Eur J Appl Physiol (2006) 98:88–96 DOI 10.1007/s00421-006-0248-z

O RI G I NAL ART I C LE

Exercise-induced breathing patterns of youth are related to age and intensity Kristin S. Ondrak · Robert G. McMurray

Accepted: 8 June 2006 / Published online: 19 July 2006 © Springer-Verlag 2006

Abstract The inXuences of sex, age, exercise intensity, and end-tidal CO2 on the inspiratory drive ([VT kg¡1]·Ti¡1) and respiratory timing (Ti·Ttot¡1) components of ventilation were examined in 295 youth (138 females, 157 males); similarly distributed 8–18 years of age. Ventilatory and metabolic measures were collected breath-by-breath at rest and during a slow walk (4.0 km h¡1), fast walk (5.6 km h¡1) and run (8.0 km h¡1). Regression modeling for drive (age, sex, and PETCO2) found that sex was signiWcant (R2 < 0.017; P < 0.05) for rest and running, but not walking. Compared to rest, drive increased by 120% for the slow walk, 217% for the fast walk and 258% for the run (P < 0.0001). Drive decreased with age (P < 0.0001): rest = 0.41 ml kg¡1 s ¡1 year ¡1; slow walk = 0.90 ml kg¡1 s¡1 year¡1; fast walk = 1.30 ml kg¡1 s¡1 year¡1; and run = 1.47 ml kg¡1 s¡1 year¡1. In the regression models for timing, sex provided » 1% of the variance during the run, but was not signiWcant during rest or walking. Timing increased with exercise intensity by approximately 0.02 units (P < 0.001), but decreased by » 0.002 units year¡1 with age for all conditions (P < 0.003). Changes in drive and timing were marginally related to end-tidal CO2 (exercise R2 < 0.063 for all models). These results suggest that in the control of inspiratory drive and timing during exercise in youth, sex is of minor importance but there are age-related changes which are marginally associated with CO2.

K. S. Ondrak (&) · R. G. McMurray Department of Exercise and Sport Science, The University of North Carolina, 209 Fetzer Gymnasium, CB #8700, Chapel Hill, NC, 27599-8700, USA e-mail: [email protected]

123

Keywords

Ventilation · Drive · Timing · CO2

Introduction DiVerences in the ventilatory patterns of children and adults at rest and during exercise are well documented (Cooper et al. 1987; Gaultier et al. 1981; GratasDelamarche et al. 1993; Robinson 1938; Sato et al. 2000). At rest, during submaximal and maximal exercise, children, compared to adults, have lower absolute minute ventilation, absolute tidal volume (McMurray et al. 2003; Robinson 1938; Rowland 1996), and end tidal CO2 (Armon et al. 1991; Cooper et al. 1987; McMurray et al. 2003; Rowland 1996). Conversely, children have a greater frequency of breathing (McMurray et al. 2003; Robinson 1938; Rowland 1996), ventilation per kilogram body mass (Gratas-Delamarche et al. 1993; Robinson 1938; Rowland 1996), and ventilatory equivalent for oxygen (McMurray et al. 2003; Rowland 1996) than adults. Respiratory mechanics also change as children grow as gains in height are associated with decreased airway resistance and increased compliance of the respiratory system (Lanteri and Sly 1993). Research also shows that children, compared to either adolescents or adults, are more sensitive to CO2 (Cooper et al. 1987; Gratas-Delamarche et al. 1993; Nagano et al. 1998). Thus, children appear to have diVering respiratory physiology from adults. Although investigators have characterized the ventilatory pattern of youth, relatively few have examined inspiratory drive and respiratory timing at rest or during exercise. Inspiratory drive has been deWned as the Xow rate calculated by dividing tidal volume by inspiratory time: VT/Ti (Milic-Emili and Grunstein 1976).

Eur J Appl Physiol (2006) 98:88–96

Inspiratory drive is inXuenced by neural factors (MilicEmili 1982), arterial pressure (Grunstein et al. 1975), and CO2 (Askanazi et al. 1979). Tidal volume is dependent upon the Xow rate and the time of inspiration, which is somewhat dependent upon CO2 (Martin and Weil 1979). Thus, a greater tidal volume could be accomplished by either increasing the Xow rate at a given inspiratory time, or by lengthening inspiration time at a given Xow rate. Respiratory timing, also known as inspiratory duty cycle, is the duration of the inspiratory portion of ventilation. Although the physiology is not well understood, timing i the inspiratory time (Ti) divided by the total time (Ttot) of each breath: Ti·T tot¡1 (Milic-Emili and Grunstein 1976). Timing, like inspiratory drive, may be related to arterial CO2, but is also inXuenced by neural components (MilicEmili and Grunstein 1976). Little information exists on inspiratory drive and timing in youth at rest or during exercise, however in one of the few studies, Gaultier et al. (1981) found that in resting youth, VT and Ti increased with age, while timing (Ti·Ttot¡1) remained constant. In contrast, drive normalized for body mass decreased. No other studies were found that described these resting responses with respect to age or sex at rest. During exercise there appears to be a greater neural input for ventilation that coexists with the CO2 drive (Eldridge 1994; Martin and Weil 1979). Studies in adults have suggested that during steady-state exercise, the neural component can supersede the CO2 drive, resulting in ventilation disproportionate to the metabolic demands (Bainton 1972; McMurray and Ahlborn 1982). To our knowledge this has not been examined in youth. Thus, it could be hypothesized that the inXuence of CO2 on drive and timing diminishes as the exercise intensity increases and therefore, inspiratory drive and timing should change. Boule et al. (1989) studied drive and timing during exercise in a small, diverse age group of youth and reported increases in both drive and timing with increasing exercise intensity. However, these authors did not report the changes in drive or timing as related to age, sex or CO2. Mercier et al. (1991) examined changes in inspiratory drive and respiratory timing during maximal exercise in 11 to 16-year-old males and found that when body weight was not considered, drive increased approximately 63%. Mercier et al. (1991) did not report any data on females, nor did they examine the role of CO2. Thus, additional research is needed to explain the drive and timing components of the breathing pattern in children and adolescents as they relate to age, exercise intensity and CO2. The objective of our study was to determine the inXuence of sex and age on the ventilation pattern of

89

youth at rest and during three exercise intensities. Furthermore, we aimed to quantify the relationship of end-tidal CO2 (PETCO2) with drive and timing at rest and during exercise. This study appears to be the Wrst to compare these variables in females and males of diVering ages at rest and three exercise intensities.

