Inter-individual differences in breathing pattern at high levels of incremental cycling exercise in healthy subjects

Respiratory Physiology & Neurobiology 189 (2013) 59–66

Contents lists available at ScienceDirect

Respiratory Physiology & Neurobiology journal homepage: www.elsevier.com/locate/resphysiol

Inter-individual differences in breathing pattern at high levels of incremental cycling exercise in healthy subjects Gilles Gravier, Stephane Delliaux, Stephane Delpierre, Regis Guieu, Yves Jammes ∗ UMR MD2 “Dysoxia/Hyperactivity”, Faculty of Medicine, Aix-Marseille University, Marseille, France

a r t i c l e

i n f o

Article history: Accepted 28 June 2013 Keywords: Maximal exercise, Break points, Breathing pattern

a b s t r a c t Interindividual differences in the rate of changes in tidal volume (VT ) and respiratory frequency (fR ) were examined during a maximal incremental cycling exercise. The gain of the inspiratory off-switch reflex was inferred from the VT vs. inspiratory duration (Ti ) relationship. Some subjects also executed a static handgrip exercise, used as a “non-dynamic” exercise trial to study patterning of breathing. Above the ventilatory threshold (VTh ), two patterns of response were identified: in group 1, the rate of change in VT significantly increased, while in group 2 the breakpoint of ventilation solely resulted from fR increase. After the respiratory compensation point, a tachypnoeic response always occurred. A leftward shift of the VT vs. Ti relationship, i.e., an inspiratory off-switch reflex, was measured during the handgrip in group 2 subjects as well as marked fR variations. Our study identifies two different patterns of breathing after the VTh . The subjects who present a tachypnoeic response to exercise above the VTh have a higher sensitivity to pulmonary inflation and their tachypnoeic response was ubiquitous during a maximal handgrip test. © 2013 Elsevier B.V. All rights reserved.

1. Introduction The time course of hyperventilation during an incremental exercise has already been intensively explored in previous studies (Wasserman et al., 1973; Jones, 1988). Three phases in the ventilatory (VE ) response can be discerned: the first one is characterized by a linear relationship between V˙ E and oxygen uptake (V˙ O2 ). In the second one, V˙ E increases more than V˙ O2 (SV1 or ventilatory threshold, VTh ), and in the third one, VE increases more than V˙ CO2 (SV2 or respiratory compensation phase, RCP). Because V˙ E represents the product of the tidal volume (VT ) and respiratory frequency (fR ), many investigators have described the mean typical changes in the pattern of breathing, during an incremental exercise (Gallagher and Younes, 1986; Jones, 1988; Scheuermann and Kowalchuk, 1999; Whipp and Pardy, 1986; Younes and Kirvinen, 1984). These studies suggest that the pattern of breathing evenly varies during an incremental exercise, the VT increase reaching a plateau when VT levels off 50% to 60% of the forced vital capacity (FVC) while fR remains stable or slightly increases. As the exercise proceeds, a further V˙ E increase occurs due to fR increase, which is called “the tachypnoeic shift“. More recently, Naranjo et al. (2005) proposed a nomogram

∗ Corresponding author at: UMR MD2, Faculty of Medicine, Aix-Marseille University, Boulevard Pierre Dramard, 13916 cedex 20 Marseille, France. E-mail address: [email protected] (Y. Jammes). 1569-9048/$ – see front matter © 2013 Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.resp.2013.06.027

based on a sample of 64 subjects, giving an exponential relationship between VT and fR . On the other hand, our experience of breathing pattern analyses during an incremental cycling exercise in healthy subjects strongly indicates a large inter-individual variability of VT and fR changes. An extensive analysis of the literature data confirms this inter-individual variability. Indeed, Blackie et al. (1991) reports in 231 normal subjects a wide range of values of the ratio between the maximal VT measured at the end of exercise and the functional residual capacity (FVC), the VTmax /FVC ratio ranging from 38% to 72%. Moreover, Lucia et al. (1999), analysing the changes in breathing pattern of highly competitive cyclists, have shown that the tachypnoeic shift was not present in all their subjects. Recently, Cross et al. (2012) have identified in 28 healthy subjects, using the polynomial spline smoothing method, two successive disproportionate changes in the fR vs. V˙ O2 relationship, the second one being closely correlated with the RCP. They also reported that the VT plateau inconstantly occurred and was measured at the VTh in 14/28 subjects and at the RCP in 18/28 subjects. These recent data support our observations of a scattering of breathing pattern changes in healthy subjects performing an incremental exercise. The first aim of this study was to identify typical changes in breathing pattern in healthy normal male subjects who underwent an incremental cycling exercise. The changes in breathing pattern after VTh and RCP were highlighted and the reproducibility of inter-individual differences was examined. The second aim was to approach the possible mechanisms of these differences. We

