Eur J Appl Physiol DOI 10.1007/s00421-014-2956-0
Original Article
Operating lung volumes are affected by exercise mode but not trunk and hip angle during maximal exercise Joseph W. Duke · Jonathon L. Stickford · Joshua C. Weavil · Robert F. Chapman · Joel M. Stager · Timothy D. Mickleborough
Received: 19 November 2013 / Accepted: 12 July 2014 © Springer-Verlag Berlin Heidelberg 2014
Abstract Introduction Despite VO2peak being, generally, greater while running compared to cycling, ventilation (VE) during maximal exercise is less while running compared to cycling. Differences in operating lung volumes (OLV) between maximal running and cycling could be one explanation for previously observed differences in VE and this could be due to differences in body position e.g., trunk/hip angle during exercise. Purpose We asked whether OLV differed between maximal running and cycling and if this difference was due to trunk/hip angle during exercise. Methods Eighteen men performed three graded maximal exercise tests; one while running, one while cycling in the drop position (i.e., extreme hip flexion), and one while cycling upright (i.e., seated with thorax upright). Resting flow-volume characteristics were measured in each body position to be used during exercise. Tidal flow-volume loops were measured throughout the exercise. Results VE during maximal running (148.8 ± 18.9 L min−1) tended to be lower than during cycling in the drop position (158.5 ± 24.7 L min−1; p = 0.07) and in the upright position (158.5 ± 23.7 L min−1; p = 0.06). End-inspiratory and endexpiratory lung volumes (EILV, EELV) were significantly larger during drop cycling compared to running (87.1 ± 4.1 and 35.8 ± 6.2 vs. 83.9 ± 6.0 and 33.0 ± 5.7 % FVC), but only EILV was larger during upright cycling compared to running (88.2 ± 3.5 % FVC). OLV and VE did not differ between cycling positions. Communicated by Guido Ferretti. J. W. Duke (*) · J. L. Stickford · J. C. Weavil · R. F. Chapman · J. M. Stager · T. D. Mickleborough Human Performance Laboratory, Department of Kinesiology, Indiana University, Bloomington, IN 47405, USA e-mail: [email protected]
Conclusion Since OLV are altered by exercise mode, but cycling position did not have a significant impact on OLV, we conclude that trunk/hip angle is likely not the primary factor determining OLV during maximal exercise. Keywords Ventilation · Pulmonary mechanics · Breathing patterns · Elite athletes Abbreviations EELV End-expiratory lung volumes EILV End-inspiratory lung volumes fb Frequency of breathing FEO2 Fraction of expired O2 FECO2− Fraction of expired CO2 FEV1 Forced expired volume in 1 s FEV1/FVC Forced expired volume in 1 s to forced vital capacity ratio FVC Forced vital capacity HR Heart rate IC Inspiratory capacity MEF50 Maximal expiratory flow at 50 % of forced vital capacity OLV Operating lung volumes PEF Peak expired flow rate rpm Revolutions per minute RER Respiratory exchange ratio Ti/TTotal Inspiratory duty cycle as a percentage of total breathing cycle time TLC Total lung capacity TTotal Total breathing cycle time VE Minute ventilation VE/VO2 Ventilatory equivalents for oxygen (O2) VE/VCO2 Ventilatory equivalents for carbon dioxide (CO2) VO2peak Peak oxygen consumption
13
VT Tidal volume W Watts
Introduction Previous studies have demonstrated that the cardiopulmonary responses to exercise differ as a function of exercise mode (Astrand and Saltin 1961; Smith et al. 1994; Gavin and Stager 1999; Elliott and Grace 2010; Tanner et al. 2014). Peak oxygen consumption (VO2peak), in particular, has been shown to be nearly 10 % greater while running compared to cycling (Hermansen et al. 1970; Gavin and Stager 1999; Tanner et al. 2014), which is likely due to the larger muscle mass recruited while running (Astrand and Saltin 1961). In addition, minute ventilation (VE) has been shown to be significantly greater during maximal cycling compared to maximal running (Astrand and Saltin 1961; Smith et al. 1994; Gavin and Stager 1999; Elliott and Grace 2010). The differing VE between maximal running and cycling may be explained by two proposed mechanisms. The first proposed mechanism is that the chemical and/or neural stimuli to breathe may be greater during maximal cycling compared to maximal running (Koyal et al. 1976; Millet et al. 2009). Of note, it has been reported that the ventilatory equivalents for both oxygen (VE/VO2) and carbon dioxide (VE/VCO2) were higher during maximal cycling compared to running, possibly implying an increased chemical (i.e., CO2) drive to breathe (Gavin and Stager 1999; Tanner et al. 2014). The second proposed mechanism, and the objective of this study, focuses on how pulmonary mechanics may influence how VE is achieved (Hopkins et al. 2000; Millet et al. 2009). Due to the mechano-elastic properties of the lung, chest wall, and respiratory muscles responsible for active inspiration and expiration, expiratory airflow rate varies as a function of the lung volumes at which ventilation takes place (Sharratt et al. 1987; Klas and Dempsey 1989; Babb et al. 1991; McClaran et al. 1999), referred to as exercise operating lung volumes (OLV). OLV are examined using end-inspiratory and end-expiratory lung volumes (EILV and EELV, respectively) and a number of studies have examined OLV in healthy individuals during running (Henke et al. 1988; Johnson et al. 1990, 1991a, 1992) or cycling (Stubbing et al. 1980; Younes and Kivinen 1984; Sharratt et al. 1987; Henke et al. 1988; Klas and Dempsey 1989; Babb et al. 1991; McClaran et al. 1999). Only recently, however, EILV and EELV were compared in individuals asked to perform both exercise modalities (Tanner et al. 2014). Although we reported significantly larger EILV and EELV (i.e., closer to total lung capacity; TLC) during maximal cycling compared to maximal treadmill running, we failed to observe a significantly larger VE
13
Eur J Appl Physiol