Methods Participants A total of 295 (138 females, 157 males) children and adolescents, 8–18 years old, participated in this study. The participants considered themselves physically activity, but none were highly trained endurance athletes. Participants were obtained from the Energy Expenditure of Physical Activity in Youth (EEPAY) study, whose methods are fully described elsewhere (Harrell et al. 2005). The distribution of participants by age and sex are presented in Table 1 along with their descriptive characteristics. Although there were relatively small numbers of children of each age and large variability among them, we had suYcient power (
= 0.05;  = 0.80) to achieve statistical signiWcance. All procedures were approved by the University’s Institutional Review Board and written informed consent and assent were obtained from the parent(s) and child, respectively, prior to data collection. Table 1 Participant characteristics (mean § SD) presented by age and sex Age (year)

Sex

N

Height (cm)

Body mass (kg)

BMI (kg m¡2)

8

F M F M F M F M F M F M F M F M F M F M F M

13 13 14 14 13 18 14 15 13 15 11 16 12 17 11 14 14 11 9 11 14 13

131.9 § 4.5 130.4 § 6.6 136.2 § 7.7 134.5 § 7.1 145.0 § 5.8 139.4 § 6.5 149.2 § 6.7 149.4 § 7.5 157.8 § 7.4 159.6 § 8.7 161.1 § 8.1 164.0 § 7.7 161.9 § 7.4 165.3 § 10.2 163.2 § 5.4 173.3 § 7.4 166.0 § 5.8 174.4 § 3.5 165.0 § 5.8 174.6 § 4.5 162.6 § 4.5 174.7 § 7.6

29.5 § 4.7 28.9 § 5.1 31.0 § 4.1 32.7 § 6.4 39.8 § 8.0 36.1 § 7.7 44.6 § 12.8 41.6 § 7.8 49.7 § 9.3 49.8 § 12.2 58.1 § 13.4 49.4 § 8.6 57.3 § 17.8 57.2 § 15.2 67.8 § 15.9 63.6 § 8.2 64.1 § 12.9 69.6 § 10.8 60.7 § 9.3 72.4 § 11.0 61.3 § 8.3 68.7 § 12.1

16.9 § 1.9 16.9 § 1.5 16.7 § 1.5 17.7 § 2.2 18.8 § 2.9 18.4 § 2.9 20.0 § 5.4 18.5 § 2.4 19.8 § 2.5 19.4 § 3.7 21.9 § 3.9 18.0 § 2.4 21.6 § 4.9 20.6 § 3.6 25.4 § 6.0 21.5 § 3.0 23.3 § 5.2 23.1 § 2.9 22.2 § 3.0 23.7 § 3.1 23.2 § 2.7 22.5 § 3.4

9 10 11 12 13 14 15 16 17 18

123

90

Data collection took place during two laboratory visits and participants fasted for three hours prior to each. Height and body mass were measured using a stadiometer and balance beam scale, respectively (Perspective Enterprises, Portage, MI, USA; Detecto Scales, Inc., Webb City, MO, USA). These measures were used to calculate body mass index (BMI, kg m¡2). Resting oxygen uptake was measured for 15 min with the child reclined in a quiet room. Ventilatory data were then collected for 10 min at each of three exercise intensities: a slow walk (4 km h¡1, 2.5 mph), a fast walk (5.6 km h¡1, 3.5 mph) and a run (8.0 km h¡1, 5 mph). Prior to data collection, ample time was devoted to accommodating the child to the mask and treadmill. For those who could not sustain a run for the entire 10 min, a minimum of three minutes of steady state data was necessary for inclusion in the analyses. The determination of steady state was made by one of the authors (RGM) in all cases. Activities were presented in order of progressive diYculty, such that rest was done before any activity and slower walks were done before faster walks. Participants were given ample rest and obtained heart rates of <100 b min¡1 before beginning the second activity. Ventilatory measures were collected breath-bybreath using a 28 mm bidirectional digital turbine which was part of the COSMED K4b2 portable metabolic analyzer (Rome, Italy). Inspiratory time (Ti) and tidal volume (VT) were measured directly from the turbine, while minute ventilation (VE), PETCO2, respiratory frequency (fR), oxygen uptake (VO2), carbon dioxide output (VCO2), and respiratory timing (Ti·Ttot¡1), were calculated by the COSMED software. Breaths were eliminated from analysis if their VT was less than 200 ml or greater than the 95th percentile. The remaining breaths were measured over the entire steady state period and the mean of those responses is reported. Minute ventilation and tidal volume per kilogram body mass (VE kg¡1; VT kg¡1), minute ventilation per unit CO2 output ratio (VE/VCO2) were derived; as was inspiratory drive per kg mass ([VT kg¡1]·Ti¡1) and timing (Ti·Ttot¡1). Analytical methods

VE/VCO2, drive, and timing. If the ANOVA results for age and intensity were signiWcant (P < 0.05), simple regression analyses were computed to describe how age and intensity inXuenced each variable. Finally, a stepwise multiple regression analysis was completed for drive and timing to quantify the variance explained by age, sex and PETCO2. Statistical analyses were carried out using SAS software (Version 8.0, SAS, Cary, NC, USA).