60

G. Gravier et al. / Respiratory Physiology & Neurobiology 189 (2013) 59–66

formulated the hypotheses that the inter-individual discrepancies between the magnitude of VT and fR increases could result from (1) a mechanical limitation of inspiration in relation with the individual value of the maximal inspiratory capacity and/or (2) different gains of the vagal inspiratory off-switch reflex mechanism which limits the pulmonary hyperinflation (Clark and von Euler, 1972). We also hypothesized that the inter-individual variations of the breathing pattern response to exercise may occur in other circumstances of stimulation of the respiratory centers. Thus, we explored in some subjects their respiratory response to a sustained static handgrip trial which constituted a second mode of exercise trial to study patterning of breathing. 2. Materials and methods 2.1. Ethical approval The protocol was approved by the Ethics Committee of our institution (CCPPRB Marseille 1). The study conformed to the standards set by the latest revision of the Declaration of Helsinki. The procedures were carried out with the adequate understanding and written consent of the subjects. 2.2. Subjects A total of twenty eight male subjects were explored. They were considered sedentary subjects because they did not participate in regular formal exercise. Their morphological characteristics and breathing pattern at rest are shown in Table 1. At their inclusion in the study, pulmonary function tests were performed. A total body pressure plethysmograph (Masterlab, Jaeger, Bunnik, The Netherlands) allowed us to determine the maximal lung

Table 1 Morphological and, respiratory characteristics, and exercise performances of healthy subjects separated in two groups according to the differences in the breathing pattern changes during incremental cycling exercise. Values are the mean ± SEM. Group 1

Group 2

N Age, year Weight, kg Height, cm fH , min−1 V˙ E , LBTPS min−1 VT , LBTPS fR , min−1 TLC, LBTPS FVC, LBTPS IC, LBTPS FRC, LBTPS

10 43.4 ± 3.7 [26–68] 80.1 ± 3.5 177.8 ± 2.4 70 ± 3.1 12.2 ± 1 0.809 ± 0.1 17 ± 1.3 7.6 ± 0.3 5.4 ± 0.2 4 ± 0.2 3.6 ± 0.3

18 32.9 ± 3.2 [19–63] 76.6 ± 2.2 180 ± 1.3 76.7 ± 2.7 11.4 ± 0.6 0.725 ± 0.04 16.8 ± 1 7.26 ± 0.33 5.1 ± 0.3 3.7 ± 0.2 3.5 ± 0.2

Ventilatory threshold Time, s Power, W V˙ O2 , mlSTPD min−1 kg−1 V˙ O2 , %VO2 max

357 ± 21 136 ± 10 22 ± 1 62 ± 2.7

441 ± 44 160 ± 31 25 ± 2 67 ± 2

Respiratory compensation point Time, s Power, W V˙ O2 , mlSTPD min−1 kg−1 V˙ O2 , %VO2 max

n=6 636 ± 53 220 ± 16 31 ± 3 85 ± 2.3

n=9 640 ± 81 222 ± 28 37 ± 4 88.6 ± 1.4

End exercise Power, W V˙ O2 max , ml min−1 kg−1

n = 10 224 ± 15.1 36 ± 2.2

n = 18 244 ± 17.3 37 ± 2.6

VE : ventilation; VT : tidal volume; fR : respiratory frequency; fH : cardiac frequency, TLC: total lung capacity, FVC: Forced vital capacity, IC: inspiratory capacity, FRC: functional residual capacity, BTPS: body temperature pressure saturated condition; V˙ O2 : oxygen uptake.

volumes, that are the forced vital capacity (FVC), the total lung capacity (TLC), and also the functional residual capacity (FRC). All these data are shown in Table 1. 2.3. Measurements of respiratory variables and heart rate For breath by breath analyses of V˙ E , VT and fR , we used a turbine flow-meter connected to a face mask (dead space: 30 ml) which was designed to form an air-tight seal over the patient’s nose and mouth. A sampling catheter connected the outlet of the mask to fast-response differential paramagnetic O2 and infrared CO2 analyzers (90% response time in 100 ms) which measured end-tidal partial pressures of O2 (PETO2 ) and CO2 (PETCO2 ), respectively (Oxycon beta, Jaeger, Bunnik, The Netherlands). The microcomputer software (Oxycon beta, Jaeger, Bunnik, The Netherlands) averaged for 5 consecutive seconds data of ventilation (V˙ E ), tidal volume (VT ), breathing frequency (fR ), inspiratory duration (Ti ), and the Ti /Ttot ratio. The software also computed the oxygen uptake (V˙ O2 ), the CO2 production (V˙ CO2 ), and the ventilatory equivalents for O2 (EqO2 = V˙ E /V˙ O2 ) and CO2 (EqCO2 = V˙ E /V˙ CO2 ). A calibration procedure for the flow meter and gas analyzer systems was carried out before each test. At rest, the data were averaged for a 10-min period. The percutaneous oxygen saturation (SpO2 ) was continuously measured throughout the exercise challenge and the recovery period using an infrared analyser (Nellcor model N3000, TX, USA). ECG was continuously recorded from standard ECG leads and the cardiac frequency (fH ) was computed. 2.4. Cycling exercise Each subject performed an incremental exercise test on an electrically braked cycle ergometre (Ergometrics ER 800, Jaeger, Bunnik, The Netherlands). In all the subjects, the exercise protocol consisted of: (1) a 10-min rest period, during which all the variables were averaged, (2) a 2-min 0-W work load period used to reach the 1 Hz pedaling frequency, (3) a work period. The work period started at a work load of 20 W and the load was increased by 20 W every 1 min until the subject stopped pedaling. Two operators visually determined the VTh and RCP from EqO2 and EqCO2 data. VTh corresponded to the V˙ O value at which EqO exhibited 2