while cycling. The effect of body position on OLV was not directly tested. The body positions used while running on a treadmill, as well as cycling on an ergometer may influence OLV. While running, an individual is in an erect posture with the thorax upright and roughly perpendicular to the treadmill. However, while cycling, individuals are seated on the ergometer with at least a slight flexion of the hip (angle less than 90°) and forward lean of the thorax. The effect of body position on lung volumes at rest has been extensively studied (see Agostoni and Hyatt 1986 for a review on this topic). In young, healthy individuals functional residual capacity (i.e., EELV at rest) is ~50 % of TLC while standing, but increases to ~57 % of TLC while seated and leaning forward with arms rested at roughly sternum height (Agostoni and Hyatt 1986). This implies that the position of the thoracic cavity (i.e., hip flexion), even at rest, has an impact on OLV and our previously published differences in OLV at rest while standing compared to seated on an ergometer support this supposition (Tanner et al. 2014). Furthermore, only a few studies have investigated the impact of body position (e.g., trunk/hip angle) during cycling on various cardiopulmonary parameters (Faria et al. 1978; Origenes et al. 1993; Berry et al. 1994). These studies compared various performance and metabolic measures between cycling positions (VO2peak, power output, etc.) and found only VE during maximal exercise to differ between cycling positions (drop > upright; Faria et al. 1978), but did not quantify OLV (Faria et al. 1978; Origenes et al. 1993; Berry et al. 1994). Therefore, the primary purpose of this investigation was to determine if OLV differed during submaximal and maximal exercise between running on a treadmill and cycling on an ergometer within the same, well-trained, individual. We also sought to determine if the observed difference on OLV between submaximal and maximal running and cycling was due to the trunk/hip angle by directly comparing OLV in drop and upright cycling. Because of the difference in trunk/hip angle and its effects on respiratory mechanics, we hypothesized that OLV would be significantly larger (i.e., EILV and EELV will be closer to TLC) during submaximal and maximal cycling in the drop position compared to running. Furthermore, we hypothesized that OLV in upright cycling would not differ from running owing to the similar trunk/hip angle.
Methods Subjects Thirty-five college-aged males volunteered to participate in the study after being advised both verbally and in writing
Eur J Appl Physiol
as to the nature of the experiments and providing written informed consent. These documents and procedures were approved by the Indiana University review board, which governs human research, and in according with the Declaration of Helsinki. Each subject was required to meet a set of inclusion criteria: a history of extensive, high-level endurance training (as determined by a questionnaire about training history), and no indication of pulmonary disease or dysfunction. Fifteen subjects did not meet the aerobic fitness criteria (VO2peak ≤ 60 mL kg−1 min−1 while cycling or ≤65 mL kg−1 min−1 while running), and two subjects voluntarily withdrew from the study prior to completion. Of the initial thirty-five screened individuals, eighteen met all of the screening criteria and were invited to participate further. Eight indicated they were primarily trained by running, six indicated cycling was their predominant training mode, and the remaining four considered themselves to be triathletes (i.e., equally trained in both running and cycling). The individuals that qualified and completed the entire study were aged 21 ± 2 years, weighed 71.3 ± 7.6 kg, and stood 180.5 ± 6.4 cm tall. All subjects were well-trained endurance athletes as indicated by their self-reported assessment of current training and fitness level. Experimental sequence Each subject performed a total of three graded exercise tests (i.e., two additional tests after the initial screening test) to volitional exhaustion; once on a motor driven treadmill and twice on an electronically braked cycle ergometer with each visit separated by at least 24 h and a maximum of 2 weeks. Following the initial screening test, the remaining two maximal exercise tests were done in a random order. The initial screening involved completing a maximal graded exercise test on either a motor driven treadmill or a cycle ergometer, depending upon which mode the subject considered himself more trained. Individuals who indicated they were more trained on a bicycle performed their initial test on the cycle ergometer in the ‘drop position’. This position was chosen for the initial testing because it reflected a cycling position to which well-trained cyclists are more accustomed compared to the upright position used in this study. In the drop position, the subject’s hands rested on the lowermost portion of the standard cycling racing handlebars with a great deal of hip flexion, such that the thorax was nearly parallel with the top tube of the cycle ergometer (i.e., in the horizontal plane). In addition, this trunk/ hip angle is most similar to that used during exercise in the previous studies from our laboratory comparing maximal running and maximal cycling (Gavin and Stager 1999; Tanner et al. 2014). The remaining cycling test was performed in the ‘upright’ position. This position required subjects to
rest their hands at sternum height on a polyvinylchloride pipe that was constructed to go over the cycle ergometer. The polyvinylchloride pipe was constructed such that it could be moved away from or towards the subject to ensure they had a comfortable reach to the pipe and that there was minimal forward lean (i.e., hip flexion). Resting their hands on this pipe kept the hips minimally flexed and the thoracic cavity vertical and near-perpendicular to the top tube of the cycle ergometer. The contrast in cycling positions was intentionally chosen to best test the impact of trunk/ hip angle on ventilatory parameters during maximal cycle ergometry. Maximal treadmill running was performed while in an erect posture with no or little hip flexion and the thoracic cavity nearly perpendicular to the treadmill belt.