Results The three-way ANOVA analysis revealed no signiWcant sex diVerences for inspiratory drive expressed per kg body mass (P = 0.065). In general, drive/kg decreased as the participants aged and increased as the activity became more intense (P < 0.0001 for both), as shown in Fig. 1. Simple regression for each of the four conditions revealed that drive decreased with age by 0.41 ml kg¡1 s¡1 year¡1 at rest, 0.90 ml kg¡1 s¡1 year¡1 during the slow walk, by 1.30 ml kg¡1 s¡1 year¡1 during the fast walk and by 1.47 ml kg¡1 s¡1 year¡1 during the run (P < 0.0001). When drive was not adjusted for body mass, no sex diVerences existed (P < 0.45), however, drive still increased with age and exercise intensity. The three-way ANOVA also found no signiWcant diVerences between males and females for timing (P = 0.72). Timing decreased slightly as the participants got older, but increased as the activity became more intense (P < 0.0001 for both), as shown in Fig. 2. Simple regression analysis for each of the four conditions showed that this age-related decrease was consistent for all activities (¡0.002 units year¡1, P < 0.003).

50

Rest Slow walk Fast walk Run

45 40

Drive/kg ([ml/kg]/s)

Procedures

Eur J Appl Physiol (2006) 98:88–96

35 30 25 20 15 10

Descriptive statistics [mean § standard deviation (SD)] were then calculated by sex and age for all ventilation variables, separately for all four conditions (rest and three exercise intensities). Three-way ANOVAs were computed to explore the inXuence of age, sex, and condition on VO2, VE kg ¡1, VT kg¡1, PETCO2,

123

5 0

8

9

10

11

12

13

14

15

16

17

18

Age (years)

Fig. 1 Mean eVect of age and exercise intensity on inspiratory drive adjusted for body mass (drive/kg = [VT kg¡1]·Ti¡1)

Eur J Appl Physiol (2006) 98:88–96

91

Timing (s/s)

0.55

Rest Slow walk Fast walk Run

0.5

0.45

0.4

0.35

8

9

10

11

12

13

14

15

16

17

between drive/kg, age, sex and PETCO2. The results, shown in Table 6, indicated that at rest the R2 was 0.507, of which 19% was attributable to PETCO2 and 31% to age (P < 0.001). Sex was found to have a minor but signiWcant relationship (P = 0.042). The results of stepwise multiple regression analysis for timing, age, sex and PETCO2 are presented in Table 7. At rest, the R2 was 0.07 with equal portions explained by PETCO2 and age (P = 0.002).

18

Age (years)

Slow walk

Fig. 2 Mean eVect of age and exercise intensity on respiratory timing (Ti·Ttot¡1)

Rest The mean (§ SD) for VO2 and the major ventilatory measures at rest are presented by sex and age in Table 2. The VO2 for males was greater than females, regardless of age. Similarly, the average VT kg¡1 and PETCO2 for males were greater than females, while the average VE/VCO2 was higher for females. Regression analysis showed that VO2 and PETCO2 increased with age (P < 0.0001). However, VE kg¡1, VE/VCO2, VT kg¡1, and fR declined with age (P < 0.005). A stepwise regression was run to explore the relationship

Table 3 presents the mean (§ SD) for VO2 and the major ventilatory measures by sex and age. The threeway ANOVA for PETCO2 indicated that the eVects of age, sex and activity were all signiWcant (P < 0.001). The average PETCO2 in males was greater than females. The PETCO2 for the males tend to increase with age, whereas PETCO2 for the females increase until approximately age 13 and then decrease (Table 3). Regression results show that VO2 increased, while VE kg¡1 decreased (P < 0.0001). Age-related decreases were also found for VT kg¡1, and fR; P < 0.0001. The models of the multiple regression analyses for drive/kg (Table 6) and timing (Table 7) were highly signiWcant. With respect to drive/kg, age accounted for 51% and PETCO2 accounted for < 4%.

Table 2 Age and sex comparisons for oxygen uptake (VO2), ventilation per kg (V E kg¡1), ventilation to carbon dioxide output (V E/VCO2), tidal volume per kg (VT kg¡1), respiratory frequency (fR) and end-tidal carbon dioxide (PETCO2) at rest Age (year)

Sex

N

VO2 (ml min¡1)

VE kg¡1 (l min¡1 kg¡1)

VE/VCO2 (l l¡1)

VT kg¡1 (ml kg¡1)

fR (b min¡1)

PETCO2 (mm Hg)

8

F M F M F M F M F M F M F M F M F M F M F M

13 13 14 14 13 18 14 15 13 15 11 16 12 17 11 14 14 11 9 11 14 13

202 § 43 192 § 37 190 § 54 200 § 60 201 § 44 214 § 53 235 § 54 209 § 40 226 § 64 274 § 63 257 § 107 235 § 42 251 § 69 251 § 68 253 § 60 282 § 65 239 § 46 307 § 59 239 § 39 279 § 58 248 § 62 277 § 35

0.25 § 0.03 0.25 § 0.04 0.22 § 0.05 0.23 § 0.08 0.18 § 0.05 0.20 § 0.04 0.17 § 0.03 0.18 § 0.05 0.16 § 0.05 0.18 § 0.06 0.14 § 0.03 0.17 § 0.06 0.16 § 0.04 0.14 § 0.03 0.13 § 0.04 0.14 § 0.03 0.13 § 0.02 0.13 § 0.02 0.14 § 0.02 0.12 § 0.02 0.13 § 0.02 0.13 § 0.02

45.2 § 8.3 42.7 § 9.4 43.7 § 10.9 39.2 § 7.4 42.0 § 12.0 38.1 § 5.1 37.8 § 4.2 39.9 § 6.5 40.2 § 8.5 38.3 § 6.1 36.8 § 5.4 38.5 § 5.0 40.3 § 5.4 37.3 § 4.7 39.3 § 3.2 37.4 § 6.2 40.3 § 6.0 35.8 § 3.6 41.3 § 5.4 36.1 § 2.7 39.7 § 8.0 39.1 § 7.6

13.2 § 3.8 11.0 § 2.6 10.9 § 4.2 12.1 § 5.3 8.6 § 2.1 11.5 § 4.6 9.4 § 2.2 10.3 § 2.9 8.4 § 2.5 9.7 § 3.2 8.8 § 3.9 10.3 § 4.6 8.2 § 3.3 8.4 § 2.5 6.7 § 1.5 9.6 § 2.4 7.8 § 1.9 9.4 § 2.5 7.8 § 1.1 7.5 § 1.5 7.8 § 2.2 8.3 § 2.7