2

a systematic increase without a concomitant increase in EqCO2 (Wasserman et al., 1973). RCP was determined if the EqCO2 curve began to rise. This corresponded to a second breakpoint on the VE versus time curve. VTh and RCP breakpoints were independently determined by the two operators. When we could not meet the three criteria we rejected the RCP data analyses. The difficulties encountered to determine the RCP breakpoint did not bias our study because, as shown in results, the inter-individual variations of the breathing pattern (fR and VT ) were only detected at the VTh . Because the data were averaged for 5 consecutive seconds, the inter-operator differences in breakpoints determination, when they occurred, never exceeded 10–15 s. In such a case, we decided to choose the intermediate epoch to determine the breakpoint. VTh and RCP were expressed in absolute value of oxygen uptake related to the body weight. The peak V˙ O2 value (V˙ O2 max ) was measured when the subject had reached his predicted maximal cardiac frequency (220 – age). The reproducibility of the breathing pattern changes during the incremental exercise was tested in nine subjects who repeated twice at a two months interval the cycling protocol. 2.5. Determination of the different ventilatory patterns The different components of the breathing pattern (V˙ E , VT , fR ) as well as the fH value were plotted against the oxygen uptake,

G. Gravier et al. / Respiratory Physiology & Neurobiology 189 (2013) 59–66

61

Fig. 1. Examples of determination of group 1 and group 2 subjects according to the changes in the slope of least square regression between VT and V˙ O2 above the VTh . Regression equations with SEM of slopes are indicated.

expressed in percentage of the V˙ O2 max measured in each subject. Then, in each subject these plots were separated into two segments separated by one or two breakpoints corresponding to the VTh and RCP, when it occurred, and they were fitted by linear least squares regressions. A Student’s t-test was used to determine significant changes in the mean values of slopes of the regression lines with the corresponding standard deviation. This allowed us to distinguish two groups of subjects. In group 1, the rate of VT change significantly increased above the VTh (positive difference between preand post-VTh slopes), while in group 2 this difference was negative. Fig. 1 gives individual examples. In fifteen subjects who presented a RCP after the VTh determination, we considered three segments separated by the two breakpoints (VTh and RCP). 2.6. Maximal static handgrip exercise Eight subjects accepted to participate to another protocol to determine their ventilatory response to a maximal handgrip exercise. The subjects were seated comfortably and wore a mask connected to the turbine flow meter allowing breath-by-breath measurements of V˙ E , VT , and fR . At rest, the data were averaged for a 10-min period. One forearm was maintained in a horizontal prone position in an anatomic device specially built for the experiment that allowed to perform an isometric handgrip (Dousset et al., 2002). At the beginning of the experiment, the subject was instructed to perform three maximal voluntary contractions (MVC) sustained for 3 s, with 1 min interval between each MVC. The highest force recording of the three contractions was considered as the MVC. A pneumatic cuff was placed, as high as possible, on the exercising upper arm to allow post-exercise circulatory occlusion. The subjects were given visual feed-back from a load cell (ZC Scaime, Annemasse, France) to keep the pre-set maximal force level constant for 60 s. Ten seconds before the cessation of exercise, the pneumatic cuff on the exercising arm was rapidly inflated to suprasystolic pressure values (250 mmHg) and the circulatory occlusion was maintained for 1 min in the post-exercise period. Half an hour after the first handgrip bout, the subject repeated the same exercise with the other arm. We chose to limit the handgrip tests to 1-min periods to compare the changes in V˙ E , VT , and fR to those measured during the first min following the determination of VTh during the incremental cycling exercises performed by the same individuals.

2.7. Statistical analyses Data are presented as means ± standard error of means (SEM). The ventilatory variables at rest, at the VTh , RCP, and the end of exercise were compared between groups using unpaired t-tests. In each group, the comparisons between pre- and post breakpoints (VTh and RCP) were made using a one way repeated measures variance analysis. A paired t-test was used to compare the reproducibility of the breathing pattern changes during the incremental cycling in subjects who repeated twice the exercise. A linear regression analysis using Pearson correlations was performed to test the associations between variables collected during cycling and also to compare the changes in ventilatory variables measured at the end of the first min period following the VTh determination and during the handgrip exercise. Significant difference was set at p < 0.05. 3. Results 3.1. Characteristics of the subjects at rest Based on inter-individual differences in VT variations above the VTh determination (Fig. 1), ten subjects were included in group 1 and eighteen in group 2. Table 1 shows that their morphological characteristics, breathing pattern at rest, pulmonary function, performances at work did not differ. 3.2. The changes in cardiorespiratory variables during the incremental cycling First, no SpO2 decrease was observed during the whole challenge in all the subjects, discarding any alteration of the oxygen extraction in some individuals. Despite this fact, marked inter-individual changes in the breathing pattern were noted. Fig. 2 gives individual examples of the evolution of V˙ E , VT and fR changes during exercise in two subjects of groups 1 and 2 for whom both the VTh and RCP were determined. The mean values of the slopes of V˙ E , VT , or fR vs. V˙ O2 regression lines are shown in Table 2. Below the VTh determination, the patterns of V˙ E , VT , and fR changes did not significantly differ between the two groups. The rate of V˙ E increase was always significant above the VTh and RCP in both groups but the changes in ventilatory pattern (VT , fR ) markedly differed between groups. Above the VTh , the rate of VT increase was significantly accentuated

