Respiratory Physiology & Neurobiology

Transcription

1 Respiratory Physiology & Neurobiology 179 (2011) Contents lists available at SciVerse ScienceDirect Respiratory Physiology & Neurobiology j our nal ho me p age: Effects of dead space loading on neuro-muscular and neuro-ventilatory coupling of the respiratory system during exercise in healthy adults: Implications for dyspnea and exercise tolerance Dennis Jensen a,b, Denis E. O Donnell b, Ruifa Li c, Yuan-Ming Luo c, a Department of Kinesiology and Physical Education, McGill University, Montreal, QC, Canada b Respiratory Investigation Unit, Department of Medicine, Division of Respirology and Critical Care Medicine, Queen s University and Kingston General Hospital, Kingston, ON, Canada c State Key Laboratory of Respiratory Disease, First Affiliated Hospital of Guangzhou Medical College, Guangzhou, China a r t i c l e i n f o Article history: Accepted 17 August 2011 Keywords: Dyspnea Exercise Diaphragm Electromyography Dead space a b s t r a c t We examined the effects of dead space loading (DSL) on ventilation ( VE), neural respiratory drive (EMGdi%max, diaphragm EMG expressed as a % of maximal EMGdi), contractile respiratory muscle effort (Pes,tidal%P Imax, tidal esophageal pressure swing expressed as a % of maximal inspiratory Pes) and exertional dyspnea intensity ratings in 11 healthy adults with normal spirometry. Subjects completed, in random order, symptom-limited incremental cycle exercise tests under control (CTRL) and DSL (500 ml) conditions. Compared with CTRL, DSL decreased exercise tolerance by 20 25%; increased exertional dyspnea intensity ratings in direct proportion to concurrent increases in EMGdi%max, Pes,tidal%P Imax and VE; and had little/no effect on the inter-relationships between EMGdi%max, Pes,tidal%P Imax and VE during exercise. In conclusion, DSL was associated with an earlier onset of intolerable dyspnea; however, neuromuscular and neuro-ventilatory coupling of the respiratory system remained relatively preserved during exercise in the presence of an increased external dead space. Under these circumstances, DSL-induced increases in exertional dyspnea intensity ratings reflected, at least in part, the awareness of increased neural respiratory drive, contractile respiratory muscle effort and ventilatory output Elsevier B.V. All rights reserved. 1. Introduction Dyspnea (respiratory discomfort) on exertion is the dominant symptom of patients with chronic cardiorespiratory disease (e.g., obstructive and restrictive pulmonary disease, pulmonary vascular disease, congestive heart failure) and contributes importantly to exercise intolerance and an impoverished health-related qualityof-life in these patients (O Donnell et al., 2006, 2007, 2009; Sajkov et al., 2010). There is increasing evidence that troublesome activityrelated dyspnea and progressive reductions in aerobic working capacity also arises as a consequence of normative aging (Ofir et al., 2008; Jensen et al., 2009). In all cases, the increase in dyspnea and decrease in exercise tolerance is associated with concurrent increases in central (neural) respiratory motor command output, primarily as a result of increases in physiological dead space and/or ventilation-perfusion mismatching (Ofir et al., 2008; Jensen et al., 2009; O Donnell et al., 2009; Van der Plas et al., 2010). However, the specific inter-relationships between increased neural respiratory Corresponding author at: State Key Laboratory of Respiratory Disease, Guangzhou Medical College, 151 Yanjian Road, Guangzhou , China. Tel.: ; fax: address: ym3698@yahoo.com.cn (Y.-M. Luo). drive, exertional dyspnea and activity-limitation are not completely understood and represent the primary focus of this study. Increases in central motor command output to the respiratory muscles have the potential to curtail exercise performance by (i) accelerating the rise in exertional dyspnea intensity ratings thereby leading to an earlier onset of intolerable dyspnea, (ii) hastening or shortening the time to reach a critical mechanical constraint on thoracic volume displacement with attendant uncoupling of the inter-relationships between neural respiratory drive, contractile respiratory muscle pressure/force generation and ventilatory output or (iii) a combination of these factors. The purpose of the present study, therefore, was to examine the acute effects of dead space loading (DSL) on ventilation ( VE), breathing pattern, multipair esophageal electrode catheter-derived measures of the diaphragm electromyogram (EMGdi), esophageal pressure (Pes)-derived measures of contractile respiratory muscle effort, exertional dyspnea intensity ratings and exercise performance in healthy, older adults. Briefly, increased external physiological dead space stresses the ventilatory control system such that a greater VE is required to affect the same arterial P CO2 (Pa CO2 ) and H + regulation at any given metabolic rate, in accordance with the alveolar gas equation for CO 2 : VE = ( V CO2 863)/(Pa CO2 [1 VD/VT]), where V CO2 and VD/VT represent the metabolic rate of CO 2 production and /$ see front matter 2011 Elsevier B.V. All rights reserved. doi: /j.resp