21.9 § 5.4 24.3 § 4.3 22.8 § 7.6 20.7 § 5.1 22.6 § 6.9 20.3 § 5.3 20.1 § 3.3 19.7 § 4.1 20.2 § 2.9 22.8 § 12.5 17.9 § 3.2 19.1 § 4.8 21.7 § 3.3 19.7 § 4.1 19.9 § 2.3 16.2 § 4.5 18.3 § 4.2 16.4 § 3.2 19.8 § 2.8 17.1 § 3.2 18.2 § 4.7 17.0 § 3.8

36.0 § 5.3 37.8 § 3.6 37.4 § 3.5 38.9 § 6.3 41.1 § 4.9 39.4 § 5.1 38.2 § 2.4 38.0 § 4.2 40.0 § 3.3 38.6 § 5.1 40.2 § 4.0 38.1 § 3.8 36.6 § 2.8 40.2 § 3.1 37.2 § 2.2 38.9 § 6.0 36.9 § 4.2 39.1 § 3.4 35.3 § 1.8 40.8 § 2.5 37.8 § 6.0 39.8 § 2.5

9 10 11 12 13 14 15 16 17 18

SigniWcant P-values for sex: VO2 = 0.017, VE/VCO2 = 0.008, VT kg¡1 = 0.013, PETCO2 = 0.018; age: VO2 = <0.0001, VE kg¡1 = <0.0001, VT kg¡1 = <0.0001, fR = <0.0001

123

92

Eur J Appl Physiol (2006) 98:88–96

Table 3 Age and sex comparisons for oxygen uptake (VO2), ventilation per kg (V E kg¡1), ventilation to carbon dioxide output (V E/VCO2), tidal volume per kg (VT kg¡1), respiratory

frequency (fR) and end-tidal carbon dioxide (PETCO2) during the slow walk (4.0 km h¡1)

Age (year)

Sex

N

VO2 (ml min¡1)

VE kg¡1 (l min¡1 kg¡1)

VE/VCO2 (l l¡1)

VT kg¡1 (ml kg¡1)

fR (b min¡1)

PETCO2 (mm Hg)

8

F M F M F M F M F M F M F M F M F M F M F M

13 14 15 13 12 16 12 15 13 15 12 17 12 16 13 17 14 10 11 11 17 14

641 § 208 566 § 89 560 § 134 637 § 175 740 § 140 689 § 168 727 § 191 666 § 109 763 § 187 818 § 188 838 § 114 724 § 133 844 § 159 818 § 223 908 § 208 923 § 154 830 § 126 827 § 142 715 § 90 932 § 97 761 § 174 868 § 150

0.61 § 0.11 0.58 § 0.07 0.47 § 0.05 0.54 § 0.05 0.48 § 0.10 0.52 § 0.09 0.42 § 0.07 0.44 § 0.06 0.41 § 0.08 0.45 § 0.08 0.37 § 0.06 0.39 § 0.06 0.38 § 0.05 0.36 § 0.06 0.36 § 0.05 0.35 § 0.04 0.33 § 0.07 0.31 § 0.05 0.32 § 0.05 0.30 § 0.03 0.32 § 0.05 0.31 § 0.04

34.8 § 3.6 35.5 § 3.0 34.0 § 3.9 33.1 § 4.6 31.1 § 4.2 33.1 § 4.3 32.3 § 2.8 33.4 § 4.9 31.1 § 3.5 32.3 § 5.2 30.9 § 3.3 31.6 § 2.8 32.3 § 3.6 30.8 § 3.8 32.5 § 5.0 31.2 § 5.1 32.7 § 4.1 31.1 § 6.5 35.6 § 2.9 29.9 § 3.2 33.8 § 6.4 32.8 § 4.4

18.2 § 3.7 15.3 § 2.9 15.5 § 4.7 16.4 § 1.9 15.4 § 3.5 14.9 § 2.8 14.0 § 2.4 14.9 § 2.4 15.4 § 3.3 15.0 § 3.4 14.3 § 3.6 15.5 § 2.8 14.0 § 3.0 14.1 § 3.1 14.1 § 2.4 15.0 § 2.0 13.0 § 2.1 16.4 § 5.6 11.9 § 1.5 13.9 § 2.5 13.9 § 3.7 13.2 § 2.2

35.4 § 6.7 39.5 § 7.0 32.6 § 6.0 34.4 § 3.3 32.3 § 6.5 36.1 § 6.8 31.8 § 5.8 31.2 § 7.2 27.3 § 4.2 32.0 § 6.4 26.9 § 5.2 26.5 § 3.6 28.9 § 4.5 28.6 § 7.0 26.8 § 3.3 24.6 § 3.7 26.6 § 5.8 22.4 § 5.9 28.4 § 3.7 23.3 § 4.1 25.6 § 6.5 25.3 § 4.4

39.1 § 3.5 38.5 § 2.5 40.3 § 3.4 40.6 § 5.1 42.5 § 4.8 40.7 § 4.1 40.0 § 1.9 39.9 § 4.1 41.1 § 3.4 41.0 § 3.6 41.0 § 3.5 40.3 § 3.1 39.1 § 3.0 42.0 § 3.4 39.8 § 2.9 42.5 § 5.3 39.0 § 4.2 42.1 § 3.8 36.8 § 2.2 42.0 § 4.1 38.9 § 5.0 40.5 § 2.6

9 10 11 12 13 14 15 16 17 18

SigniWcant P-values for sex: PETCO2 = 0.009; age: VO2 = <0.0001, VE kg¡1 = <0.0001, VT kg¡1 = 0.001, fR = <0.0001

The model for timing resulted in the total variance of 7%, of which » 5% was attributed to age and < 2% was due to PETCO2 (P = 0.023). Fast walk The mean (§ SD) values for VO2 and the major ventilatory variables are presented by sex and age in Table 4. The average PETCO2 was greater in males than females. No other signiWcant sex-related diVerences were found. VO2 and PETCO2 increased signiWcantly (P < 0.0001) with age. Conversely, VE kg¡1 decreased as the children got older (P < 0.002). Similar age-related declines were found for VE/VCO2, VT kg¡1, and fR; P < 0.002 for all. The multiple regression analysis for drive during the fast walk (Table 6) indicated that age was highly signiWcant accounting for the majority of the variance; however, PETCO2 was also signiWcant and accounted for 3% of the variance. Age was the only signiWcant factor in the model for timing (Table 7), accounting for » 4% of the variance. Running Table 5 presents the means (§ SD) for during the run. The average VO2, VT kg¡1, and PETCO2 for males were