62

G. Gravier et al. / Respiratory Physiology & Neurobiology 189 (2013) 59–66

Fig. 2. Individual examples of the rate of changes in tidal volume (VT ), respiratory frequency (fR ), and cardiac frequency (fH ) in group 1 and group 2 subjects for whom both the VTh and RCP were determined.

in group 1 but it was reduced in group 2, for which the V˙ E breakpoint solely resulted from fR increase. Above the RCP, the V˙ E breakpoint only resulted from an accentuated tachypnoeic response. Table 2 also shows that the rate of fH changes did not vary after VTh but decreased slightly though significantly after RCP in both groups. We also noted intergroup differences in the respiratory timing, i.e., the Ti /Ttot ratio. Fig. 3 shows that this ratio was significantly lower in group 1 subjects at rest and throughout the exercise challenge. Thus, the expiratory duration was lengthened in group 1 subjects who preferred to recruit VT and not fR above the VTh . 3.3. Reproducibility of breathing pattern changes during exercise Table 3 shows a high reproducibility of VT and fR variations above the VTh in the subjects of each group who repeated twice the exercise challenge. 3.4. The influence of maximal pulmonary capacities on VT changes Table 1 clearly indicates no intergroup differences in TLC, FVC, and IC. When VT variations measured after VTh and RCP, and at V˙ O2 max were related to the corresponding values of TLC, no significant intergroup differences were noted at VTh . However, the VT /TLC ratio measured at RCP and V˙ O2 max was significantly higher in group 1 subjects who were able to breathe at higher volume in their

Fig. 3. The differences in the respiratory timing, that is the ratio of inspiratory duration (Ti ) to the total breath duration (Ttot ), between groups 1 and 2 subjects at rest then during the incremental cycling exercise. Values are the means ± SEM. Asterisks indicate the Ti /Ttot ratio significantly increases during exercise (*p < 0.05). Symbol $ denotes significant intergroup differences ($ p < 0.05; $$ p < 0.01).

inspiratory capacity than the group 2 ones: VT (RCP)/TLC = 40 ± 2% in group 1 vs. 32 ± 2% in group 2, p < 0.05; VT (V˙ O2 max )/TLC = 41 ± 2% in group 1 vs. 35 ± 1% in group 2.

G. Gravier et al. / Respiratory Physiology & Neurobiology 189 (2013) 59–66 Table 2 The changes in cardiorespiratory variables measured below and above the ventilatory threshold (VTh ) and the RCP in group 1 and 2 subjects. VTh :

Group 1(n = 10)

Group 2 (n = 18)

V˙ E changes: Slope V˙ E vs. V˙ O2 , LBTPS mlO2 −1 kg Pre VTh 1.57 ± 0.14 3.56 ± 0.42*** Post VTh VT changes: Slope VT vs. V˙ O2 , LBTPS mlO2 −1 kg min Pre VTh 0.06 ± 0.01 Post VTh 0.10 ± 0.01** fR changes: Slope fR vs. V˙ O2 , kg mlO2 BTPS−1 Pre VTh 0.22 ± 0.10 Post VTh 0.34 ± 0.11 fH changes Slope fH vs. V˙ O2 , kg mlO2 BTPS−1 Pre VTh 3.10 ± 0.26 3.20 ± 0.33 Post VTh RCP:

1.67 ± 0.1 3.70 ± 0.37***

$$$

$$$

63

3.5. The role played by the inspiratory off-switch mechanism In each group, the corresponding mean values ± SEM of VT and Ti measured at rest, VTh , RCP, and end of exercise (Max) were plotted together to build the relationship initially described by Clark and von Euler (1972). Compared to group 1, the amplitude of breath in group 2 was markedly limited above the VTh (Fig. 4). The inspiratory duration was significantly lengthened at rest in group 2

0.08 ± 0.01 0.04 ± 0.01***

0.17 ± 0.07 1.16 ± 0.12***

3.20 ± 0.20 2.80 ± 0.20

Group 1(n = 6)

Group 2 (n = 9)

