Eur J Appl Physiol DOI 10.1007/s00421-015-3192-y
ORIGINAL ARTICLE
Heart rate recovery and parasympathetic modulation in boys and girls following maximal and submaximal exercise J. P. Guilkey1 · M. Overstreet1 · A. D. Mahon1
Received: 13 October 2014 / Accepted: 18 May 2015 © Springer-Verlag Berlin Heidelberg 2015
Abstract Purpose This study examined heart rate recovery (HRR) and heart rate variability (HRV) following submaximal and maximal exercise in boys (n = 13; 10.1 ± 0.8 years) and girls (n = 12; 10.1 ± 0.7 years). Methods Participants completed 10 min of supine rest followed by a graded exercise test to maximal effort. On a separate day, participants performed submaximal exercise at ventilatory threshold. Immediately following both exercise bouts, 1-min HRR was assessed in the supine position. HRV variables were analyzed under controlled breathing in the time and frequency domains over the final 5 min of rest and recovery. Results There were no significant differences in HRR following maximal and submaximal exercise between boys (58 ± 8 and 59 ± 8 beats min−1, respectively) and girls (54 ± 6 and 52 ± 19 beats min−1, respectively). There also were no significant interactions between groups from rest to recovery from maximal exercise for any HRV variables. However, there was a difference in the response between sexes from rest to recovery from submaximal exercise for log transformed standard deviation of NN intervals (lnSDNN) and log transformed total power (lnTP). No differences were observed for lnSDNN at rest (boys = 4.61 ± 0.28 vs. girls = 4.28 ± 0.52 ms) or during recovery (lnSDNN: boys 3.78 ± 0.46 vs. girls 3.87 ± 0.64 ms and lnTP: boys 7.33 ± 1.09 vs. girls; 7.44 ± 1.24 ms2). Post hoc pairwise comparisons showed
a significant difference between boys and girls for lnTP at rest (boys = 9.14 ± 0.42 vs. girls = 8.30 ± 1.05 ms2). Conclusion Parasympathetic modulation was similar between boys and girls at rest and during recovery from exercise, which could explain similarities observed in HRR. Keywords Cardiovascular regulation · Heart rate variability · Pediatrics · Sex differences Abbreviations HF High-frequency power HFnu High-frequency power normalized units HR Heart rate HRR Heart rate recovery HRV Heart rate variability ln Natural log pNN50 Percentage of consecutive NN intervals differing by more than 50 ms PNS Parasympathetic nervous system RER Respiratory exchange ratio RMSSD Root mean square of successive RR intervals RPE Ratings of perceived exertion SDNN Standard deviation of NN intervals TP Total power VE Pulmonary ventilation VO2 Oxygen consumption VO2max Maximal oxygen consumption VT Ventilatory threshold W Watts
Communicated by Massimo Pagani. * J. P. Guilkey [email protected] 1
Human Performance Laboratory, Ball State University, Muncie, IN 47303, USA
Introduction Heart rate recovery (HRR) has been found to be greater in boys compared to girls following maximal exercise (Pels
13
et al. 1981; Washington et al. 1988). Pels et al. (1981) showed that young (~7.5 years) boys had a greater decrease in heart rate (HR) during the first 3 min of recovery compared to girls of a similar age. Similarly, Washington et al. (1988) found 1-min HRR greater in older (7–13 years) boys compared to girls, regardless of body size. More recently, Singh et al. (2008) used a multivariable linear regression model and showed sex was a significant predictor of 1-min HRR following maximal exercise in a large cohort of healthy children between 9 and 18 years old. Following submaximal exercise, Mahon et al. (2003) observed a greater HR in girls compared to boys 1 min following an absolute intensity (70 W), suggesting greater HRR in boys. However, this difference in HRR was eradicated following submaximal exercise at a relative intensity (85–90 % of VO2max). The mechanism responsible for these observed differences in HRR between boys and girls is presently unclear. Pels et al. (1981) and Washington