Experimental procedures Prior to each test, three to five forced vital capacity (FVC) manoeuvres were performed in the body position in which exercise would take place. From each FVC manoeuvre, forced expiratory volume in 1 s (FEV1), peak expiratory flow rate (PEF), and maximal expiratory flow at 50 % of vital capacity (MEF50) were calculated. Tests were done in accordance with the standards set by the American Thoracic Society for repeatability (Miller et al. 2005). Flow and volume values were corrected to body temperature, pressure, saturated, and the largest FVC, FEV1, PEF, and MEF50 selected. Of the three repeatable FVC manoeuvres performed, the manoeuvre that produced the largest FVC and FEV1 was selected as a representation of the subject’s pulmonary function. Treadmill test After collecting resting metabolic data for 5 min, the speed of the treadmill (model 18–60, Quinton, Bothell, WA, USA) was increased to one that would allow the test to last approximately 10–15 min. The selected speed was based on the individual subject’s training history and treadmill running experience. Speeds ranged from 9.2 to 14.5 km h−1 (5.3–9.0 mi h−1) and remained constant throughout the duration of the test. Subjects began the test by running on a flat treadmill (0 % grade). After 2 min, the treadmill grade was increased to 4 %, and the grade continued to be raised 2 % every 2 min thereafter until volitional fatigue (Balke and Ware 1959). Cycle ergometry tests Similar to the treadmill test, the maximal cycling protocol began with 5 min of seated resting measurements.
13
Following rest, the subject began cycling on an electronically braked cycle ergometer (Velotron, Elite Model, Racer Mate, Seattle, WA, USA) at a workload of 100 W. The workload was increased by 25 W every minute thereafter until the subject could no longer continue or cadence decreased below ~60 rpm. Both cycling positions followed identical exercise protocols. During the maximal exercise, test subjects were verbally encouraged to exercise as long as possible. VO2peak was assessed using the following criteria: (1) a heart rate ≥90 % of the age-predicted maximal heart rate (220 − age), (2) a respiratory exchange ratio (RER) ≥1.10, and (3) identification of a plateau (≤150 ml) in VO2 with an increase in workload. If two of the three criteria were met, the highest 1 min average VO2 was chosen as the subject’s VO2peak. During the maximal exercise tests, heart rate (HR) was measured using a telemetry transmitter affixed to the subject’s chest (Polar Electro Inc., Lake Success, NY, USA) and recorded at the end of every minute. Ventilatory and metabolic variables were continuously measured during rest and exercise via open-circuit calorimetry. Subjects breathed through a low resistance, two-way non-rebreathing valve (model 2700, Hans Rudolph, Shawnee, KS, USA), from which expired gases entered a 5 L mixing chamber. Fractional concentrations of O2 and CO2 (FEO2 and FECO2, respectively) were sampled from the mixing chamber at a constant rate (300 mL × min−1) with O2 quantified using an Applied Electrochemistry S-3A (Ametek, Thermox Instruments, Pittsburgh, PA, USA) oxygen analyzer and CO2 with a CD-3A carbon dioxide analyzer (Ametek, Thermox Instruments, Pittsburgh, PA, USA). Analyzers were calibrated immediately pre-test with a gas of a known concentration in the physiological range and checked/corrected for any drift immediately following the test’s completion. A pneumotachograph (series 3,813 Hans Rudolph, Shawnee, KS, USA) placed on the inspired side was used to measure inspired airflow, which was integrated and converted to yield VE. Ventilatory and metabolic variables were averaged over each minute of exercise. The above variables, as well as, FEO2 and FECO2, frequency of breathing (fb), and tidal volume (VT), were continuously measured and monitored with a data acquisition software (DASYLab, Measurement Computing, Norton, MA, USA) sampling at 50 Hz. Flow‑volume relationships Flow-volume loops were collected during all progressive exercise tests. Maximal flow-volume loops were collected pre- and post-exercise and obtained in triplicate by having the subject perform FVC manoeuvres, again in accordance with ATS standards (Miller et al. 2005). The largest maximal flow-volume loop, regardless of whether it was
13
Eur J Appl Physiol