VE changes: Slope V˙ E vs. V˙ O2 , LBTPS mlO2 −1 kg 3.00 ± 0.30 Pre RCP Post RCP 3.80 ± 0.15* VT changes: Slope VT vs. V˙ O2 , LBTPS mlO2 −1 kg min Pre RCP 0.11 ± 0.02 0.06 ± 0.03 Post RCP fR changes: Slope fR vs. V˙ O2 , kg.mlO2 BTPS−1 0.21 ± 0.08 Pre RCP 0.67 ± 0.10*** Post RCP fH changes Slope fH vs. V˙ O2 , kg mlO2 BTPS−1 Pre RCP 2.38 ± 0.45 Post RCP 1.38 ± 0.36*

2.54 ± 0.07 3.10 ± 0.20*

$$

0.03 ± 0.01 0.03 ± 0.01

$$

0.68 ± 0.05 0.85 ± 0.03**

$

1.47 ± 0.26 0.88 ± 0.22*

Values are the mean ± SEM. V˙ E : ventilation; VT : tidal volume; fR : respiratory frequency; fH : cardiac frequency. For each variable, are reported the slope of least square linear regressions computed before (pre: from 20 W to VTh ) and after (post: from VTh to V˙ O2 max ) the VTh determination. For subjects where RCP could be determined, the slopes of least square linear regressions are computed between the VTh and RCP (Pre RCP) and between RCP and V˙ O2 max (Post RCP). * Asterisks denote significant differences between pre- and post VTh or pre- and post-RCP values: p < 0.05. ** Asterisks denote significant differences between pre- and post VTh or pre- and post-RCP values: p < 0.01. *** Asterisks denote significant differences between pre- and post VTh or pre- and post-RCP values: p < 0.001. $ Symbol $ indicates significant intergroup differences: p < 0.05. $$ Symbol $ indicates significant intergroup differences: p < 0.01. $$$ Symbol $ indicates significant intergroup differences: p < 0.001.

Fig. 4. The VT versus Ti relationships obtained in groups 1 and 2 subjects. The mean values ± SEM of both variables were calculated at rest, at determination of the VTh and RCP, and at V˙ O2 max .

Table 3 Reproducibility of breathing pattern changes in group 1 and 2 subjects. For each variable, are reported the slope of least square linear regressions computed before (pre: from 20 W to VTh ) and after (post: from VTh to V˙ O2 max ) the VTh determination. Values are the means ± SEM. No significant differences were noted between the two trials. Group 1 (n = 3) First trial VE changes: Slope V˙ E vs. V˙ O2 , LBTPS mlO2 −1 kg 1.91 Pre VTh 4.36 Post VTh VT changes: Slope VT vs. V˙ O2 , LBTPS mlO2 −1 kg min Pre VTh 0.06 0.13 Post VTh fR changes: Slope fR vs. V˙ O2 , kg mlO2 BTPS−1 Pre VTh 0.39 0.28 Post VTh * ** ***

Group 2 (n = 6) Second trial

First trial

Second trial

± 0.22 ± 0.10***

1.97 ± 0.26 4.70 ± 0.13***

1.71 ± 0.14 3.01 ± 0.19**

1.74 ± 0.12 3.18 ± 0.37**

± 0.01 ± 0.03*

0.06 ± 0.01 0.10 ± 0.02*

0.07 ± 0.01 0.03 ± 0.01*

0.07 ± 0.01 0.03 ± 0.01*

± 0.08 ± 0.08

0.36 ± 0.07 0.40 ± 0.04

0.28 ± 0.06 0.86 ± 0.10***

0.25 ± 0.10 0.80 ± 0.10***

Asterisks denote significant post VTh variations in each trial: p < 0.05. Asterisks denote significant post VTh variations in each trial: p < 0.01. Asterisks denote significant post VTh variations in each trial: p < 0.001.

64

G. Gravier et al. / Respiratory Physiology & Neurobiology 189 (2013) 59–66

4.2. Limitations of the study To avoid factors which could contribute to inter-individual pattern differences during exercise, we chose to only explore male subjects who did not participate in regular formal exercise and thus in a particular sport training program. This allowed to eliminate the occurrence of possible already demonstrated sex differences (Guenette et al., 2009) and mostly learning processes which could intervene in the observed differences in breathing pattern. Indeed, Lucia et al. (1999) have reported that professional cyclists present a lack of the “tachypnoeic shift” during the intense phase of incremental exercise and thus, recruit more VT to finish the exercise. The role of the well-known cycling-respiratory coupling (Bechbache and Duffin, 1977) in the fR changes must be discarded because throughout the cycling bout each subject was asked to pedal at a constant rhythm (60 cpm) and we were mindful of its maintenance. 4.3. Hypothetical considerations on the intergroup differences Fig. 5. fR changes measured at the end of the maximal handgrip trial and the end of the 1-min period following the VTh determination were measured in 8 subjects (group 1: n = 3; group 2: n = 5) who repeated twice the handgrip. The least square regression line with 95% confidence interval is shown.

subjects (1.70 ± 0.08 s vs. 1.52 ± 0.05 s in group 1; p < 0.05) and at VTh (1.35 ± 0.06 s vs. 1.22 ± 0.04 s in group 1; p < 0.05), but not at RCP and VO2 max .