2 220 D. Jensen et al. / Respiratory Physiology & Neurobiology 179 (2011) the physiological dead space-to-tidal volume ratio, respectively. We postulated that if significant dynamic respiratory mechanical constraints exist at peak exercise in this population (Ofir et al., 2008), then the otherwise harmonious inter-relationships between increased neural activation of the diaphragm, contractile respiratory muscle effort and ventilatory output will be disrupted, with attendant intensification of dyspnea at any given VE near endexercise. Alternatively, if healthy adults possess large ventilatory reserve, as some previous studies suggest (Leblanc et al., 1988; Dempsey et al., 1990; Johnson & Dempsey, 1991; Johnson et al., 1991; O Donnell et al., 2000), then indices of neuro-muscular and neuro-ventilatory coupling of the respiratory system will remain intact throughout exercise, even in the presence of an increased external dead space. Under these circumstances, DSL will be associated with an accelerated rise in exertional dyspnea intensity ratings and an earlier onset of intolerable dyspnea; however, dyspnea intensity ratings will not increase at any given VE during exercise near the limits of tolerance. 2. Methods 2.1. Ethical approval The study protocol and consent form were approved by the First Affiliated Hospital of Guangzhou Medical College Ethics Committee in accordance with the standards set by the Declaration of Helsinki. Written informed consent was obtained from all participants prior to study participation Experimental design This was a controlled, randomized, cross-over study in which subjects visited the laboratory on 3 separate occasions over a period of <2 weeks. Visit 1 included screening for eligibility, anthropometric measurements, pulmonary function testing and a symptom-limited incremental cardiopulmonary exercise test (CPET) for familiarization purposes. Visits 2 and 3 included CPET with added measurements of EMGdi and Pes under one of two conditions, randomized to order: unloaded control (CTRL) or increased external dead space (DSL). Each visit was separated by 24 hrs and all visits were conducted at approximately the same time of day (±1 h) for each subject. Subjects avoided caffeine, heavy meals, alcohol and major physical exertion entirely on the day of each test Subjects Subjects included 11 healthy, non-smoking, normal weight, Chinese men (n = 3) and women (n = 8) between 49 and 67 years of age (mean ± SEM 56.5 ± 1.6 years) with normal spirometry (forced expiratory volume in 1 s (FEV 1 ) 80%predicted; FEV 1 -to-forced vital capacity ratio (FEV 1 /FVC) >70%) and no history of respiratory, cardiovascular, neuromuscular, metabolic and/or musculoskeletal disease Pulmonary function testing Routine spirometry and constant-volume body plethysmography were performed (Miller et al., 2005a, 2005b; Wanger et al., 2005) using automated equipment (Q-Box, COSMED, Italy) and expressed as percentages of predicted normal values (Hou et al., 1990). A 12-s maximal voluntary ventilation maneuver was also performed (Miller et al., 2005b) Cardiopulmonary exercise testing Exercise tests were conducted on an electronically braked cycle ergometer (Ergoline 100P, COSMED, Italy) using an automated CPET system (ML870B80 Exercise Physiology System, ADInstruments, Castle Hill, Australia) and consisted of a steady-state resting period of 6 min, followed by 20 W increases in cycle work rate every 2-min to the point of symptom-limitation. At rest and during exercise, subjects breathed through a mouthpiece (with nasal passages occluded by noseclip) that was connected, in series, to a 2-way non-rebreathing valve and a pneumotachograph (apparatus dead space [VD app ] = 545 ml) to continuously measure inspiratory and expiratory airflow, which was then integrated to obtain volume. Mixed expired gases were measured at a port located in a mixing chamber using a calibrated O 2 /CO 2 analyzer. Oxyhemoglobin saturation and end-tidal P CO2 (PET CO2 ) were measured continuously by finger pulse oximetry and capnography (Capnocheck Plus, model , BCI, Waukesha, WI, USA), respectively. Changes in endexpiratory lung volume (EELV) were estimated from inspiratory capacity (IC) maneuvers performed at rest, at the end of every 2- min interval during exercise and at end-exercise. Assuming that total lung capacity does not change during exercise (Stubbing et al., 1980), changes in IC reflect changes in dynamic EELV. Pedaling cadence was maintained between 50 and 70 rpm, and subjects were verbally encouraged to cycle to the point of symptom-limitation. Signals were sampled at a rate of 200 Hz using a PowerLab 16/30 analog-to-digital converter running LabChart Pro Version 5.4 software (ADInstruments, Castle Hill, Australia); and were displayed in real time and saved for further off-line analysis Dead space loading A dead space of 500 ml was added to the breathing apparatus using a length of 35-mm tubing placed between the mouthpiece and the pneumotachograph (VD app = 1045 ml) Electromyogram of the diaphragm: measurement and analysis The electromyogram of the crural diaphragm (EMGdi) was recorded from a multipair esophageal electrode catheter and used as an index of neural respiratory drive (Luo et al., 2008). As illustrated in Fig. 1 of Luo et al. (2009), the catheter was 90 cm in length, 2.6 mm in diameter, and consisted of ten 1-cm silver coils and a latex balloon. The esophageal balloon was 7 cm in length and 2.4 cm in diameter with a 1 cm gap between it and the ground electrode (coil 0). The diameter of the inner lumen of the tube used for esophageal pressure measurement was 1 mm. The most proximal coil (coil 0) was connected to the ground and was 2 cm away from coil 1. Coils 1 9 were used for recording and were 1 cm in length with a gap of 1 mm between adjacent coils. Five consecutive recording pairs were formed with an inter-electrode distance of 3.4 cm within a pair. The combined electrode-balloon catheter was passed through the nose and carefully positioned based on the strength of EMGdi recorded simultaneously from different pairs of electrodes during spontaneous breathing. Briefly, the position of the electrode catheter was known to be optimal (that is, electrode 5 positioned at the electrically active center of the diaphragm) when the amplitude of EMGdi during inspiration was greatest in electrode pairs 1 and 5, and lowest in electrode pair 3. The placement and position of the catheter was the same under CTRL and DSL conditions. The EMGdi signals were sampled at a rate of 2000 Hz using a PowerLab 16/30 analog-to-digital converter running LabChart Pro Version 5.4 software; amplified and band-pass filtered between 20 and 1000 Hz (bio-amplifier model RA-8, Guangzhou Yinghui Medical Equipment