obtained pre- or post-exercise, was chosen for further analysis using the same criteria outlined with respect to pulmonary function above. Of the 54 maximal flow-volume loops (18 subjects × 3 trials = 54) obtained, 9 were constructed from post-exercise manoeuvres and 45 from pre-exercise manoeuvres. To measure FVC, subjects were told to expire to residual volume, maximally inspire to TLC and maximally expire to residual volume. Inspired and expired flow rates were measured using pneumotachographs on both the inspired and expired sides. The pneumotachograph on the expired side was heated to allow for calculation of body, temperature, pressure, saturated values despite changes in expired gas temperature during exercise. Exercise flowvolume data were collected over the last 30 s of each stage of exercise using a technique described elsewhere (Derchak et al. 2000; Tanner et al. 2014). Briefly, tidal breath flowvolume loops were averaged over approximately 12–15 breaths during each minute. Average tidal breath flowvolume loops were placed into the proper position on the volume axis within the maximal flow-volume loop using an in-house data analysis program specifically designed for this purpose (Clipper 5.2, Computer Associates, Islandia, NY, USA). Inspiratory capacity manoeuvres were performed at 30 and 55 s of each stage and were used to estimate average inspiratory reserve volume during this period as previously described (Babb 1997). EELV and EILV were mathematically determined by, first, subtracting inspiratory capacity volume from FVC (EELV = FVC−IC) and adding VT to EELV (EILV = EELV + VT). Prior to the initial exercise test, the procedure and importance of the inspiratory capacity manoeuvre were described in detail to each subject. Correct performance was demonstrated and subjects performed several unmeasured practice manoeuvres. During exercise, strong encouragement was provided immediately prior to and during each manoeuvre during exercise. Our data acquisition software allowed visualization of real-time measurement so if a manoeuvre appeared inadequate; the subject was prompted to perform another. The in-house data analysis program also provided measures of total breathing cycle time, from which the inspiratory duty cycle as a percentage of total breathing cycle time was calculated (Ti/TTotal). Expiratory flow limitation was considered to be present when the tidal flow-volume loop met or exceeded the maximal flow-volume loop and the extent of expiratory flow limitation was calculated as a percentage of the expired tidal volume that met or exceeded the maximal flow-volume loop (Johnson et al. 1995). Statistical analysis To address the primary purpose of the study (maximal exercise), all variables (OLV, VE, VO2peak, VCO2, VT, and fb) were analyzed using a priori planned comparisons
Eur J Appl Physiol Table 1 Resting pulmonary function data Running
Drop cycle
Upright cycle
FVC (L) FEV1 (L) FEV1/FVC (%) PEF (L sec−1)
5.22 ± 0.52 (90.6 ± 6.7) 4.61 ± 0.49 (96.4 ± 7.3) 88.5 ± 4.3 (105.6 ± 4.8) 10.05 ± 1.26
5.27 ± 0.56 (91.4 ± 6.8) 4.66 ± 0.50 (97.3 ± 6.9) 88.5 ± 4.0 (105.6 ± 4.5) 10.19 ± 1.58
5.28 ± 0.55 (91.6 ± 7.3) 4.60 ± 0.53 (96.2 ± 8.2) 87.3 ± 4.5 (104.2 ± 5.1) 10.46 ± 1.44
MEF50 (L sec−1)
4.98 ± 1.26
5.32 ± 1.26
5.50 ± 1.89
No variables were significantly different between trials. Data are displayed as mean ± SD. Values in parentheses are mean ± SD percent predicted for each pulmonary function parameter FVC forced vital capacity, FEV1 forced expiratory volume in 1 s, MEF50 maximal expiratory flow at 50 % of vital capacity
between (1) running and drop cycling, (2) drop cycling and upright cycling, and (3) running and upright cycling. For these comparisons, the familywise error rate (αfam) was set to equal 0.05 and was adjusted to control Type I error (αcomp) using the Bonferroni adjustment, such that αcomp = αfam/c, where c is the number of comparisons to be made. We chose to analyze the maximal exercise data in this manner because we wanted to utilize our statistical power for the pairwise comparisons that would allow us to answer our primary research question rather than test for a significant overall test first. To address the secondary, more exploratory, purpose (submaximal exercise) we utilized a more conservative statistical approach and multiple one-way repeated measures ANOVAs were computed (i.e., (1) between running, drop cycling, and upright cycling at 70 % of VO2peak and (2) between running, drop cycling, and upright cycling at 85 % of VO2peak). When a significant omnibus test was observed a Tukey posttest was computed to determine where pairwise differences existed. This allowed us to determine if differences observed during maximal exercise were apparent at lower intensity exercise or did not appear until maximal exercise. The prevalence of expiratory flow limitation was tested using a two-sample z test (because proportions follow a Normal distribution) for (1) running and drop cycling, (2) drop cycling and upright cycling, and (3) running and upright cycling. The Bonferroni adjustment was made as described above. Statistical analyses were performed using PASW version 19.0 for Windows (IBM, Chicago, IL, USA).
Results Pulmonary function Absolute and percent predicted values (mean ± SD) for pulmonary function in the respective exercise mode are displayed in Table 1. All values were within the normal ranges for healthy men aged 18–35 years (Hankinson et al. 1999).