123

greater than for the females. However, average values for VE/VCO2 were greater in females than males. Regression showed signiWcant (P < 0.04) age-related increases in: VO2, VT kg¡1, and PETCO2. In contrast, VE kg¡1, fR and VE/VCO2 declined from age 8 to 18 years (P < 0.0001). In the multiple regression for drive, age, PETCO2, and sex were signiWcant factors (Table 6); with age accounting for the majority (< 43%) of the variance and PETCO2 accounting for only 6.2%. The inXuence of age and sex were also present in the analyses for timing, accounting for a combined total of 6.4% of the variance (Table 7). PETCO2 was not signiWcant (P > 0.15).

Discussion Our results show that age and exercise intensity modiWed the drive and timing for ventilation of children and adolescents. Although sex had statistical signiWcance in the regression models, from a clinical perspective the amount of variance accounted for by sex (< 2%) was not meaningful. The similarity of ventilatory drive and timing among the sexes has not been clearly demonstrated in previous literature. Our results also show that changes in ventilatory pattern were related to

Eur J Appl Physiol (2006) 98:88–96

93

Table 4 Age and sex comparisons for oxygen uptake (VO2), ventilation per kg (V E kg¡1), ventilation to carbon dioxide output (V E/VCO2), tidal volume per kg (VT kg¡1), respiratory

frequency (fR) and end-tidal carbon dioxide (PETCO2) during the fast walk (5.6 km h¡1)

Age (year)

Sex

N

VO2 (ml min¡1)

VE kg¡1 (l min¡1 kg¡1)

VE/VCO2 (l l¡1)

VT kg¡1 (ml kg¡1)

fR (b min¡1)

PETCO2 (mm Hg)

8

F M F M F M F M F M F M F M F M F M F M F M

13 14 15 15 13 18 14 16 14 15 13 17 12 17 12 17 15 11 12 11 17 13

850 § 288 767 § 109 768 § 165 883 § 164 986 § 197 916 § 240 960 § 204 904 § 167 1002 § 197 1034 § 213 1078 § 172 964 § 158 1113 § 198 1127 § 266 1257 § 287 1250 § 193 1151 § 174 1112 § 205 1020 § 176 1264 § 102 1035 § 213 1170 § 183

0.86 § 0.14 0.82 § 0.10 0.68 § 0.10 0.77 § 0.10 0.69 § 0.13 0.72 § 0.10 0.59 § 0.09 0.62 § 0.08 0.54 § 0.10 0.58 § 0.08 0.51 § 0.09 0.53 § 0.08 0.52 § 0.06 0.50 § 0.07 0.50 § 0.05 0.48 § 0.06 0.47 § 0.12 0.42 § 0.05 0.45 § 0.06 0.41 § 0.04 0.45 § 0.09 0.41 § 0.06

34.7 § 3.9 35.8 § 2.8 33.3 § 3.3 31.5 § 3.3 31.4 § 4.4 32.4 § 3.5 31.0 § 3.2 33.0 § 6.0 30.1 § 3.6 31.5 § 6.2 30.1 § 3.1 30.3 § 3.0 30.7 § 3.2 28.8 § 2.9 30.0 § 3.9 29.2 § 4.3 30.4 § 4.0 29.7 § 7.5 33.0 § 3.2 27.9 § 3.3 31.3 § 5.4 29.6 § 3.6

20.0 § 5.1 17.4 § 3.3 17.5 § 4.2 20.8 § 4.7 18.2 § 4.1 17.7 § 3.3 16.6 § 2.9 16.7 § 2.8 17.5 § 2.9 17.2 § 3.8 16.3 § 3.0 17.9 § 3.2 16.4 § 4.2 17.1 § 3.9 16.6 § 3.3 18.1 § 3.1 15.9 § 2.9 19.5 § 7.7 15.1 § 2.3 17.0 § 3.1 16.7 § 4.1 15.9 § 2.6

45.2 § 7.1 49.7 § 9.1 41.6 § 7.0 39.1 § 3.1 39.9 § 9.3 42.0 § 8.7 37.2 § 7.3 40.6 § 11.3 32.2 § 4.9 37.0 § 9.4 32.1 § 5.4 31.1 § 4.4 34.0 § 6.1 32.1 § 7.9 31.4 § 3.9 28.3 § 4.5 31.5 § 8.0 26.7 § 7.0 31.2 § 3.7 25.7 § 4.6 30.1 § 9.4 28.8 § 8.6

38.4 § 3.8 37.8 § 2.5 40.0 § 3.5 40.7 § 4.0 41.5 § 5.0 39.8 § 3.6 40.3 § 2.8 40.4 § 4.9 41.6 § 3.6 41.7 § 3.9 41.4 § 3.6 41.1 § 3.5 40.3 § 3.4 42.7 § 3.4 41.1 § 3.2 44.0 § 5.0 40.6 § 4.7 43.0 § 4.4 37.7 § 2.8 43.9 § 4.5 40.4 § 5.4 42.3 § 3.0

9 10 11 12 13 14 15 16 17 18

SigniWcant P-values for sex: PETCO2 = 0.005; age: VO2 = <0.0001, VE kg¡1 = <0.0001, VE/VCO2 = <0.0001, fR = <0.0001, PETCO2 = 0.017

Table 5 Age and sex comparisons for oxygen uptake (VO2), ventilation per kg (V E kg¡1), ventilation to carbon dioxide output (VE/VCO2), tidal volume per kg (VT kg¡1), respiratory frequency (fR) and end-tidal carbon dioxide (PETCO2) during the run (8.0 km h¡1) Age (year)

Sex

N

VO2 (ml min¡1)