et al. (1988) suggested faster HRR in boys could be due to greater cardiorespiratory fitness. Cardiorespiratory fitness effects on HRR is well documented in adults (Aubert et al. 2003; Darr et al. 1988; Hagberg et al. 1980; Imai et al. 1994) and to a lesser extent in children (Singh et al. 2008), in such a manner that greater cardiorespiratory fitness results in greater HRR. Contrary to this notion, the study by Mahon et al. (2003) showed resting HR, not cardiorespiratory fitness, accounted for a greater percentage of variation in HRR in their population. This suggests that autonomic regulation of resting HR, particularly the parasympathetic nervous system (PNS), could be a primary influence of HRR, which has been previously proposed by Ohuchi et al. (2000). Relatively few studies have examined sex differences in autonomic regulation of HR at rest in children and the findings between studies have been equivocal. One study observed no sex differences (Goto et al. 1997), others have found an overall sex difference independent of age (Faulkner et al. 2003; Umetani et al. 1998), and another study found an age-dependent difference (Silvetti et al. 2001). Recently, Michels et al. (2013) studied a large cohort of children and found that boys had greater PNS modulation [high-frequency power (HF), root mean square of successive RR intervals (RMSSD), and percentage of consecutive NN intervals differing by more than 50 ms (pNN50)] at rest than girls. The reasons for the variations in findings between studies are currently not clear. Moreover, these variable findings could suggest that resting PNS modulation may not fully account for sex differences in HRR in children, as originally suggested by Mahon et al. (2003). HRR is mediated by the interaction of PNS reactivation and sympathetic withdrawal. In adults, it has been proposed that measuring autonomic modulation during recovery from maximal exercise, compared to rest, may better explain HRR (Javorka et al. 2002), although this has
13
Eur J Appl Physiol
been untested in children. Therefore, measuring autonomic regulation during recovery could give greater insight into the mechanism of sex differences of HRR in children. There is little known about the influence of sex on the autonomic response following exercise. In adults, Mendonca et al. (2010) showed no differences in PNS modulation (HF and detrended fluctuation analysis) during recovery from a Wingate test, although differences were present at rest. To our knowledge, there have been no studies that have examined the PNS response between boys and girls following exercise. In adults, slowed HRR has shown to independently increase risk of mortality (Cole et al. 1999), which is thought to be due to changes in the PNS responsiveness following exercise (Davrath et al. 2006). Moreover, recent evidence has suggested a relationship between slowed HRR and cardiometabolic risk factors in children and adolescents (Laguna et al. 2013; Simhaee et al. 2013). Therefore, it may be important to understand the mechanism that mediates sex differences in HRR to guide pointed preventative strategies in childhood and adolescence. The purpose of this study was to examine HRR and PNS modulation during the recovery from maximal and submaximal exercise in boys and girls. PNS modulation was measured at rest and during recovery from exercise via heart rate variability (HRV) (Goldberger et al. 2006; Kannankeril et al. 2004), which has been previously used to assess PNS modulation (Camm et al. 1996; Kleiger et al. 1992). For this study, it was hypothesized that boys would have greater HRR following maximal exercise, but similar HRR following a relative intensity submaximal exercise bout. Secondly, it was hypothesized that HRV variables would reflect greater PNS modulation in boys during recovery from maximal exercise and similar PNS modulation during recovery from submaximal exercise, corresponding to differences in HRR.