No differences in pulmonary function existed between exercise modes or cycling positions (p > 0.05). Operating lung volumes during exercise EILV and EELV expressed as % of FVC are displayed in Fig. 1. Tidal flow-volume loops within a maximal flowvolume loop during maximal exercise in each trial are displayed in Fig. 2. EILV and EELV were significantly larger during maximal cycling in the drop position compared to maximal running (p 0.05). EILV was significantly larger during maximal cycling in the upright position compared to maximal running (p 0.05). EILV was significantly greater during submaximal drop and upright cycling compared to while running. Submaximal and maximal tidal flow-volume loops for each exercise trial for a single subject are displayed in Fig. 3. The prevalence of extent of expiratory flow limitation during maximal exercise did not differ between exercise trials (p > 0.05; Table 2). Submaximal and maximal exercise response data Metabolic and ventilatory values for the three exercise tests are presented in Table 2 and Fig. 4. VE during maximal exercise did not differ between maximal running and maximal cycling in either position (drop; p = 0.07 and upright; p = 0.06). VE during submaximal exercise was significantly greater during upright cycling compared to drop cycling at 70 % of VO2peak only. No other differences existed on VE between exercise trials. VT did not differ between maximal running and drop cycling (p > 0.05), but was significantly larger during upright cycling compared to running and drop cycling (p 0.05), but did differ between cycling positions (p
Original Article
Operating lung volumes are affected by exercise mode but not trunk and hip angle during maximal exercise Joseph W. Duke · Jonathon L. Stickford · Joshua C. Weavil · Robert F. Chapman · Joel M. Stager · Timothy D. Mickleborough
Received: 19 November 2013 / Accepted: 12 July 2014 © Springer-Verlag Berlin Heidelberg 2014
Abstract Introduction Despite VO2peak being, generally, greater while running compared to cycling, ventilation (VE) during maximal exercise is less while running compared to cycling. Differences in operating lung volumes (OLV) between maximal running and cycling could be one explanation for previously observed differences in VE and this could be due to differences in body position e.g., trunk/hip angle during exercise. Purpose We asked whether OLV differed between maximal running and cycling and if this difference was due to trunk/hip angle during exercise. Methods Eighteen men performed three graded maximal exercise tests; one while running, one while cycling in the drop position (i.e., extreme hip flexion), and one while cycling upright (i.e., seated with thorax upright). Resting flow-volume characteristics were measured in each body position to be used during exercise. Tidal flow-volume loops were measured throughout the exercise. Results VE during maximal running (148.8 ± 18.9 L min−1) tended to be lower than during cycling in the drop position (158.5 ± 24.7 L min−1; p = 0.07) and in the upright position (158.5 ± 23.7 L min−1; p = 0.06). End-inspiratory and endexpiratory lung volumes (EILV, EELV) were significantly larger during drop cycling compared to running (87.1 ± 4.1 and 35.8 ± 6.2 vs. 83.9 ± 6.0 and 33.0 ± 5.7 % FVC), but only EILV was larger during upright cycling compared to running (88.2 ± 3.5 % FVC). OLV and VE did not differ between cycling positions. Communicated by Guido Ferretti. J. W. Duke (*) · J. L. Stickford · J. C. Weavil · R. F. Chapman · J. M. Stager · T. D. Mickleborough Human Performance Laboratory, Department of Kinesiology, Indiana University, Bloomington, IN 47405, USA e-mail: [email protected]
Conclusion Since OLV are altered by exercise mode, but cycling position did not have a significant impact on OLV, we conclude that trunk/hip angle is likely not the primary factor determining OLV during maximal exercise. Keywords Ventilation · Pulmonary mechanics · Breathing patterns · Elite athletes Abbreviations EELV End-expiratory lung volumes EILV End-inspiratory lung volumes fb Frequency of breathing FEO2 Fraction of expired O2 FECO2− Fraction of expired CO2 FEV1 Forced expired volume in 1 s FEV1/FVC Forced expired volume in 1 s to forced vital capacity ratio FVC Forced vital capacity HR Heart rate IC Inspiratory capacity MEF50 Maximal expiratory flow at 50 % of forced vital capacity OLV Operating lung volumes PEF Peak expired flow rate rpm Revolutions per minute RER Respiratory exchange ratio Ti/TTotal Inspiratory duty cycle as a percentage of total breathing cycle time TLC Total lung capacity TTotal Total breathing cycle time VE Minute ventilation VE/VO2 Ventilatory equivalents for oxygen (O2) VE/VCO2 Ventilatory equivalents for carbon dioxide (CO2) VO2peak Peak oxygen consumption
13
VT Tidal volume W Watts
Introduction Previous studies have demonstrated that the cardiopulmonary responses to exercise differ as a function of exercise mode (Astrand and Saltin 1961; Smith et al. 1994; Gavin and Stager 1999; Elliott and Grace 2010; Tanner et al. 2014). Peak oxygen consumption (VO2peak), in particular, has been shown to be nearly 10 % greater while running compared to cycling (Hermansen et al. 1970; Gavin and Stager 1999; Tanner et al. 2014), which is likely due to the larger muscle mass recruited while running (Astrand and Saltin 1961). In addition, minute ventilation (VE) has been shown to be significantly greater during maximal cycling compared to maximal running (Astrand and Saltin 1961; Smith et al. 1994; Gavin and Stager 1999; Elliott and Grace 2010). The differing VE between maximal running and cycling may be explained by two proposed mechanisms. The first proposed mechanism is that the chemical and/or neural stimuli to breathe may be greater during maximal cycling compared to maximal running (Koyal et al. 1976; Millet et al. 2009). Of note, it has been reported that the ventilatory equivalents for both oxygen (VE/VO2) and carbon dioxide (VE/VCO2) were higher during maximal cycling compared to running, possibly implying an increased chemical (i.e., CO2) drive to breathe (Gavin and Stager 1999; Tanner et al. 2014). The second proposed mechanism, and the objective of this study, focuses on how pulmonary mechanics may influence how VE is achieved (Hopkins et al. 2000; Millet et al. 2009). Due to the mechano-elastic properties of the lung, chest wall, and respiratory muscles responsible for active inspiration and expiration, expiratory airflow rate varies as a function of the lung volumes at which ventilation takes place (Sharratt et al. 1987; Klas and Dempsey 1989; Babb et al. 1991; McClaran et al. 1999), referred to as exercise operating lung volumes (OLV). OLV are examined using end-inspiratory and end-expiratory lung volumes (EILV and EELV, respectively) and a number of studies have examined OLV in healthy individuals during running (Henke et al. 1988; Johnson et al. 1990, 1991a, 1992) or cycling (Stubbing et al. 1980; Younes and Kivinen 1984; Sharratt et al. 1987; Henke et al. 1988; Klas and Dempsey 1989; Babb et al. 1991; McClaran et al. 1999). Only recently, however, EILV and EELV were compared in individuals asked to perform both exercise modalities (Tanner et al. 2014). Although we reported significantly larger EILV and EELV (i.e., closer to total lung capacity; TLC) during maximal cycling compared to maximal treadmill running, we failed to observe a significantly larger VE