VE kg¡1 (l min¡1 kg¡1)

VE/VCO2 (l l¡1)

VT kg¡1 (ml kg¡1)

fR (b min¡1)

PETCO2 (mm Hg)

8

F M F M F M F M F M F M F M F M F M F M F M

10 11 13 14 10 15 12 12 12 13 12 15 9 12 13 13 11 12 11 10 10 13

1149 § 107 1142 § 187 1156 § 166 1349 § 180 1594 § 290 1376 § 246 1573 § 256 1540 § 258 1841 § 326 1841 § 385 1878 § 466 1840 § 333 1889 § 275 1951 § 342 1972 § 391 2272 § 370 2008 § 229 2341 § 397 1980 § 245 2483 § 237 1822 § 367 2105 § 438

1.34 § 0.19 1.36 § 0.19 1.32 § 0.15 1.41 § 0.22 1.22 § 0.18 1.25 § 0.13 1.16 § 0.25 1.25 § 0.15 1.13 § 0.12 1.10 § 0.13 1.05 § 0.09 1.12 § 0.17 0.98 § 0.13 1.02 § 0.14 0.99 § 0.14 1.03 § 0.14 0.92 § 0.16 0.95 § 0.16 0.93 § 0.17 0.93 § 0.10 1.01 § 0.26 0.83 § 0.16

36.8 § 3.4 40.1 § 6.7 37.2 § 4.2 36.6 § 5.2 33.0 § 4.1 33.9 § 2.3 33.1 § 3.4 33.7 § 2.5 32.6 § 3.9 32.9 § 4.4 31.2 § 3.2 30.1 § 3.5 31.2 § 2.0 31.2 § 3.7 34.4 § 12.1 29.7 § 4.9 32.3 § 6.5 28.8 § 4.7 32.6 § 5.5 29.8 § 6.8 33.0 § 6.4 28.1 § 4.3

21.3 § 2.9 19.4 § 2.8 21.1 § 3.1 23.3 § 5.0 24.2 § 3.5 22.7 § 3.2 21.4 § 4.0 21.6 § 2.5 22.7 § 4.9 21.4 § 4.5 22.5 § 3.1 25.6 § 5.4 20.3 § 3.2 23.0 § 3.5 21.1 § 4.1 25.2 § 5.5 21.0 § 2.9 24.5 § 5.6 22.6 § 5.7 24.7 § 5.8 23.1 § 3.3 23.1 § 5.5

64.8 § 8.6 72.0 § 10.6 66.3 § 11.7 63.0 § 12.0 51.7 § 6.9 57.5 § 8.3 55.9 § 10.7 59.2 § 7.8 52.2 § 10.4 54.4 § 11.0 48.3 § 8.5 45.8 § 10.0 49.6 § 7.2 46.1 § 10.6 47.1 § 5.9 43.5 § 10.5 45.4 § 7.9 40.8 § 6.6 43.9 § 8.3 42.5 § 10.0 45.2 § 12.6 40.3 § 8.1

36.1 § 4.2 34.0 § 5.0 35.9 § 3.1 37.2 § 3.7 37.8 § 3.7 37.8 § 2.5 37.4 § 3.3 37.5 § 2.8 38.6 § 4.4 38.3 § 4.2 39.4 § 3.9 40.5 § 3.2 38.9 § 2.0 40.4 § 2.6 38.9 § 3.8 40.8 § 5.1 39.1 § 7.3 42.7 § 5.3 38.7 § 5.5 43.4 § 5.2 37.2 § 5.6 41.8 § 3.1

9 10 11 12 13 14 15 16 17 18

SigniWcant P-values for sex: VO2 = 0.001, VE/VCO2 = 0.045, VT kg¡1 = 0.027, PETCO2 = 0.005; age: VO2 = <0.0001, VE kg¡1 = <0.0001, VE/VCO2 = <0.0001, fR = <0.0001, PETCO2 = <0.0001

123

94

Eur J Appl Physiol (2006) 98:88–96

Table 6 Multiple regression of the inXuence of age, sex and PETCO2 on inspiratory drive at rest and during walking and running

Rest Age PETCO2 Sex Slow walk Age PETCO2 Fast walk Age PETCO2 Run Age PETCO2 Sex

Partial R2

Model R2



0.309 0.191 0.008

0.309 0.499 0.507

¡0.395 ¡0.236 0.394

< 0.0001 < 0.0001 0.042

0.505 0.037

0.505 0.542

¡0.893 ¡0.196

< 0.0001 < 0.0001

0.566 0.030

0.566 0.597

¡1.236 ¡0.233

< 0.0001 < 0.0001

0.426 0.062 0.016

0.426 0.488 0.504

¡1.260 ¡0.450 1.817

< 0.0001 < 0.0001 0.005

P value

Drive = ([VT kg¡1]·Ti¡1). Sex coded for males = 1, females = 0

Table 7 Multiple regression of the inXuence of age, sex and PETCO2 on respiratory timing at rest and during walking and running

Rest PETCO2 Age Slow walk Age PETCO2 Fast walk Age Run Age Sex

Partial R2

Model R2



0.034 0.034

0.034 0.068

0.0018 ¡0.0025

0.002 0.002

0.053 0.017

0.053 0.070

¡0.0022 0.0010

< 0.001 0.023

0.041

0.041

¡0.0016

< 0.001

0.052 0.012

0.052 0.064

¡0.0017 ¡0.0050

< 0.001 0.072

P value

Timing = (Ti·Ttot¡1). Sex coded for males = 1, females = 0

some degree to PETCO2. However, the association of PETCO2 with ventilatory control diminished as the intensity of the exercise increased. Drive The age-related decrease in drive/kg agrees with previous results (Gaultier et al. 1981). The question remains as to which component of ventilatory drive, VT or Ti, is most responsible for the age-related decrease and how are these related to PETCO2. At rest and for all intensities of exercise, VT kg¡1 decreased with age, while Ti increased with age. Generally, the decrease in VT kg¡1 was less than the increase in Ti (» 11 vs. » 42%). This would suggest that Ti has a larger inXuence on the changes in drive with age than does VT kg¡1. To answer the second part of the proposed question, we computed the correlations between PETCO2 and the

