trained athletes. J Physiol 587: 3665–3677,2009.

57. Vogiatzis I, Athanasopoulos D, Habazettl H, Aliverti A, Louvaris Z,Cherouveim E, Wagner H, Roussos C, Wagner PD, Zakynthinos S.Intercostal muscle blood flow limitation during exercise in chronic ob-structive pulmonary disease. Am J Respir Crit Care Med 182: 1105–1113,2010.

58. Vogiatzis I, Habazettl H, Aliverti A, Athanasopoulos D, Louvaris Z,Lo Mauro A, Wagner H, Roussos C, Wagner PD, Zakynthinos S.Effect of helium breathing on intercostal and quadriceps muscle bloodflow during exercise in COPD patients. Am J Physiol Regul Integr CompPhysiol 300: R1549–R1559, 2011.



at Universiteit M

aastricht on Decem


Dow

nloaded from


Heliox increases quadriceps muscle oxygen delivery during exercise in COPD patients with and without...

Documents

Transcript of Heliox increases quadriceps muscle oxygen delivery during exercise in COPD patients with and without...