13
Eur J Appl Physiol
while cycling. The effect of body position on OLV was not directly tested. The body positions used while running on a treadmill, as well as cycling on an ergometer may influence OLV. While running, an individual is in an erect posture with the thorax upright and roughly perpendicular to the treadmill. However, while cycling, individuals are seated on the ergometer with at least a slight flexion of the hip (angle less than 90°) and forward lean of the thorax. The effect of body position on lung volumes at rest has been extensively studied (see Agostoni and Hyatt 1986 for a review on this topic). In young, healthy individuals functional residual capacity (i.e., EELV at rest) is ~50 % of TLC while standing, but increases to ~57 % of TLC while seated and leaning forward with arms rested at roughly sternum height (Agostoni and Hyatt 1986). This implies that the position of the thoracic cavity (i.e., hip flexion), even at rest, has an impact on OLV and our previously published differences in OLV at rest while standing compared to seated on an ergometer support this supposition (Tanner et al. 2014). Furthermore, only a few studies have investigated the impact of body position (e.g., trunk/hip angle) during cycling on various cardiopulmonary parameters (Faria et al. 1978; Origenes et al. 1993; Berry et al. 1994). These studies compared various performance and metabolic measures between cycling positions (VO2peak, power output, etc.) and found only VE during maximal exercise to differ between cycling positions (drop > upright; Faria et al. 1978), but did not quantify OLV (Faria et al. 1978; Origenes et al. 1993; Berry et al. 1994). Therefore, the primary purpose of this investigation was to determine if OLV differed during submaximal and maximal exercise between running on a treadmill and cycling on an ergometer within the same, well-trained, individual. We also sought to determine if the observed difference on OLV between submaximal and maximal running and cycling was due to the trunk/hip angle by directly comparing OLV in drop and upright cycling. Because of the difference in trunk/hip angle and its effects on respiratory mechanics, we hypothesized that OLV would be significantly larger (i.e., EILV and EELV will be closer to TLC) during submaximal and maximal cycling in the drop position compared to running. Furthermore, we hypothesized that OLV in upright cycling would not differ from running owing to the similar trunk/hip angle.
Methods Subjects Thirty-five college-aged males volunteered to participate in the study after being advised both verbally and in writing
Eur J Appl Physiol
as to the nature of the experiments and providing written informed consent. These documents and procedures were approved by the Indiana University review board, which governs human research, and in according with the Declaration of Helsinki. Each subject was required to meet a set of inclusion criteria: a history of extensive, high-level endurance training (as determined by a questionnaire about training history), and no indication of pulmonary disease or dysfunction. Fifteen subjects did not meet the aerobic fitness criteria (VO2peak ≤ 60 mL kg−1 min−1 while cycling or ≤65 mL kg−1 min−1 while running), and two subjects voluntarily withdrew from the study prior to completion. Of the initial thirty-five screened individuals, eighteen met all of the screening criteria and were invited to participate further. Eight indicated they were primarily trained by running, six indicated cycling was their predominant training mode, and the remaining four considered themselves to be triathletes (i.e., equally trained in both running and cycling). The individuals that qualified and completed the entire study were aged 21 ± 2 years, weighed 71.3 ± 7.6 kg, and stood 180.5 ± 6.4 cm tall. All subjects were well-trained endurance athletes as indicated by their self-reported assessment of current training and fitness level. Experimental sequence Each subject performed a total of three graded exercise tests (i.e., two additional tests after the initial screening test) to volitional exhaustion; once on a motor driven treadmill and twice on an electronically braked cycle ergometer with each visit separated by at least 24 h and a maximum of 2 weeks. Following the initial screening test, the remaining two maximal exercise tests were done in a random order. The initial screening involved completing a maximal graded exercise test on either a motor driven treadmill or a cycle ergometer, depending upon which mode the subject considered himself more trained. Individuals who indicated they were more trained on a bicycle performed their initial test on the cycle ergometer in the ‘drop position’. This position was chosen for the initial testing because it reflected a cycling position to which well-trained cyclists are more accustomed compared to the upright position used in this study. In the drop position, the subject’s hands rested on the lowermost portion of the standard cycling racing handlebars with a great deal of hip flexion, such that the thorax was nearly parallel with the top tube of the cycle ergometer (i.e., in the horizontal plane). In addition, this trunk/ hip angle is most similar to that used during exercise in the previous studies from our laboratory comparing maximal running and maximal cycling (Gavin and Stager 1999; Tanner et al. 2014). The remaining cycling test was performed in the ‘upright’ position. This position required subjects to
rest their hands at sternum height on a polyvinylchloride pipe that was constructed to go over the cycle ergometer. The polyvinylchloride pipe was constructed such that it could be moved away from or towards the subject to ensure they had a comfortable reach to the pipe and that there was minimal forward lean (i.e., hip flexion). Resting their hands on this pipe kept the hips minimally flexed and the thoracic cavity vertical and near-perpendicular to the top tube of the cycle ergometer. The contrast in cycling positions was intentionally chosen to best test the impact of trunk/ hip angle on ventilatory parameters during maximal cycle ergometry. Maximal treadmill running was performed while in an erect posture with no or little hip flexion and the thoracic cavity nearly perpendicular to the treadmill belt.