123

components of drive, adjusting for age and sex, and found that the correlations between PETCO2 and Ti were approximately twice as high as that for VT kg¡1 (r = 0.183 vs. 0.095). These correlations suggest that PETCO2 has a greater inXuence on the duration of inspiration than the tidal volume, and the inXuence of PETCO2 is quite small during exercise. Also, one cannot overlook the simple eVect of size on ventilatory control, because the children are becoming larger and size is known to inXuence the control of ventilation (Aitken et al. 1986). The ANOVA results suggested that in children and adolescents, when drive was adjusted for body mass, it decreased with age but did not diVer by sex. However, the regression analyses suggest that sex does have a small role at rest and during the run, accounting for < 2% of the variance in drive. Since our ANOVA results found lower PETCO2 values in the females compared to the males at all intensities (P < 0.03), this suggests that drive during exercise is partially attributable to sex diVerences in CO2. It appears that the females became more sensitive to CO2 than males around midpuberty, which resembles what is found in adults as women have lower PETCO2 values than men (Aitken et al. 1986). The combined eVect of age and intensity on drive is seen in Fig. 1. The ventilatory drive at rest was » 31% lower in the 18 year-olds compared to the 8 year-olds, while drive for running decreased by 46.6% when from age 8 to 18. These intensity-related changes in drive appeared to be impacted diVerentially by CO2. The relationship between PETCO2 and drive was approximately three times stronger at rest than during any of the exercise intensities (Table 6). This suggests that the role of PETCO2 in the control of drive during exercise diminishes with intensity, while the importance of neural input appears to increase (Eldridge 1994; Martin and Weil 1979). Another variable, VCO2, was signiWcantly correlated with drive in all conditions and the relationships became stronger as intensity increased from rest (r = 0.775) to running (r = 0.845). This may have been more related to metabolic demands increasing CO2 output, since PETCO2 did not change in a similar manner. Another way to examine the role of CO2 on ventilatory control is to evaluate CO2 sensitivity, which can be indirectly estimated from VE/VCO2 (Cooper et al. 1987). Regression analyses, adjusted for age and sex, indicated that drive was marginally associated with VE/VCO2, accounting for < 3% of the variance during rest and walking, and » 8% of the variance during running. This result suggests that for children and adolescents, CO2 has only a minor role in the control of ventilatory drive during exercise.

Eur J Appl Physiol (2006) 98:88–96

Timing Respiratory timing increased with exercise intensity and decreased slightly with age, regardless of sex. This Wnding has not been reported previously, but the lack of a sex-diVerence is credible since the percentage of change in Ti and Ttot was similar between females and males. Age was the major contributor in all regression models for timing, although the total R2 values were small and the year to year changes amounted to 0.04% or a » 5% change between an 8 and 18-year old. The age-related decline in timing was not supported in previous research, as others have reported no change (Gaultier et al. 1981; Mercier et al. 1991). However, our sample size was larger and contained a greater diversity of ages than previous studies, which probably contributed to our signiWcance. When considering the eVects of intensity, our results show that as the activity became more intense, timing increased signiWcantly, due to decreases in both Ti and Ttot. For example, comparing the resting and running values, Ti decreased by 56%, while Ttot declined by 63%, on average. This Wnding concurs with a similar study in children (Boule et al. 1989). Many factors may play a role in the control of respiratory timing. Milic-Emili and Grunstein (1976) suggested that timing is inXuenced by CO2 levels. Thus, we examined the relationship between PETCO2 and timing. The correlations between timing and PETCO2 were signiWcant, although generally weak (r = 0.193 at rest, 0.131 for the slow walk) and became negative during the run (r = ¡0.176). When examining the inXuence of VCO2 on timing, signiWcant negative correlations were found at rest (r = ¡0.184) and during the run (r = ¡0.128). These relationships actually accounted for less than 4% of the total variance. Further analyses of the VE/VCO2 ratio (an indirect estimate of CO2 sensitivity) and timing resulted in no signiWcant correlations (P > 0.08). Thus, CO2 appears to have limited importance to the control of timing.

95

increased as the exercise became more intense; consistent with previous research (Robinson 1938; Rowland 1996). The VE kg¡1 and VT kg¡1 also decreased with age, and increased with intensity (Gratas-Delamarche et al. 1993; Robinson 1938; Rowland 1996). This agerelated decrease in VT kg¡1 is in opposition to previous works of Rowland and Cunningham (1997) who reported no change. This dissimilarity is not well understood, but may be due to diVerences in the participants and study design, as Rowland and Cunningham (1997) studied only 20 children, but followed them longitudinally for 5 years. Limitations A possible limitation of our study was the use of absolute, rather than relative exercise intensities. Each workload may have elicited greater ventilatory responses in the younger and smaller participants, compared to the older and larger participants. Also, the mass of the COSMED unit (» 800 g) may have a greater inXuence on the VO2 of the smaller children. However, research has shown that imposing a vertical load of 5–10% of one’s body mass does not change children’s VO2 while running on a treadmill (Cooke et al. 1991). In the current study, the additional weight of the equipment ranged from <1 to 3% of the participants’ body weight and thus, its inXuences were minimal. Another limitation was the lack of control for menstrual cycle phase in the pubertal female participants. Studies have suggested that at rest, minute ventilation is slightly higher during the luteal phase of the menstrual cycle, due to an increased inspiratory drive (Das 1998; White et al. 1983); however, the majority of studies found no eVects of the menstrual cycle on minute ventilation at rest or during exercise (Dombovy et al. 1987; Jurkowski et al. 1981; Lebrun et al. 1995; Matsuo et al. 2003).