Methods Participants Twelve girls between 9 and 11 years of age were recruited to participate in this study. Additionally, 13 boys who participated in a previous study (Guilkey et al. 2014) were included for comparison purposes. All children who participated were apparently healthy and recreationally active, but not well trained, based on their responses on a health history questionnaire. Pubertal status was determined and based on parental assessment of pubic hair growth based on the criteria established by Tanner (1962). Participants were excluded if they had any cardiorespiratory or metabolic diseases or were taking medications that are known to alter cardiac function. Prior to participation in the study,
Eur J Appl Physiol
child participants gave written assent and parents gave written consent. This study was approved by the Institutional Review Board at Ball State University. Procedures The first visit to the laboratory was used to familiarize participants to all aspects of the study. Upon arrival to the laboratory, the children practiced paced breathing (0.25 Hz) that was utilized in subsequent visits to control for respiratory sinus arrhythmia, which can confound HRV analysis (Brown et al. 1993). Furthermore, the children were oriented to the OMNI rating of perceived exertion (RPE) scale. Participants also completed a three-stage stepwise incremental exercise test on an electronically braked cycle ergometer. Oxygen consumption (VO2) was measured continuously during the exercise test. The VO2 during the final minute of each stage was averaged to develop a VO2–work rate relationship, which was used to establish a work rate during the third visit. The second and third visits took place between 7:00 and 10:00 am. The participants were fasted for at least 10 h and abstained from vigorous physical activity for at least 12 h, as confirmed verbally. The second visit consisted of a 10-min supine resting period to determine the participants resting HR and HRV. Following the resting period, participants completed a graded exercise test to maximal effort. This test started at 20 W for 2 min and the work rate increased by 10 W every minute until a maximal voluntary effort was achieved. HR, using a Polar monitor, and pulmonary gas exchange, using a mouthpiece breathing valve and nose clip, were measured continuously throughout the test. RPE was recorded after the first 2 min and every minute thereafter. In lieu of a plateau in VO2, secondary criteria were used to establish the achievement of a maximal effort. The criteria used were as follows: (1) failure to maintain a pedal rate >50 rpm; (2) respiratory exchange ratio (RER) ≥1.00; (3) OMNI RPE scale ≥8 and (4) peak HR at the termination of test ≥95 % of age predicted maximal (208 − 0.7 × age). If two of the four criteria were met, maximal exercise was considered to have been achieved (Guilkey et al. 2014; Rogowski et al. 2012). Within 30 s, following the conclusion of the graded exercise test, participants assumed a supine position on an examination table. The participants remained supine for 10 min, avoiding talking and excessive movement during this time period. On the third visit, participants arrived at the laboratory at approximately the same time of day as visit two and completed a 12-min exercise bout at an intensity equivalent to the VO2 (±5 %) at ventilatory threshold (VT). Prior to the exercise bout, participants warmed up for 3 min at a light intensity followed by exercise at the desired intensity (work load estimated from VO2–work rate relationship). If
necessary, the work load was adjusted in the first 6 min of the bout, in order to produce a VO2 within the acceptable range. HR and pulmonary gas exchange were measured continuously after the second minute of exercise and RPE was assessed at the sixth and twelfth minute of exercise. Following the 12 min of exercise, the recovery period followed the same protocol as the previous visit. VT determination Pulmonary ventilation (VE) was plotted against VO2 and a third-order polynomial trend line was fitted. The graphs were then analyzed by three independent investigators. VT was defined as the point during exercise at which there was a non-linear increase in VE in relation to VO2. The results of the three investigators were pooled and the coefficient of variation between the values was determined. If the coefficient of variation was less than 5 %, all three values were averaged; if the coefficient of variation was greater than 5 %, the two closest values were averaged (Anderson and Mahon 2007; Guilkey et al. 2014). HRR HRR from maximal exercise was determined as the absolute and percent difference in HR from the peak HR (greatest 30 s average) at the end of exercise to the HR at the first minute of recovery (HRR = peak HR − HR