Experimental procedures Prior to each test, three to five forced vital capacity (FVC) manoeuvres were performed in the body position in which exercise would take place. From each FVC manoeuvre, forced expiratory volume in 1 s (FEV1), peak expiratory flow rate (PEF), and maximal expiratory flow at 50 % of vital capacity (MEF50) were calculated. Tests were done in accordance with the standards set by the American Thoracic Society for repeatability (Miller et al. 2005). Flow and volume values were corrected to body temperature, pressure, saturated, and the largest FVC, FEV1, PEF, and MEF50 selected. Of the three repeatable FVC manoeuvres performed, the manoeuvre that produced the largest FVC and FEV1 was selected as a representation of the subject’s pulmonary function. Treadmill test After collecting resting metabolic data for 5 min, the speed of the treadmill (model 18–60, Quinton, Bothell, WA, USA) was increased to one that would allow the test to last approximately 10–15 min. The selected speed was based on the individual subject’s training history and treadmill running experience. Speeds ranged from 9.2 to 14.5 km h−1 (5.3–9.0 mi h−1) and remained constant throughout the duration of the test. Subjects began the test by running on a flat treadmill (0 % grade). After 2 min, the treadmill grade was increased to 4 %, and the grade continued to be raised 2 % every 2 min thereafter until volitional fatigue (Balke and Ware 1959). Cycle ergometry tests Similar to the treadmill test, the maximal cycling protocol began with 5 min of seated resting measurements.
13
Following rest, the subject began cycling on an electronically braked cycle ergometer (Velotron, Elite Model, Racer Mate, Seattle, WA, USA) at a workload of 100 W. The workload was increased by 25 W every minute thereafter until the subject could no longer continue or cadence decreased below ~60 rpm. Both cycling positions followed identical exercise protocols. During the maximal exercise, test subjects were verbally encouraged to exercise as long as possible. VO2peak was assessed using the following criteria: (1) a heart rate ≥90 % of the age-predicted maximal heart rate (220 − age), (2) a respiratory exchange ratio (RER) ≥1.10, and (3) identification of a plateau (≤150 ml) in VO2 with an increase in workload. If two of the three criteria were met, the highest 1 min average VO2 was chosen as the subject’s VO2peak. During the maximal exercise tests, heart rate (HR) was measured using a telemetry transmitter affixed to the subject’s chest (Polar Electro Inc., Lake Success, NY, USA) and recorded at the end of every minute. Ventilatory and metabolic variables were continuously measured during rest and exercise via open-circuit calorimetry. Subjects breathed through a low resistance, two-way non-rebreathing valve (model 2700, Hans Rudolph, Shawnee, KS, USA), from which expired gases entered a 5 L mixing chamber. Fractional concentrations of O2 and CO2 (FEO2 and FECO2, respectively) were sampled from the mixing chamber at a constant rate (300 mL × min−1) with O2 quantified using an Applied Electrochemistry S-3A (Ametek, Thermox Instruments, Pittsburgh, PA, USA) oxygen analyzer and CO2 with a CD-3A carbon dioxide analyzer (Ametek, Thermox Instruments, Pittsburgh, PA, USA). Analyzers were calibrated immediately pre-test with a gas of a known concentration in the physiological range and checked/corrected for any drift immediately following the test’s completion. A pneumotachograph (series 3,813 Hans Rudolph, Shawnee, KS, USA) placed on the inspired side was used to measure inspired airflow, which was integrated and converted to yield VE. Ventilatory and metabolic variables were averaged over each minute of exercise. The above variables, as well as, FEO2 and FECO2, frequency of breathing (fb), and tidal volume (VT), were continuously measured and monitored with a data acquisition software (DASYLab, Measurement Computing, Norton, MA, USA) sampling at 50 Hz. Flow‑volume relationships Flow-volume loops were collected during all progressive exercise tests. Maximal flow-volume loops were collected pre- and post-exercise and obtained in triplicate by having the subject perform FVC manoeuvres, again in accordance with ATS standards (Miller et al. 2005). The largest maximal flow-volume loop, regardless of whether it was
13
Eur J Appl Physiol
obtained pre- or post-exercise, was chosen for further analysis using the same criteria outlined with respect to pulmonary function above. Of the 54 maximal flow-volume loops (18 subjects × 3 trials = 54) obtained, 9 were constructed from post-exercise manoeuvres and 45 from pre-exercise manoeuvres. To measure FVC, subjects were told to expire to residual volume, maximally inspire to TLC and maximally expire to residual volume. Inspired and expired flow rates were measured using pneumotachographs on both the inspired and expired sides. The pneumotachograph on the expired side was heated to allow for calculation of body, temperature, pressure, saturated values despite changes in expired gas temperature during exercise. Exercise flowvolume data were collected over the last 30 s of each stage of exercise using a technique described elsewhere (Derchak et al. 2000; Tanner et al. 2014). Briefly, tidal breath flowvolume loops were averaged over approximately 12–15 breaths during each minute. Average tidal breath flowvolume loops were placed into the proper position on the volume axis within the maximal flow-volume loop using an in-house data analysis program specifically designed for this purpose (Clipper 5.2, Computer Associates, Islandia, NY, USA). Inspiratory capacity manoeuvres were performed at 30 and 55 s of each stage and were used to estimate