3 D. Jensen et al. / Respiratory Physiology & Neurobiology 179 (2011) Co. Ltd, Guangzhou, China); and displayed in real time and saved for further off-line analysis. The raw EMGdi signal was converted to root mean square (RMS) using a time constant of 100 ms and a moving window. The maximum RMS value of each inspired breath was then determined by manually selecting EMGdi signals falling between QRS complexes to avoid the influence of ECG artifact on EMGdi. The RMS selected was that from the electrode pair with the largest EMGdi amplitude for each breath. Maximum EMGdi (EMGdi,max) was identified as the largest of all the RMS values obtained from IC maneuvers performed either at rest or during exercise. The ratio of EMGdi to EMGdi,max (EMGdi%max) was used as an index of neural respiratory drive Esophageal pressure: measurement and analysis Inspiratory (Pes,insp) and expiratory (Pes,expir) esophageal pressures were collected on a breath-by-breath basis using a balloon containing 0.8 ml of air and mounted on the electrode catheter. When the catheter was optimally positioned based on the strength of EMGdi (see above), the balloon was positioned in the midesophagus, as confirmed by maximal inspiratory sniff maneuvers. Esophageal pressure was sampled continuously at a rate of 200 Hz using a differential pressure transducer (model DP15-34, Validyne Engineering, Northridge, CA, USA), signal conditioner (model CD280-4, Validyne Engineering) and a PowerLab 16/30 analog-todigital converter running LabChart Pro Version 5.4 software. Signals were displayed in real time and saved for further off-line analysis. Inspiratory sniff maneuvers were performed pre-exercise at rest (ATS/ERS, 2002) to obtain maximal values for inspiratory Pes (P Imax ). Pes,insp and Pes,expir were taken as the peak inspiratory and expiratory esophageal pressure during tidal inspiration and expiration, respectively. The tidal Pes swing (Pes,tidal) was calculated as the difference between Pes,insp and Pes,expir. The ratio of Pes,tidal to P Imax (Pes,tidal%P Imax ) was used as an index of contractile respiratory muscle effort Symptom evaluation Using Borg s 0 10 category ratio scale (1982), subjects rated the intensity of their perceived breathing (dyspnea) and leg discomfort at rest, within the last 30 s of every 2nd minute during exercise, and at end-exercise. Before each CPET, subjects were familiarized with the Borg scale and its endpoints were anchored such that 0 represented no breathing (or leg) discomfort and 10 represented the most severe breathing (or leg) discomfort you have ever experienced or could ever imagine experiencing. Upon exercise cessation, subjects were asked to verbalize their main reason for stopping exercise (i.e., dyspnea; leg discomfort; both dyspnea and leg discomfort; other) Analysis of exercise end-points Measured parameters were averaged in 30-s intervals at rest and during exercise. These parameters, collected over the first 30-s period of every 2nd minute during exercise, were linked with Borg ratings and IC measurements collected in the latter 30-s of the same minute. Pre-exercise rest was the average of the last 30-s of the steady-state period after at least 3-min of quiet breathing before the start of exercise. Iso-work was the highest equivalent work rate achieved during both of the CTRL and DSL cycle exercise tests performed by a given subject. Peak exercise was the average of the last 30-s of loaded pedaling. Peak work rate was the highest cycle work rate that the subject was able to maintain for at least 30-s. The slope of the relationship between increasing dyspnea intensity ratings and each of VE, EMGdi%max and Pes,tidal%P Imax during exercise was calculated for each subject under CTRL and DSL conditions by Table 1 Subject characteristics and pulmonary function test parameters. Parameter Age, years 56.5 ± 1.6 Height, cm ± 1.8 BMI, kg/m ± 0.8 Pulmonary function test parameters FEV 1, l [% predicted] 2.22 ± 0.09 [102 ± 5] FEV 1/FVC, % [% predicted] 79.8 ± 1.4 [102 ± 2] FVC, l [% predicted] 2.79 ± 0.12 [107 ± 5] PEFR, l/s [% predicted] 6.06 ± 0.31 [103 ± 4] FEF 25 75%, l/s [% predicted] 2.17 ± 0.16 [75 ± 5] MVV, l/min 97.1 ± 5.3 TLC, l [% predicted] 5.21 ± 0.29 [107 ± 4] SVC, l [% predicted] 3.22 ± 0.15 [116 ± 6] IC, l [% predicted] 2.06 ± 0.14 [126 ± 10] FRC, l [% predicted] 3.20 ± 0.17 [116 ± 4] RV, l [% predicted] 2.15 ± 0.20 [116 ± 10] Values are means ± SEM. BMI, body mass index; FEV 1, forced expiratory volume in one second; FVC, forced vital capacity; PEFR, peak expiratory flow rate; FEF 25 75%, forced expiratory flow between 25 and 75% of the FVC maneuvre; MVV, maximal voluntary ventilation; TLC, total lung capacity; SVC, slow vital capacity; IC, inspiratory capacity; FRC, functional residual capacity; RV, residual volume. linear regression analysis using dyspnea intensity ratings 0.5 Borg units. Similarly, the slope of the relationship between increasing VE and V CO2 was calculated for each subject under CTRL and DSL conditions by linear regression analysis using the 30-s averaged data from rest through to end-exercise