after 1 min recovery; HRR % Δ = absolute HRR/peak HR × 100). HRR from submaximal exercise was determined as the absolute and percent difference in HR from the end of exercise (average HR over final minute) and the HR at the first minute of recovery (HRR = HR last 60 s − HR after 1 min recovery; HRR % Δ = absolute HRR/HR last 60 s × 100). Percent change was utilized to account for differences in HR observed at maximal and submaximal exercise between boys and girls (Turley 1997). HRV An electrocardiogram was continuously recorded during the entire 10 min of the resting period and for 10 min following both exercise bouts at a sampling frequency of 1000 Hz using a Biopac MP35 system. Participants adhered to a breathing frequency of 15 breaths per minute (0.25 Hz) for the entire resting period and the final 6 min of each recovery period to control for respiratory sinus arrhythmia (Fleming et al. 2011; Guilkey et al. 2014). The final 5 min of the electrocardiogram recording during the resting and recovery periods were used for the HRV analysis, in order to ensure stationarity of the signal (Camm et al. 1996). This approach has been done in previous studies in adults (Javorka et al. 2002; Mendonca et al. 2010) and children (Goulopoulou et al. 2006) during
13
recovery from exercise with no stated violations of stationarity in the HR data. The electrocardiogram recording was imported into WinCPRS software for analysis of HRV in the time and frequency domain. R-waves were automatically identified. In addition, R-waves were manually inspected to ensure proper identification and inclusion in the analysis. Moreover, ectopic heart beats were identified and corrected using the interpolation method (Lippman et al. 1994). Standard deviation of NN intervals (SDNN), RMSSD, and pNN50 were determined in the time domain. SDNN has been shown to be an index of overall variation of HR, but it has been suggested during short-term signals the variations between heart beats is primarily due to the PNS (Goldberger et al. 2001; Kleiger et al. 1992; Shaffer et al. 2014). However, this view is not universally accepted and others view SDNN as an indicator of global autonomic modulation (Camm et al. 1996). In the present study, SDNN will be used to give insight into PNS modulation at rest and during recovery from exercise, as it has been used in previous studies of this nature. RMSSD and pNN50 will be used as measures of PNS modulation at rest and during recovery from exercise (Camm et al. 1996; Kleiger et al. 1992). In the frequency domain, the fast Fourier transform was used to determine the power spectrum of HRV. The power spectrum was then analyzed for total power (TP) (0.0–0.4 Hz), which give insights in global autonomic modulation (Camm et al. 1996). Additionally, HF (0.15– 0.4 Hz) was determined and is agreed upon as a measure of PNS modulation. HF was normalized to TP minus very low frequency (HFnu), which also represents PNS modulation. The low-frequency components of HRV and the ratio of low-frequency power and HF were excluded from this study because these variables do not give insight into PNS modulation (Camm et al. 1996).
Eur J Appl Physiol Table 1 Subject characteristics
Age (years) Height (cm) Weight (kg)
Girls
Boys
(n = 12)
(n = 13)
10.1 ± 0.7 142.5 ± 8.5
10.1 ± 0.8 144.2 ± 6.9
38.7 ± 12.2
36.9 ± 9.0
Values presented as mean ± SD
responses. At both exercise intensities, the Wilks’ Lambda showed significant effects (p 0.05). Ventilatory threshold VT was determined on 24 of the 25 participants and indeterminate on one boy, who was excluded from the
Eur J Appl Physiol Table 2 Physiological and perceptual responses at maximal exercise Girls (n = 12)
Boys (n = 13)
VO2 (L min−1)
1.51 ± 0.32
1.63 ± 0.32
VO2 (ml kg−1 min−1) HR (beats min−1) RER
40.1 ± 4.8
45.6 ± 4.8a
193 ± 9 1.05 ± 0.06
191 ± 8 1.07 ± 0.05
9.3 ± 1.7
9.2 ± 0.9
RPE Values presented as mean ± SD
HR heart rate, RER respiratory exchange ratio, RPE ratings of perceived exertion
Heart rate recovery Following maximal exercise, HRR was similar between sexes (p > 0.05) (Fig. 1). The percent decline of HR during the first minute of recovery in boys was 30.3 ± 5.0 % and in girls was 28.2 ± 4.2 %, which also was similar (p > 0.05). Additionally, there was no significant difference between the groups in HRR following submaximal exercise (Fig. 1). When expressed as percent decline of HR, the boys recovered by 38.0 ± 6.0 %. Whereas, the girls recovered by 34.3 ± 11.2 %. These differences were not significant.
a
Different from girls (p 0.05). The HR was similar between the boys and girls (154 ± 13 vs. 151 ± 20 beats min−1) (p > 0.05). Likewise, the RER and RPE were not significantly different between boys (0.90 ± 0.04 and 4.8 ± 1.7, respectively) and girls (0.91 ± 0.04 and 4.5 ± 2.1, respectively).