average inspiratory reserve volume during this period as previously described (Babb 1997). EELV and EILV were mathematically determined by, first, subtracting inspiratory capacity volume from FVC (EELV = FVC−IC) and adding VT to EELV (EILV = EELV + VT). Prior to the initial exercise test, the procedure and importance of the inspiratory capacity manoeuvre were described in detail to each subject. Correct performance was demonstrated and subjects performed several unmeasured practice manoeuvres. During exercise, strong encouragement was provided immediately prior to and during each manoeuvre during exercise. Our data acquisition software allowed visualization of real-time measurement so if a manoeuvre appeared inadequate; the subject was prompted to perform another. The in-house data analysis program also provided measures of total breathing cycle time, from which the inspiratory duty cycle as a percentage of total breathing cycle time was calculated (Ti/TTotal). Expiratory flow limitation was considered to be present when the tidal flow-volume loop met or exceeded the maximal flow-volume loop and the extent of expiratory flow limitation was calculated as a percentage of the expired tidal volume that met or exceeded the maximal flow-volume loop (Johnson et al. 1995). Statistical analysis To address the primary purpose of the study (maximal exercise), all variables (OLV, VE, VO2peak, VCO2, VT, and fb) were analyzed using a priori planned comparisons
Eur J Appl Physiol Table 1 Resting pulmonary function data Running
Drop cycle
Upright cycle
FVC (L) FEV1 (L) FEV1/FVC (%) PEF (L sec−1)
5.22 ± 0.52 (90.6 ± 6.7) 4.61 ± 0.49 (96.4 ± 7.3) 88.5 ± 4.3 (105.6 ± 4.8) 10.05 ± 1.26
5.27 ± 0.56 (91.4 ± 6.8) 4.66 ± 0.50 (97.3 ± 6.9) 88.5 ± 4.0 (105.6 ± 4.5) 10.19 ± 1.58
5.28 ± 0.55 (91.6 ± 7.3) 4.60 ± 0.53 (96.2 ± 8.2) 87.3 ± 4.5 (104.2 ± 5.1) 10.46 ± 1.44
MEF50 (L sec−1)
4.98 ± 1.26
5.32 ± 1.26
5.50 ± 1.89
No variables were significantly different between trials. Data are displayed as mean ± SD. Values in parentheses are mean ± SD percent predicted for each pulmonary function parameter FVC forced vital capacity, FEV1 forced expiratory volume in 1 s, MEF50 maximal expiratory flow at 50 % of vital capacity
between (1) running and drop cycling, (2) drop cycling and upright cycling, and (3) running and upright cycling. For these comparisons, the familywise error rate (αfam) was set to equal 0.05 and was adjusted to control Type I error (αcomp) using the Bonferroni adjustment, such that αcomp = αfam/c, where c is the number of comparisons to be made. We chose to analyze the maximal exercise data in this manner because we wanted to utilize our statistical power for the pairwise comparisons that would allow us to answer our primary research question rather than test for a significant overall test first. To address the secondary, more exploratory, purpose (submaximal exercise) we utilized a more conservative statistical approach and multiple one-way repeated measures ANOVAs were computed (i.e., (1) between running, drop cycling, and upright cycling at 70 % of VO2peak and (2) between running, drop cycling, and upright cycling at 85 % of VO2peak). When a significant omnibus test was observed a Tukey posttest was computed to determine where pairwise differences existed. This allowed us to determine if differences observed during maximal exercise were apparent at lower intensity exercise or did not appear until maximal exercise. The prevalence of expiratory flow limitation was tested using a two-sample z test (because proportions follow a Normal distribution) for (1) running and drop cycling, (2) drop cycling and upright cycling, and (3) running and upright cycling. The Bonferroni adjustment was made as described above. Statistical analyses were performed using PASW version 19.0 for Windows (IBM, Chicago, IL, USA).
Results Pulmonary function Absolute and percent predicted values (mean ± SD) for pulmonary function in the respective exercise mode are displayed in Table 1. All values were within the normal ranges for healthy men aged 18–35 years (Hankinson et al. 1999).
No differences in pulmonary function existed between exercise modes or cycling positions (p > 0.05). Operating lung volumes during exercise EILV and EELV expressed as % of FVC are displayed in Fig. 1. Tidal flow-volume loops within a maximal flowvolume loop during maximal exercise in each trial are displayed in Fig. 2. EILV and EELV were significantly larger during maximal cycling in the drop position compared to maximal running (p 0.05). EILV was significantly larger during maximal cycling in the upright position compared to maximal running (p 0.05). EILV was significantly greater during submaximal drop and upright cycling compared to while running. Submaximal and maximal tidal flow-volume loops for each exercise trial for a single subject are displayed in Fig. 3. The prevalence of extent of expiratory flow limitation during maximal exercise did not differ between exercise trials (p > 0.05; Table 2). Submaximal and maximal exercise response data Metabolic and ventilatory values for the three exercise tests are presented in Table 2 and Fig. 4. VE during maximal exercise did not differ between maximal running and maximal cycling in either position (drop; p = 0.07 and upright; p = 0.06). VE during submaximal exercise was significantly greater during upright cycling compared to drop cycling at 70 % of VO2peak only. No other differences existed on VE between exercise trials. VT did not differ between maximal running and drop cycling (p > 0.05), but was significantly larger during upright cycling compared to running and drop cycling (p 0.05), but did differ between cycling positions (p