Statistical analysis Paired t-tests were used to examine the effects of DSL vs. CTRL on measured parameters at rest, at standardized submaximal work rates during exercise, and at end-exercise (SigmaStat for Windows Version 3.10, Systat Software, Inc., San Jose, CA). A p < 0.05 level of statistical significance was used for all analyses. The results are reported as mean ± SEM. 3. Results 3.1. Subjects Physical characteristics and pulmonary function test parameters are presented in Table 1. Spirometric parameters and plethysmographic lung volumes were within normal age, sex and height predicted limits Symptom-limited incremental cycle exercise Compared to CTRL, DSL significantly decreased aerobic working capacity: peak work rate, peak V O2 and the duration of loaded pedaling decreased by 24 ± 7 W ( 22%), 4.8 ± 1.4 ml/kg/min ( 20%) and 2.5 ± 0.7 min ( 25%), respectively (Table 2). However, V O2 /work rate (10.6 ± 0.7 vs ± 0.8 ml/min/w, p = 0.634) and V CO2 /work rate (12.4 ± 0.7 vs ± 0.09 ml/min/w, p = 0.096) slopes were not significantly different during incremental cycle exercise under CTRL and DSL conditions. Values of P Imax ( 62.5 ± 6.4 vs ± 5.8 cmh 2 O, p = 0.706) and EMGdi,max (225.4 ± 17.9 vs ± 13.9 V, p = 0.262) were similar under DSL vs. CTRL conditions. Compared with CTRL, DSL was associated with consistent increases in PET CO2, VE, tidal volume (VT), breathing frequency (fr), Pes,tidal%P Imax and EMGdi%max at rest and at any given submaximal work rate during exercise (Fig. 1, Table 2); an increase in the slope of the VE V CO2 relationship during exercise (41.0 ± 3.0 vs ± 5.7, p = 0.005); no change in the behavior of dynamic IC (or EELV) during exercise (Table 2); a discernible plateau in VT expansion

4 222 D. Jensen et al. / Respiratory Physiology & Neurobiology 179 (2011) Table 2 Effects of dead space loading on physiological and perceptual responses at rest, at the highest equivalent work rate (iso-work) and at the symptom-limited peak of symptomlimited incremental cycle exercise. Parameter REST ISO-WORK PEAK CTRL DSL CTRL DSL CTRL DSL Time, min 6.7 ± ± ± ± 0.6 * Work rate, W 67.3 ± ± ± ± 6.3 * Dyspnea intensity, Borg scale units 0.1 ± ± ± ± 0.8 * 6.3 ± ± 0.7 Leg discomfort, Borg scale units 0.0 ± ± ± ± 0.9 * 6.0 ± ± 1.0 V O2, ml/kg/min 4.3 ± ± ± ± ± ± 1.6 * V CO2, ml/kg/min 3.1 ± ± ± ± ± ± 1.8 * VE, l/min 15.6 ± ± 0.9 * 45.4 ± ± 3.5 * 69.5 ± ± 3.4 VE, %MVV 16.4 ± ± 1.8 * 48.2 ± ± 4.4 * 73.3 ± ± 4.9 VT, l 0.88 ± ± 0.04 * 1.56 ± ± 0.06 * 1.68 ± ± 0.09 fr, breaths/min 18.2 ± ± 0.9 * 29.1 ± ± 1.7 * 41.6 ± ± 1.5 PET CO2, mmhg 44.1 ± ± 1.2 * 48.1 ± ± 1.6 * 48.1 ± ± 1.8 * Change in PET CO2 from rest, mmhg 4.0 ± ± 0.9 * 4.0 ± ± 1.0 * SpO 2, % 97.8 ± ± ± ± ± ± 2.4 IC, l 2.04 ± ± ± ± ± ± 0.11 Change in IC from rest, l 0.12 ± ± ± ± 0.14 Pes,insp, cmh 2O 11.2 ± ± 1.2 * 15.5 ± ± 2.1 * 18.4 ± ± 2.1 * Pes,expir, cmh 2O 2.1 ± ± ± ± 1.8 * 10.2 ± ± 3.0 Pes,tidal, cmh 2O 9.1 ± ± 1.0 * 17.7 ± ± 3.0 * 28.6 ± ± 4.3 * Pes,tidal, %P Imax 16.0 ± ± 1.7 # 30.1 ± ± 4.6 * 50.0 ± ± 6.2 EMGdi, V 42.4 ± ± 4.5 * ± ± 12.1 * ± ± 11.3 EMGdi, %EMGdi,max 20.2 ± ± 2.6 * 45.1 ± ± 2.9 * 65.8 ± ± 3.2 Values are means ± SEM. CTRL, control condition; DSL, dead space loading condition; V O2, rate of oxygen consumption; V CO2, rate of carbon dioxide production; VE, minute ventilation; MVV, maximal voluntary ventilation; VT, tidal volume, fr, breathing frequency; PET CO2, partial pressure of end-tidal carbon dioxide; SpO 2, oxyhemoglobin saturation; IC, inspiratory capacity; Pes,insp and Pes, expir, inspiratory and expiratory esophageal pressure, respectively; Pes,tidal, tidal esophageal pressure swing; P Imax, maximal inspiratory esophageal pressure; EMGdi, root mean square of the diaphragm electromyogram; EMGdi,max, maximal root mean square of the diaphragm electromyogram. * p < 0.05 CTRL vs. DSL. # p = CTRL vs. DSL. between iso-work and peak exercise (Fig. 1, Table 2); and a significantly greater ( 2-fold) mean change in PET CO2 from rest to end-exercise (Fig. 1, Table 2). Mean values of VE, Pes,tidal%P Imax and EMGdi%max at end-exercise under DSL vs. CTRL conditions were increased by 3.7 ± 2.0 l/min (p = 0.088), 10.7 ± 5.3% P Imax (p = 0.071) and 8.6 ± 4.3% EMGdi,max (p = 0.073), respectively; however, these differences were not statistically significant (Fig. 1, Table 2). Compared with CTRL, DSL had little/no effect on the interrelationships between EMGdi%max, Pes,tidal%P Imax and VE during symptom-limited incremental cycle exercise (Fig. 2), indicating relative preservation of neuro-muscular and neuro-ventilatory coupling of the respiratory system. Dyspnea was identified as the primary symptom limiting exercise in 9 vs. 4 subjects under DSL vs. CTRL conditions, respectively. The intensity of perceived dyspnea and leg discomfort was similar at end-exercise under DSL vs. CTRL conditions (Fig. 3, Table 2). Compared with CTRL, DSL was associated with significant increases in Borg ratings of perceived dyspnea (Fig. 3, Table 2) and leg discomfort (Table 2) at any given submaximal work rate. As illustrated in Fig. 3, however, DSL had no statistically significant effect on the slope of the relationship between increasing dyspnea intensity ratings and each of EMGdi%max (CTRL, 0.16 ± 0.02 vs. DSL, 0.21 ± 0.04; p = 0.292); Pes,tidal%P Imax (CTRL, 0.23 ± 0.04 vs. DSL, 0.25 ± 0.07; p = 0.839); and VE (CTRL, 0.22 ± 0.06 vs. DSL, 0.18 ± 0.02; p = 0.377) during symptom-limited incremental cycle exercise. 