Time and frequency domain variables at rest and during recovery from maximal and submaximal exercise are shown in Tables 4 and 5, respectively. For the maximal exercise responses, there were no significant interactions between sex and exercise for any of the time and frequency domain variables. In addition, there were no main effects of sex for any of the variables (p > 0.05). However, there were significant main effects for exercise for each variable. Specifically, lnSDNN, lnRMSSD, pNN50, lnTP, lnHF, and lnHFnu were all greater at rest than during recovery from maximal exercise (p recovery (main effect of exercise, p
ORIGINAL ARTICLE
Heart rate recovery and parasympathetic modulation in boys and girls following maximal and submaximal exercise J. P. Guilkey1 · M. Overstreet1 · A. D. Mahon1
Received: 13 October 2014 / Accepted: 18 May 2015 © Springer-Verlag Berlin Heidelberg 2015
Abstract Purpose This study examined heart rate recovery (HRR) and heart rate variability (HRV) following submaximal and maximal exercise in boys (n = 13; 10.1 ± 0.8 years) and girls (n = 12; 10.1 ± 0.7 years). Methods Participants completed 10 min of supine rest followed by a graded exercise test to maximal effort. On a separate day, participants performed submaximal exercise at ventilatory threshold. Immediately following both exercise bouts, 1-min HRR was assessed in the supine position. HRV variables were analyzed under controlled breathing in the time and frequency domains over the final 5 min of rest and recovery. Results There were no significant differences in HRR following maximal and submaximal exercise between boys (58 ± 8 and 59 ± 8 beats min−1, respectively) and girls (54 ± 6 and 52 ± 19 beats min−1, respectively). There also were no significant interactions between groups from rest to recovery from maximal exercise for any HRV variables. However, there was a difference in the response between sexes from rest to recovery from submaximal exercise for log transformed standard deviation of NN intervals (lnSDNN) and log transformed total power (lnTP). No differences were observed for lnSDNN at rest (boys = 4.61 ± 0.28 vs. girls = 4.28 ± 0.52 ms) or during recovery (lnSDNN: boys 3.78 ± 0.46 vs. girls 3.87 ± 0.64 ms and lnTP: boys 7.33 ± 1.09 vs. girls; 7.44 ± 1.24 ms2). Post hoc pairwise comparisons showed
a significant difference between boys and girls for lnTP at rest (boys = 9.14 ± 0.42 vs. girls = 8.30 ± 1.05 ms2). Conclusion Parasympathetic modulation was similar between boys and girls at rest and during recovery from exercise, which could explain similarities observed in HRR. Keywords Cardiovascular regulation · Heart rate variability · Pediatrics · Sex differences Abbreviations HF High-frequency power HFnu High-frequency power normalized units HR Heart rate HRR Heart rate recovery HRV Heart rate variability ln Natural log pNN50 Percentage of consecutive NN intervals differing by more than 50 ms PNS Parasympathetic nervous system RER Respiratory exchange ratio RMSSD Root mean square of successive RR intervals RPE Ratings of perceived exertion SDNN Standard deviation of NN intervals TP Total power VE Pulmonary ventilation VO2 Oxygen consumption VO2max Maximal oxygen consumption VT Ventilatory threshold W Watts
Communicated by Massimo Pagani. * J. P. Guilkey [email protected] 1
Human Performance Laboratory, Ball State University, Muncie, IN 47303, USA
Introduction Heart rate recovery (HRR) has been found to be greater in boys compared to girls following maximal exercise (Pels
13
et al. 1981; Washington et al. 1988). Pels et al. (1981) showed that young (~7.5 years) boys had a greater decrease in heart rate (HR) during the first 3 min of recovery compared to girls of a similar age. Similarly, Washington et al. (1988) found 1-min HRR greater in older (7–13 years) boys compared to girls, regardless of body size. More recently, Singh et al. (2008) used a multivariable linear regression model and showed sex was a significant predictor of 1-min HRR following maximal exercise in a large cohort of healthy children between 9 and 18 years old. Following submaximal exercise, Mahon et al. (2003) observed a greater HR in girls compared to boys 1 min following an