3.6. Comparison between the ventilatory responses to cycling exercise and maximal handgrip Three group 1 and five group 2 subjects accepted to participate to another protocol to test their ventilatory response to a maximal static handgrip exercise. In all these subjects, the ventilatory response to the handgrip test was always maximal at the end of the 90-s period of sustained static effort, then it declined during the consecutive 1-min period of arterial occlusion. When compared to breathing pattern changes measured above the VTh during cycling exercise, the fR increase measured at the end of the handgrip, before the arterial occlusion, were positively correlated with the fR increase measured for the 1-min period following the VTh determination during cycling (Fig. 5).

4. Discussion 4.1. Data summary The main result of this study was the existence of opposite but reproducible changes in the breathing pattern above the ventilatory threshold in healthy subjects performing an incremental maximal cycling exercise. In group 1 subjects, the rate of VT change, but not that of fR , significantly increased above the VTh . On the other hand, in group 2 the rate of VT changes decreased and that of fR increased above the VTh . In both groups, the rate of fR changes increased above the RCP. After the VTh , we observed a leftward shift of the VT vs. Ti relationship in the group 2 subjects, which was in favor of a higher gain of the vagal inspiratory off-switch reflex in these individuals. We also noted that the subjects presenting the highest fR increase above the VTh during cycling (group 2) also had the highest fR variations during the handgrip test. This indicates that intergroup differences in breathing pattern response to exercise are reproduced in another experimental stimulation of the respiratory centers.

Several mechanisms may be proposed to explain the intergroup differences in the breathing pattern variations during incremental cycling, mostly the limitation of VT increase and the tachypnoeic response measured in the group 2 subjects. 4.3.1. Inter-relationships between the pulmonary mechanics and breathing pattern changes at work In parallel to the increase in workload during heavy exercise, several studies have shown in healthy subjects a rise in the end-expiratory lung volume due to an expiratory flow limitation, leading to a dynamic pulmonary hyperinflation (McClaran et al., 1999; Pellegrino et al., 1993). We did not measure such an expiratory flow limitation in our subjects which would lengthen the expiratory duration and thus reduce the Ti /Ttot ratio. Indeed, the Ti /Ttot ratio significantly increased throughout the exercise challenge in all the subjects. Previous works have also established relationships between the work of breathing (WOB) and the breathing pattern during exercise, defining the minimum WOB which corresponds to the optimal value of fR used by the subjects at rest and at work (Mead, 1960; Gallagher et al., 1987). At low and moderate levels of exercise, the ventilation increases primarily through an increase in tidal volume, when the energy cost of the respiratory muscles is minimized, whereas at high levels of exercise fR changes are dominant and tend to minimize the elastic work of breathing with the enhanced stiffness of the respiratory system (Gallagher et al., 1987). Thus, during an incremental exercise a breaking point might exist when the mechanical adaptation of the respiratory system to hyperventilation becomes less efficient (increased WOB), forcing the subject to utilize more fR than VT . However, the occurrence of inter-individual difference in the WOB changes during incremental cycling seems doubtful in our subjects who have the same physical fitness, the same mechanical characteristics of the respiratory apparatus, and the same maximal performances (VTh , VO2 max , maximal power) at work. 4.3.2. Effect of the inspiratory off-switch reflex mechanism on the breathing pattern changes during cycling The control of VT and fR during the exercise hyperventilation could depend on the activation of the vagal pulmonary stretch receptors which limit the VT increase, that is the “inspiratory off-switch” mechanism described by Clark and von Euler (1972). During a rebreathing challenge eliciting hyperventilation, these authors reported a leftward shift of the VT vs. Ti relationship limiting the VT increase in some subjects who were supposed to have an efficient vagal control of breathing. In our study, the group 1 subjects who continued to increase VT above the VTh , had a minimal leftward shift of the VT vs. Ti curve, compared to the

G. Gravier et al. / Respiratory Physiology & Neurobiology 189 (2013) 59–66

group 2 subjects who rapidly limited the VT variations above the VTh . The hypothesis of an inter-individual difference in the strength of the Hering–Breuer reflex may be evoked and was already shown in healthy subjects (Jammes et al., 1976a, 1976b). Several studies have also noted marked differences in the ventilatory response to exercise between healthy subjects and heart-lung transplant patients, who had a vagal denervation. Compared to normal subjects, the transplanted patients had a disproportionate VT increase and reduced fR changes (Grassi et al., 1993; Sciurba et al., 1988). 4.3.3. The role of the chemoreflex control of breathing in the rate of fR change at RCP The breathing pattern differences between the two groups disappeared after the RCP where fR increased in all subjects. It is well known that the exercise-induced metabolic acidosis increases fR through the chemoreflex of breathing (Mateika and Duffin, 1995; Péronnet and Aguilanu, 2006; Whipp, 2007). The blood pH and lactate concentration were not here measured but we recently confirmed the well documented observations of an abrupt lactate increase after the VTh determination during cycling in some subjects who participated to the present study (Gravier et al., 2013). Blood acidosis constitutes a potent chemostimulation of the respiratory centers which represents a hypothetical explanation of the further fR increase above the RCP in all our subjects. Duffin et al. (2000) have proposed a model of ventilatory response to chemostimulation which incorporates tidal volume and frequency responses. They showed that ventilation increased linearly with the rise of PCO2 , with tidal volume usually contributing more than frequency to the increase. However, they also reported interindividual differences in the breathing pattern response, some individuals recruiting more fR than VT . When breathing was further driven in hypoxia, a second breakpoint was often observed, with frequency usually contributing more than tidal volume to the increase. Duffin’s observations are useful to explain the constant fR increase measured in our subjects at high exercise levels when a chemostimulation probably occurred but they resulted from different mechanisms because no SpO2 decrease, and thus no hypoxia, could be detected in our subjects even at V˙ O max . 2