4. Discussion To our knowledge, the present study is the first to investigate specifically whether stressing the ventilatory control system via DSL affects exertional dyspnea and exercise tolerance by disrupting the otherwise harmonious inter-relationships between increased neural activation of the diaphragm (EMGdi%max), contractile respiratory muscle effort (Pes,tidal%P Imax ) and ventilatory output ( VE) during incremental cycle exercise in healthy adults. The main findings of this study are as follows: (1) DSL provoked an earlier onset of intolerable dyspnea (with concomitant reductions in exercise tolerance) by accelerating the rise in exertional dyspnea intensity ratings; (2) DSL-induced increases in exertional dyspnea intensity ratings rose linearly with increases in neural respiratory drive, contractile respiratory muscle effort and ventilatory output; and (3) neuro-muscular and neuro-ventilatory coupling of the respiratory system was relatively preserved during DSL vs. CTRL exercise. These findings are consistent with the hypothesis that healthy adults possess large ventilatory reserve and that significant dynamic respiratory mechanical constraints do not exist during exercise in this population Increased neural respiratory drive, exertional dyspnea and activity-limitation DSL-induced reductions in aerobic working capacity (by 20 25%) occurred in tandem with an accelerated rise in exertional dyspnea intensity ratings and an earlier onset of intolerable dyspnea: dyspnea intensity ratings were uniformly increased at any given submaximal work rate during DSL vs. CTRL exercise (e.g., by 2.2 ± 0.3 Borg scale units at iso-work); and subjects reached their symptom-limited peak at an earlier time and at a relatively reduced V O2 and work rate under DSL vs. CTRL conditions. In keeping with the results of previous studies (Jones et al., 1971; Ward & Whipp, 1980; Poon, 1992; McClaren et al., 1999; O Donnell et al., 2000; Wood et al., 2008, 2010), we also found that, compared with CTRL, DSL significantly increased VE, EMGdi%max and Pes,tidal%P Imax at any given work rate during exercise (e.g., by 17.2 ± 2.2 l/min, 19.2 ± 3.0%EMGdi,max and 17.2 ± 4.5%P Imax at isowork, respectively). By contrast, increased external dead space had little/no effect on the behavior of dynamic IC (or EELV) or on the inter-relationships between EMGdi%max, Pes,tidal%P Imax and VE during symptom-limited incremental cycle exercise. These findings corroborate and extend those of O Donnell et al. (2000), and provide evidence to suggest that, in healthy adults with normal pulmonary function, the increased ventilatory requirements

5 D. Jensen et al. / Respiratory Physiology & Neurobiology 179 (2011) Fig. 1. Effects of dead space loading (500 ml) on ventilatory, breathing pattern, neural respiratory drive and contractile respiratory muscle effort responses to symptomlimited incremental cycle exercise. Data points are means ± SEM values at rest, at standardized submaximal work rates and at peak exercise. Open circles ( ) = dead space loading (DSL) condition. Solid black squares ( ) = unloaded control (CTRL) condition. PET CO2, partial pressure of end-tidal carbon dioxide; EMGdi%max, root mean square (RMS) of the diaphragm electromyogram (EMGdi) expressed as a percentage of the maximal RMS EMGdi, representing neural respiratory motor drive; Pes,tidal%P Imax, tidal esophageal pressure (Pes) swing expressed as a percentage of maximal inspiratory Pes, representing contractile respiratory muscle effort. *p < 0.05 CTRL vs. DSL, paired t-test; # p = CTRL vs. DSL, paired t-test. during exercise in the setting of an increased external dead space were accommodated within the most compliant (linear) portion of the respiratory system s sigmoid pressure-volume relation, where neuro-muscular and neuro-ventilatory coupling of the respiratory system is harmoniously preserved. Based on our current understanding of the neurophysiology of exertional dyspnea (ATS, 1999; O Donnell et al., 2006, 2009; Jensen et al., 2009), it is reasonable to conclude that, under these circumstances, DSL-induced increases in exertional dyspnea intensity ratings likely reflect the awareness of (i) increased central motor command output to the respiratory muscles, as sensed by increased central corollary discharge to the somatosensory cortex; (ii) increased feedback information from sensory afferent nerves located in the lungs, airways and respiratory musculature, secondary to increased ventilatory output and/or contractile (global) respiratory muscle force/pressure generation; or (iii) some combination thereof. Indeed, as illustrated in Fig. 3, the slope of the linear relationship between increasing dyspnea intensity ratings and each of EMGdi%max, Pes,tidal%P Imax and VE was similar throughout exercise under DSL vs. CTRL conditions. Thus, dyspnea intensity ratings rose linearly (and to a similar symptomlimited peak compared with CTRL) in association with concomitant increases in neural respiratory drive, contractile respiratory muscle effort and ventilatory output during exercise in the setting of an increased external dead space. Unfortunately, the relative contribution of these dyspneogenic stimuli could not be ascertained. Based on the collective results of previous studies (Harms et al., 1997, 1998, 2000; Wetter et al., 1999; Romer et al., 2006), it is reasonable to speculate that DSL-induced increases in contractile respiratory muscle effort (and presumably, therefore, the work and O 2 cost of breathing) may have contributed, at least in part, to (i) increases in the intensity of perceived leg discomfort (Table 2) and (ii) reductions in exercise performance by causing a redistribution of total blood flow and O 2 from the legs to the respiratory muscles (i.e., the respiratory muscle steal hypothesis), particularly toward the higher intensities of incremental exercise. This possibility requires further investigation.