Conclusion Other ventilatory measures Although the timing and drive of the ventilatory pattern were similar between the sexes, other ventilatory parameters diVered. The absolute values for VE, VO2 and VT increased with age and exercise intensity in males and females as expected (McMurray et al. 2003; Robinson 1938; Rowland 1996). The VE/VO2 decreased as the children got older and as the exercise became more intense, which is in agreement with previous Wndings (McMurray et al. 2003; Rowland 1996). The respiratory frequency (fR), decreased with age, but

With children and adolescents, CO2 appears to have signiWcant, but minor roles in drive and timing, and its inXuence decreases from rest to high intensity submaximal exercise. At rest PETCO2 contributed to both drive and timing, but as the exercise intensity increased, the inXuence of PETCO2 diminished; suggesting the importance of neural drive as the speed of ambulation increases. Drive and timing were minimally (< 2%) diVerent between the sexes, even though the females became more sensitive to CO2 than the males during adolescence. On the other hand drive,

123

96

Eur J Appl Physiol (2006) 98:88–96

and to a minor extent, timing decreased with age. These results suggest that in youth there are agerelated changes in the signiWcance of CO2 in the control of inspiratory drive and timing during exercise. Acknowledgment # NR04564.

This research was supported by NINR Grant

References Aitken ML, Franklin JL, Pierson DJ, Schoene RB (1986) InXuence of body size and gender on control of ventilation. J Appl Physiol 60(6):1894–1899 Armon Y, Cooper DM, Zanconato S (1991) Maturation of ventilatory responses to 1-minute exercise. Pediatr Res 29:362– 368 Askanazi J, Milic-Emili J, Broell JR, Hyman AI, Kinney JM (1979) InXuence of exercise and CO2 on breathing pattern of normal man. J Appl Physiol 47(1):192–196 Bainton CR (1972) EVect of speed vs. grade and shivering on ventilation in dogs during active exercise. J Appl Physiol 33(6):778–787 Boule M, Gaultier C, Girard F (1989) Breathing pattern during exercise in untrained children. Respir Physiol 75:225–234 Cooke CB, McDonagh MJN, Nevill AM, Davies CTM (1991) EVects of load on oxygen intake in trained boys and men during treadmill running. J Appl Physiol 71(4):1237–1244 Cooper DM, Kaplan MR, Baumgarten L, Weiler-Ravell D, Whipp BJ, Wasserman K (1987) Coupling of ventilation and CO2 production during exercise in children. Pediatr Res 21:568–572 Das TK (1998) EVects of the menstrual cycle on timing and depth of breathing at rest. Indian J Physiol Pharmacol 42(4):498– 502 Dombovy ML, Bonekat HW, Williams TJ, Staats BA (1987) Exercise performance and ventilatory response in the menstrual cycle. Med Sci Sports Exerc 19(2):111–117 Eldridge FL (1994) Central integration of mechanisms in exercise hyperpnea. Med Sci Sports Exerc 26(3):319–327 Gaultier C, Perret L, Boule M, Buvry A, Girard F (1981) Occlusion pressure and breathing pattern in healthy children. Respir Physiol 46:71–80 Gratas-Delamarche A, Mercier J, Ramonatxo M, Dassonville J, Prefaut C (1993) Ventilatory response of prepubertal boys and adults to carbon dioxide at rest and during exercise. Eur J Appl Physiol 66:25–30 Grunstein MM, Derenne JP, Milic-Emili J (1975) Control of depth and frequency of breathing during baroreceptor stimulation in cats. J Appl Physiol 39(3):395–404

123

Harrell JS, McMurray RG, Baggett CD, Pennell ML, Pearce PF, Bangdiwala SI (2005) Energy costs of physical activities in children and adolescents. Med Sci Sports Exerc 37(2):329– 336 Jurkowski JEH, Jones NL, Toews CJ, Sutton JR (1981) EVects of menstrual cycle on blood lactate, O2 delivery, and performance during exercise. J Appl Physiol 51(6):1493–1499 Lanteri CJ, Sly PD (1993) Changes in respiratory mechanics with age. J Appl Physiol 74(1):369–378 Lebrun CM, McKenzie DC, Prior JC, Taunton JE (1995) EVects of menstrual cycle phase on athletic performance. Med Sci Sports Exerc 27(3):437–444 Martin BJ, Weil JV (1979) CO2 and exercise tidal volume. J Appl Physiol 46(2):322–325 Matsuo H, Katayama K, Ishida K, Muramatsu T, Miyamura M (2003) EVect of menstrual cycle and gender on ventilatory and heart rate responses at the onset of exercise. Eur J Appl Physiol 90:100–108 McMurray RG, Ahlborn SW (1982) Respiratory responses to running and walking at the same metabolic rate. Respira Physiol 47:257–265 McMurray RG, Baggett C, Pennell M, Bangdiwala S, Harrell J (2003) Gender diVerences in ventilatory responses of youth are related to exercise intensity. Port J Sport Sci 3:101–102 Mercier J, Varray A, Ramonatxo M, Mercier B, Prefaut C (1991) InXuence of anthropometric characteristics on changes in maximal exercise ventilation and breathing pattern during growth in boys. Eur J Appl Physiol 63:235–241 Milic-Emili J (1982) Recent advances in clinical assessment of control of breathing. Lung 160:1–17 Milic-Emili J, Grunstein MM (1976) Drive and timing components of ventilation. Chest 70(1 Suppl):131–133 Nagano Y, Baba R, Kuraishi K, Yasuda T, Ikoma M, Nishibata K, Yokota M, Nagashima M (1998) Ventilatory control during exercise in normal children. Pediatr Res 43:704–707 Robinson S (1938) Experimental studies of physical Wtness in relation to age. Arbeitsphysiologie 10:251–323 Rowland TW (1996) Response to endurance exercise: ventilation. In: Gilly H, Johnson C, Blakley J, Hooper L (eds) Developmental exercise physiology. Human Kinetics, Champaign, pp 141–157 Rowland TW, Cunningham LN (1997) Development of ventilatory responses to exercise in normal white children. Chest 111:327–332 Sato Y, Katayama K, Ishida K, Miyamura M (2000) Ventilatory and circulatory responses at the onset of voluntary exercise and passive movement in children. Eur J Appl Physiol 83:516–523 White DP, Douglas NJ, Pickett CK, Weil JV, Zwillich CW (1983) Sexual inXuence on the control of breathing. J Appl Physiol 54(4):874–879