5. Comparison between breathing pattern changes during cycling and maximal static handgrip exercise The changes in ventilation during sustained static muscle contractions are well documented. Iellamo et al. (1999) have reported that ventilation increased during sustained contractions then rapidly decreased during the post-exercise circulatory occlusion. We also observed in healthy humans sustaining a static handgrip at 50% of maximal for 90 s that ventilation increased whether or not an arterial occlusion was superimposed to the exercise bout (Hug et al., 2004). VT and fR changes during the handgrip were only measured by Muza et al. (1983) and Fontana et al. (1993) who reported opposing findings, fR increasing in the study by Fontana and coworkers but not in that by Muza et al. (1983). These discrepancies could be due to the different populations of subjects. Indeed, in our present study fR only increased during the handgrip in subjects having a tachypnoeic response to cycling. During inframaximal (50%) handgrip sustained until exhaustion, we already determined a ventilatory threshold characterized by an abrupt EqO2 increase which also occurred during a complete arterial blood flow interruption suppressing any venous return from the exercising forearm (Hug et al., 2004). Thus, the maximal handgrip bout executed in the present study must correspond to an activation of the respiratory centers above the VTh as in the maximal cycling protocol. This could explain the similarities here reported between the fR changes in the two circumstances of respiratory stimulation. Thus,

65

the static hand grip experiments were only considered as a second mode of exercise thru which breathing patterning could be examined. We here reported similarities between breathing pattern changes in the same subjects performing a maximal cycling exercise or a maximal static handgrip. The mechanistic source of the inter-individual differences in the respiratory drive cannot be explained. Indeed, the role played by the different muscle afferents in the respiratory response to static efforts could be only supposed and this respiratory response also involves central mechanisms including the central command, and also behavioral and/or cognitive influences. 6. Conclusions The present study brings new insights on the changes in breathing pattern during an incremental maximal cycling exercise. This clearly shows two opposite patterns of response, some subjects preferring to recruit VT and others fR after the ventilatory threshold. Different explanations were examined but the more consistent findings are that the respiratory control by peripheral vagal afferents seems to prevail in subjects who increase fR and not VT after VTh . The others continue to increase VT and seem to neglect the peripheral information coming from the lungs and the exercising muscles. Such inter-individual differences in the respective part played by the vagal afferents and the central respiratory command were already reported in previous works (Jammes et al., 1976a, 1976b). Conflict of interest statement There is no conflict of interest. References Bechbache, R.R., Duffin, J., 1977. The entrainment of breathing frequency by exercise rhythm. Journal of Physiology 272, 553–561. Blackie, S.P., Fairbarn, M.S., McElvaney, N.G., Wilcox, P.G., Morrison, N.J., Pardy, R.L., 1991. Normal values and ranges for ventilation and breathing pattern at maximal exercise. Chest 100, 136–142. Clark, F.J., von Euler, C., 1972. On the regulation of depth and rate of breathing. Journal of Physiology 222, 267–295. Cross, T.J., Morris, N.R., Schneider, D.A., Sabapathy, S., 2012. Evidence of breakpoints in breathing pattern at the gas-exchange thresholds during incremental cycling in young, healthy subjects. European Journal of Applied Physiology 112, 1067–1076. Dousset, E., Steinberg, J.G., Faucher, M., Jammes, Y., 2002. Acute hypoxemia does not increase the oxidative stress in resting and contracting muscle in humans. Free Radical Research 36, 701–704. Duffin, J., Mohan, R.M., Vasiliou, P., Stephenson, R., Mahamed, S., 2000. A model of the chemoreflex control of breathing in humans: model parameters measurement. Respiration Physiology 120, 13–26. Fontana, G.A., Pantaleo, T., Bongianni, F., Cresci, F., Manconi, R., Panuccio, P., 1993. Respiratory and cardiovascular responses to static handgrip exercise in humans. Journal of Applied Physiology 75, 2789–2796. Gallagher, C.G., Younes, M., 1986. Breathing pattern during and following maximal exercise in patients with chronic obstructive lung disease, interstitial lung disease and cardiac disease and in normal subjects. American Review of Respiratory Disease 133, 581–586. Gallagher, C.G., Brown, E., Younes, M., 1987. Breathing pattern during maximal exercise and during submaximal exercise with hypercapnia. Journal of Applied Physiology 63, 238–244. Grassi, B., Ferretti, G., Xi, L., Rieu, M., Meyer, M., Marconi, C., Cerretelli, P., 1993. Ventilatory response to exercise after heart and lung denervation in humans. Respiration Physiology 92, 289–304. Gravier, G., Steinberg, J.G., Lejeune, P.J., Delliaux, S., Guieu, R., Jammes, Y., 2013. Exercise-induced oxidative stress influences the motor control during maximal incremental cycling exercise in healthy humans. Respiratory Physiology and Neurobiology 186, 265–272. Guenette, J.A., Querido, J.S., Eves, N.D., Chua, R., Sheel, A.W., 2009. Sex differences in the resistive and elastic work of breathing during exercise in endurancetrained athletes. American Journal of Physiology – Regulatory, Integrative and Comparative Physiology 297, R166–R175. Hug, F., Faucher, M., Marqueste, T., Guillot, C., Kipson, N., Jammes, Y., 2004. Electromyographic signs of neuromuscular fatigue are concomitant with further