6 224 D. Jensen et al. / Respiratory Physiology & Neurobiology 179 (2011) alveolar P CO2 during the respiratory cycle, increases above resting levels during exercise. In the present study, exercise-induced increases in V CO2 (and presumably, therefore, mixed-venous P CO2 ) were similar under DSL vs. CTRL conditions; however, VT was consistently increased at rest and during exercise in the former (Fig. 2, Table 2). These differences in breathing pattern may explain, at least in part, the greater ( 2-fold) mean change in PET CO2 from rest to end-exercise in the presence of an increased external dead space (Fig. 1, Table 2). From a methodological standpoint, the relative preservation of VE-EMGdi%max and VE-Pes,tidal%P Imax relationships during DSL vs. CTRL exercise (Fig. 2) provides evidence to suggest that simple, routine and non-invasive assessment of VE may be used as a valid and reliable index of neural respiratory drive and contractile respiratory muscle effort during incremental cycle exercise, at least in healthy, normal weight adults with normal respiratory mechanical/muscular function. However, this may not apply under pathological conditions of the lungs, airways, chest wall and/or respiratory musculature (Sinderby et al., 2001; Steier et al., 2009; Qin et al., 2010; Reilly et al., 2011) Critique of methods Fig. 2. Effects of dead space loading (500 ml) on the inter-relationships between neural respiratory drive, contractile respiratory muscle effort and ventilatory output during symptom-limited incremental cycle exercise. Data points are means ± SEM values at rest, at standardized submaximal work rates and at peak exercise. Open circles ( ) = dead space loading (DSL) condition. Solid black squares ( ) = unloaded control (CTRL) condition. EMGdi%max, root mean square (RMS) of the diaphragm electromyogram (EMGdi) expressed as a percentage of the maximal RMS EMGdi, representing neural respiratory motor drive; Pes,tidal%P Imax, tidal esophageal pressure (Pes) swing expressed as a percentage of maximal inspiratory Pes, representing contractile respiratory muscle effort. We increased external dead space by 500 ml, which increased VT by 250 ml, 200 ml and 100 ml vs. CTRL at rest, iso-work and end-exercise, respectively (Fig. 2, Table 2). It follows that the VD/VT ratio was higher and alveolar ventilation lower under DSL vs. CTRL conditions, thereby explaining the increase in PET CO2 (Fig. 2, Table 2). With exercise there is an increase in VT, V CO2 and mixedvenous P CO2, such that the within-breath fluctuations of alveolar gas composition progressively increase relative to rest (Wasserman et al., 2005). Because of the increased rate of CO 2 delivery to the lung associated with the increased V CO2 of exercise, alveolar P CO2 increases toward mixed-venous P CO2 levels during expiration more rapidly during exercise than at rest (Jones et al., 1979; Wasserman et al., 2005). As a result, PET CO2, which represents the highest Criticisms of using a multipair esophageal electrode catheter positioned at the crus of the diaphragm to assess neural respiratory drive in humans have been described in detail elsewhere (Luo et al., 2001, 2008; ATS/ERS, 2002; Qin et al., 2010). Briefly, the use of EMGdi as a representative measure of neural respiratory drive is based on the fact that (i) the phrenic nerve is the only motor nerve of the diaphragm and (ii) there is a strong positive correlation between simultaneously measured changes in phrenic nerve EMG and diaphragm EMG during both normal and obstructed breathing (Lourenco et al., 1966; Aubier et al., 1981; Hussain et al., 1985). In addition, both animal (D Angelo et al., 2010) and human (Amirjani et al., 2011) studies have indicated uniform changes in the level of crural and costal diaphragm EMG activity during various respiratory maneuvers. It could be argued that DSL-induced changes in the behavior of dynamic operating lung volumes (e.g., EELV), the velocity of shortening of the diaphragm and/or the spatial recruitment of other respiratory muscles (e.g., intercostals, accessory muscles) during exercise may have influenced our measurement of EMGdi and subsequent interpretation of results. However, in the present study, no significant difference in the behavior of dynamic IC (or EELV) was observed during exercise under DSL vs. CTRL conditions (Table 2). Furthermore, Luo et al. (2000) previously found that experimentally induced increases in EELV of up to 2 l above normal resting levels had little effect on multipair esophageal electrode catheter derived measures of the amplitude of the diaphragm compound muscle action potential. Moreover, the results of an earlier study by Beck et al. (1998) provided no evidence for an influence of increased inspiratory flow rates of up to 1.4 l/s on multipair esophageal electrode catheter derived measures of EMGdi in healthy, young volunteers. In light of the above, it is unlikely that DSL-induced alterations in dynamic operating lung volumes and/or the velocity of diaphragm shortening influenced our measurements and interpretation of results. Nevertheless, we cannot rule out the possibility that differences in the spatial recruitment of other respiratory muscles during DSL vs. CTRL exercise may have helped to unload the diaphragm and, in this manner, preserve the inter-relationships between EMGdi%max, Pes,tidal%P Imax, VE and dyspnea intensity ratings, particularly near the limits of tolerance. Unlike the commonly observed response of a reduction in PET CO2 from mild-to-peak exercise (Wasserman et al., 2005), there was no drop in PET CO2 during the latter stages of exercise under both CTRL and DSL conditions (Fig. 1, Table 2). The reason for

7 D. Jensen et al. / Respiratory Physiology & Neurobiology 179 (2011) Fig. 3. Effects of dead space loading (500 ml) on the relationship between increasing dyspnea intensity ratings and each of cycle work rate, neural respiratory drive, contractile respiratory muscle effort and ventilatory output during symptom-limited incremental cycle exercise. Data points are means ± SEM values at rest, at standardized submaximal work rates and at peak exercise. Open circles ( ) = dead space loading (DSL) condition. Solid black squares ( ) = unloaded control (CTRL) condition. EMGdi%max, root mean square (RMS) of the diaphragm electromyogram (EMGdi) expressed as a percentage of the maximal RMS EMGdi, representing neural respiratory motor drive; Pes,tidal%P Imax, tidal esophageal pressure (Pes) swing expressed as a percentage of maximal inspiratory Pes, representing contractile respiratory muscle effort. Compared with CTRL, DSL had no statistically significant effect on the slope of the following relationships during exercise: dyspnea intensity-emgdi%max (CTRL, 0.16 ± 0.02 vs. DSL, 0.21 ± 0.04; p = 0.292, paired t-test); dyspnea intensity-pes,tidal%p Imax (CTRL, 0.23 ± 0.04 vs. DSL, 0.25 ± 0.07; p = 0.839, paired t-test); and dyspnea intensity-ventilation (CTRL, 0.22 ± 0.06 vs. DSL, 0.18 ± 0.02; p = 0.377, paired t-test). * p < 0.05 CTRL vs. DSL, paired t-test. these atypical responses is unclear, but may reflect the relatively large VD app (545 ml and 1045 ml under CTRL and DSL conditions, respectively), which, in the setting of progressive exercise-induced increases in VT, V CO2 and mixed-venous P CO2, may have served to increase the P CO2 of the gas that remained in the breathing apparatus at end-expiration during exercise. The incremental cycle exercise test employed in the current study provided only a short period of time (i.e., 2-min) in which the subjects were exposed to the conditions that may have predisposed them to dynamic respiratory mechanical constraints. In this regard, a series of high-intensity, constant work rate cycle exercise tests (e.g., 60%, 80%, 100% of the maximal incremental work rate) may have provided a better opportunity to determine whether DSL affects the inter-relationships between EMGdi%max, Pes,tidal%P Imax and VE in healthy adults. Finally, we must acknowledge that the generalizability of our results may be limited to the characteristics of the subjects studied here (i.e., healthy, non-smoking, normal weight, Chinese adults the majority of whom were women). 5. Conclusions In conclusion: (1) DSL decreased exercise tolerance, accelerated the rise in exertional dyspnea intensity ratings, and had little/no demonstrable effect on neuro-muscular and neuro-ventilatory coupling of the respiratory system during exercise in healthy adults; and (2) DSL-induced increases in exertional dyspnea intensity ratings rose linearly in association with concomitant increases in neural respiratory drive, contractile respiratory muscle effort and ventilatory output. Authors contributions DJ contributed to conception and design of the experiment; collection, analysis and interpretation of data; and drafting of the article. RL contributed to the collection and analysis of data. DEO contributed to the interpretation of data and drafting of the article. YML contributed to conception and design of the experiment; the interpretation of data; and drafting of the article. Acknowledgments This study was supported by the Chinese Natural Scientific Foundation (Grant # ). D. Jensen was supported by a Canadian Thoracic Society/Canadian Lung Association Post-Doctoral Research Fellowship Award and a Queen s University Post-Doctoral Fellow Excellence in Research Award. References American Thoracic Society, Dyspnea: mechanisms, assessment, and management: a consensus statement. Am. J. Respir. Crit. Care Med. 159, American Thoracic Society/European Respiratory Society, ATS/ERS Statement on respiratory muscle testing. Am. J. Respir. Crit. Care Med. 166,

8 226 D. Jensen et al. / Respiratory Physiology & Neurobiology 179 (2011) Amirjani, N., Hudson, A.L., Butler, J.E., Gandevia, S.C., Comparison of EMG in the costal and crural diaphragm with increments in tidal volume. Am. J. Respir. Crit. Care Med. 183, Aubier, M., Trippenbach, T., Roussos, C., Respiratory muscle fatigue during cardiogenic shock. J. Appl. Physiol. 51, Beck, J., Sinderby, C., Lindstrom, L., Grassino, A., Crural diaphragm activation during contractions at various inspiratory flow rates. J. Appl. Physiol. 85, Borg, G.A., Psychophysical bases of perceived exertion. Med. Sci. Sports Exerc. 14, D Angelo, E., Monaco, A., D Angelo, E., Pecchiari, M., Motor control of the diaphragm in anesthetized rabbits. Respir. Physiol. Neurobiol. 170, Dempsey, J.A., Johnson, B.D., Saupe, K.W., Adaptations and limitations in the pulmonary system during exercise. Chest 97, 81S 87S. Harms, C.A., Babcock, M.A., McClaran, S.R., Pegelow, D.F., Nickele, G.A., Nelson, W.B., Dempsey, J.A., Respiratory muscle work compromises leg blood flow during maximal exercise. J. Appl. Physiol. 82, Harms, C.A., Wetter, T.J., McClaran, S.R., Pegelow, D.F., Nickele, G.A., Nelson, W.B., Hanson, P., Dempsey, J.A., Effects of respiratory muscle work on cardiac output and its distribution during maximal exercise. J. Appl. Physiol. 85, Harms, C.A., Wetter, T.J., St Croix, C.M., Pegelow, D.F., Dempsey, J.A., Effects of respiratory muscle work on exercise performance. J. Appl. Physiol. 89, Hou, S., Zhan, Y.G., Guo, S.C., He, J., In: Nationwide normal values of lung function, Mu, K.J., Liu, S.W. (Eds.), Normal values of lung function in the population of Guangdong. PUMC & Beijing Medical University Publication, pp Hussain, S.A., Simkus, G., Roussos, C., Respiratory muscle fatigue: a cause of ventilatory failure in septic shock. J. Appl. Physiol. 58, Jensen, D., Ofir, D., O Donnell, D.E., Effects of pregnancy, obesity and aging on the intensity of perceived breathlessness during exercise in healthy humans. Respir. Physiol. Neurobiol. 