66

G. Gravier et al. / Respiratory Physiology & Neurobiology 189 (2013) 59–66

increase in ventilation during static handgrip. Clinical Physiology and Functional Imaging 24, 25–32. Iellamo, F., Massaro, M., Raimondi, G., Peruzzi, G., Legramante, J.M., 1999. Role of muscular factors in cardiorespiratory responses to static exercise: contribution of reflex mechanisms. Journal of Applied Physiology 86, 174–180. Jammes, Y., Prefaut, C., Delpierre, S., Mosse, P., Grimaud, C., 1976a. Relationships between eupnoeic pattern of breathing and ventilatory control in man: I. Response to airways occlusion during active lung inflation. Archives Internationales de Physiologie et de Biochimie 84, 955–968. Jammes, Y., Guillot, C., Grimaud, C., 1976b. Relationship between the control of the duration of inspiration and the resting ventilation pattern in conscious man. Comptes rendus des séances de la Société de biologie et de ses filiales 170, 1064–1068. Jones, N.L., 1988. Clinical Exercise Testing. Saunders WB Company, Philadelphia, London, Toronto, Montreal, Sydney, Tokyo, pp. 325. Lucia, A., Carvajal, C., Calderon, F.J., Alfonso, A., Chicharro, J.L., 1999. Breathing pattern in highly competitive cyclists during incremental exercise. European Journal of Applied Physiology 79, 512–521. Mateika, J.H., Duffin, J., 1995. A review of the control of breathing during exercise. European Journal of Applied Physiology 71, 1–27. McClaran, S.R., Wetter, T.J., Pegelow, D.F., Dempsey, J.A., 1999. Role of expiratory flow limitation in determining lung volumes and ventilation during exercise. Journal of Applied Physiology 86, 1357–1366. Mead, J., 1960. Control of respiratory frequency. Journal of Applied Physiology 15, 325–336. Muza, S.R., Lee, L.Y., Wiley, R.L., McDonald, S., Zechman, F.W., 1983. Ventilatory responses to static handgrip exercise. Journal of Applied Physiology 54, 1457–1462.

Naranjo, J., Centeno, R.A., Galiano, D., Beaus, M., 2005. A nomogram for assessment of breathing patterns during treadmill exercise. British Journal of Sports Medicine 39, 80–83. Pellegrino, R., Brusasco, V., Rodarte, J., Bab, T., 1993. Expiratory flow limitation and regulation of end-expiratory lung volume during exercise. Journal of Applied Physiology 74, 2552–2558. Péronnet, F., Aguilanu, B., 2006. Lactic acid buffering, nonmetabolic CO2 and exercise hyperventilation: a critical reappraisal. Respiratory Physiology and Neurobiology 150, 4–18. Scheuermann, B.W., Kowalchuk, J.M., 1999. Breathing pattern during slow and fast ramp exercise in man. Experimental Physiology 84, 109–120. Sciurba, F.C., Owens, G.R., Sanders, M.H., Griffith, B.P., Hardesty, R.L., Paradis, I.L., Costantino, J.P., 1988. Evidence of an altered pattern of breathing during exercise in recipients of heart-lung transplants. New England Journal of Medicine 319, 1186–1192. Wasserman, K., Whipp, B.J., Koyal, S.N., Beaver, W.L., 1973. Anaerobic threshold and respiratory gas exchange during exercise. Journal of Applied Physiology 35, 236–243. Whipp, B.J., Pardy, R.L., 1986. Breathing during exercise. In: Fishman, A.P., Cherniack, N.S., Widdicombe, J.G., Geiger, S.R. (Eds.), Handbook of Physiology, Section 3, The Respiratory System, vol. 3, Control of Breathing. American Physiological Society, Bethesda, pp. 595–619. Whipp, B.J., 2007. Physiological mechanisms dissociating pulmonary CO2 and O2 exchange dynamics during exercise in humans. Experimental Physiology 92, 347–355. Younes, M., Kirvinen, G., 1984. Respiratory mechanics and breathing pattern during and following maximal exercise. Journal of Applied Physiology 57, 1773–1778.