167, Johnson, B.D., Dempsey, J.A., Demand vs. capacity in the aging pulmonary system. Exerc. Sport Sci. Rev. 19, Johnson, B.D., Reddan, W.G., Seow, K.G., Dempsey, J.A., Mechanical constraints on exercise hyperpnea in a fit aging population. Am. Rev. Respir. Dis. 143, Jones, N.L., Levine, G.B., Robertson, D.G., Epstein, S.W., The effect of added dead space on the pulmonary response to exercise. Respiration 28, Jones, N.L., Robertson, D.G., Kane, J.W., Difference between end-tidal and arterial P CO2 in exercise. J. Appl. Physiol. 47, Leblanc, P., Summers, E., Inman, M.D., Jones, N.L., Campbell, E.J., Killian, K.J., Inspiratory muscles during exercise: a problem of supply and demand. J. Appl. Physiol. 64, Lourenco, R.V., Cherniack, N.S., Malm, J.R., Fishman, A.P., Nervous output from the respiratory center during obstructed breathing. J. Appl. Physiol. 21, Luo, Y.M., Hart, N., Mustfa, N., Lyall, R.A., Polkey, M.I., Moxham, J., Effect of diaphragm fatigue on neural respiratory drive. J. Appl. Physiol. 90, Luo, Y.M., Lyall, R.A., Harris, M.L., Hawkins, P., Hart, N., Polkey, M.I., Moxham, J., Effect of lung volume on the esophageal diaphragm EMG assessed by magnetic phrenic nerve stimulation. Eur. Respir. J. 15, Luo, Y.M., Moxham, J., Polkey, M.I., Diaphragm electromyography using an oesophageal catheter: current concepts. Clin. Sci. 115, Luo, Y.M., Tang, J., Jolley, C., Steier, J., Zhong, N.S., Moxham, J., Polkey, M.I., Distinguishing obstructive from central sleep apnea events: diaphragm electromyogram and esophageal pressure compared. Chest 135, McClaren, S.R., Wetter, T.J., Pegelow, D.F., Dempsey, J.A., Role of expiratory flow limitation in determining lung volumes and ventilation during exercise. J. Appl. Physiol. 86, Miller, M.R., Crapo, R., Hankinson, J., Brusasco, V., Burgos, F., Casaburi, R., Coates, A., Enright, P., van der Grinten, C.P., Gustafsson, P., Jensen, R., Johnson, D.C., MacIntyre, N., McKay, R., Navajas, D., Pedersen, O.F., Pellegrino, R., Viegi, G., Wanger, J., ATS/ERS Task Force, 2005a. General considerations for lung function testing. Eur. Respir. J. 26, Miller, M.R., Hankinson, J., Brusasco, V., Burgos, F., Casaburi, R., Coates, A., Crapo, R., Enright, P., van der Grinten, C.P., Gustaffson, P., Jensen, R., MacIntyre, N., McKay, R., Navajas, D., Pedersen, O.F., Pellegrino, R., Viegi, G., Wanger, J., ATS/ERS Task Forces, 2005b. Standardisation of spirometry. Eur. Respir. J. 26, O Donnell, D.E., Banzett, R.B., Carrieri-Kohlman, V., Casaburi, R., Davenport, P.W., Gandevia, S.C., Gelb, A.F., Mahler, D.A., Webb, K.A., Pathophysiology of dyspnea in chronic obstructive pulmonary disease: a roundtable. Proc. Am. Thorac. Soc. 4, O Donnell, D.E., Hong, H.H., Webb, K.A., Respiratory sensation during chest wall restriction and dead space loading in exercising men. J. Appl. Physiol. 88, O Donnell, D.E., Ofir, D., Laveneziana, P., Patterns of cardiopulmonary response to exercise in lung diseases. Eur. Respir. Mon. 40, O Donnell, D.E., Ora, J., Webb, K.A., Laveneziana, P., Jensen, D., Mechanisms of activity-related dyspnea in pulmonary diseases. Respir. Physiol. Neurobiol. 167, Ofir, D., Laveneziana, P., Webb, K.A., Lam, Y.M., O Donnell, D.E., Sex differences in the perceived intensity of breathlessness during exercise with advancing age. J. Appl. Physiol. 104, Poon, C.S., Potentiation of exercise ventilatory response by airway CO 2 and dead space loading. J. Appl. Physiol. 73, Qin, Y.Y., Steier, J., Jolley, C., Moxham, J., Zhong, N.S., Luo, Y.M., Efficiency of neural drive during exercise in patients with COPD and healthy subjects. Chest 138, Reilly, C.C., Ward, K., Jolley, C.J., Lunt, A.C., Steier, J., Elston, C., Polkey, M.I., Rafferty, G.F., Moxham, J., Neural respiratory drive, pulmonary mechanics and breathlessness in patients with cystic fibrosis. Thorax 66, Romer, L.M., Lovering, A.T., Haverkamp, H.C., Pegelow, D.F., Dempsey, J.A., Effect of inspiratory muscle work on peripheral fatigue of locomotor muscles in healthy humans. J. Physiol. 571, Sajkov, D., Petrovsky, N., Palange, P., Management of dyspnea in advanced pulmonary arterial hypertension. Curr. Opin. Support Palliat. Care. 4, Sinderby, C., Spahija, J., Beck, J., Kaminski, D., Yan, S., Comtois, N., Sliwinsky, P., Diaphragm activation during exercise in chronic obstructive pulmonary disease. Am. J. Respir. Crit. Care Med. 163, Steier, J., Jolley, C.J., Seymour, J., Roughton, M., Polkey, M.I., Moxham, J., Neural respiratory drive in obesity. Thorax 64, Stubbing, D.G., Pengelly, L.D., Morse, J.L.C., Jones, N.L., Pulmonary mechanics during exercise in normal males. J. Appl. Physiol. 49, Van der Plas, M.N., Reesnik, H.J., Roos, C.M., van Steenwijk, R.P., Kloek, J.J., Bresser, P., Pulmonary endarterectomy improves dyspnea by the relief of dead space ventilation. Ann. Thorac. Surg. 89, Wanger, J., Clausen, J.L., Coates, A., Pedersen, O.F., Brusasco, V., Burgos, F., Casaburi, R., Crapo, R., Enright, P., van der Grinten, C.P., Gustaffson, P., Hankinson, J., Jensen, R., Johnson, D., MacIntyre, N., McKay, R., Miller, M.R., Navajas, D., Pellegrino, R., Viegi, G., Standardisation of the measurement of lung volumes. Eur. Respir. J. 26, Ward, S.A., Whipp, B.J., Ventilatory control during exercise with increased external dead space. J. Appl. Physiol. 48, Wasserman, K., Hansen, J.E., Sue, D.Y., Stringer, W.W., Whipp, B.J., Principles of Exercise Testing and Interpretation, 4th ed. Lippincott Williams & Wilkins, Baltimore, MD, USA. Wetter, T.J., Harms, C.A., Nelson, W.B., Pegelow, D.F., Dempsey, J.A., Influence of respiratory muscle work on VO 2 and leg blood flow during submaximal exercise. J. Appl. Physiol. 87, Wood, H.E., Mitchell, G.S., Babb, T.G., Short-term modulation of the exercise ventilatory response in young men. J. Appl. Physiol. 104, Wood, H.E., Mitchell, G.S., Babb, T.G., Short-term modulation of the exercise ventilatory response in older men. Respir. Physiol. Neurobiol. 173,