absolute intensity (70 W), suggesting greater HRR in boys. However, this difference in HRR was eradicated following submaximal exercise at a relative intensity (85–90 % of VO2max). The mechanism responsible for these observed differences in HRR between boys and girls is presently unclear. Pels et al. (1981) and Washington et al. (1988) suggested faster HRR in boys could be due to greater cardiorespiratory fitness. Cardiorespiratory fitness effects on HRR is well documented in adults (Aubert et al. 2003; Darr et al. 1988; Hagberg et al. 1980; Imai et al. 1994) and to a lesser extent in children (Singh et al. 2008), in such a manner that greater cardiorespiratory fitness results in greater HRR. Contrary to this notion, the study by Mahon et al. (2003) showed resting HR, not cardiorespiratory fitness, accounted for a greater percentage of variation in HRR in their population. This suggests that autonomic regulation of resting HR, particularly the parasympathetic nervous system (PNS), could be a primary influence of HRR, which has been previously proposed by Ohuchi et al. (2000). Relatively few studies have examined sex differences in autonomic regulation of HR at rest in children and the findings between studies have been equivocal. One study observed no sex differences (Goto et al. 1997), others have found an overall sex difference independent of age (Faulkner et al. 2003; Umetani et al. 1998), and another study found an age-dependent difference (Silvetti et al. 2001). Recently, Michels et al. (2013) studied a large cohort of children and found that boys had greater PNS modulation [high-frequency power (HF), root mean square of successive RR intervals (RMSSD), and percentage of consecutive NN intervals differing by more than 50 ms (pNN50)] at rest than girls. The reasons for the variations in findings between studies are currently not clear. Moreover, these variable findings could suggest that resting PNS modulation may not fully account for sex differences in HRR in children, as originally suggested by Mahon et al. (2003). HRR is mediated by the interaction of PNS reactivation and sympathetic withdrawal. In adults, it has been proposed that measuring autonomic modulation during recovery from maximal exercise, compared to rest, may better explain HRR (Javorka et al. 2002), although this has
13
Eur J Appl Physiol
been untested in children. Therefore, measuring autonomic regulation during recovery could give greater insight into the mechanism of sex differences of HRR in children. There is little known about the influence of sex on the autonomic response following exercise. In adults, Mendonca et al. (2010) showed no differences in PNS modulation (HF and detrended fluctuation analysis) during recovery from a Wingate test, although differences were present at rest. To our knowledge, there have been no studies that have examined the PNS response between boys and girls following exercise. In adults, slowed HRR has shown to independently increase risk of mortality (Cole et al. 1999), which is thought to be due to changes in the PNS responsiveness following exercise (Davrath et al. 2006). Moreover, recent evidence has suggested a relationship between slowed HRR and cardiometabolic risk factors in children and adolescents (Laguna et al. 2013; Simhaee et al. 2013). Therefore, it may be important to understand the mechanism that mediates sex differences in HRR to guide pointed preventative strategies in childhood and adolescence. The purpose of this study was to examine HRR and PNS modulation during the recovery from maximal and submaximal exercise in boys and girls. PNS modulation was measured at rest and during recovery from exercise via heart rate variability (HRV) (Goldberger et al. 2006; Kannankeril et al. 2004), which has been previously used to assess PNS modulation (Camm et al. 1996; Kleiger et al. 1992). For this study, it was hypothesized that boys would have greater HRR following maximal exercise, but similar HRR following a relative intensity submaximal exercise bout. Secondly, it was hypothesized that HRV variables would reflect greater PNS modulation in boys during recovery from maximal exercise and similar PNS modulation during recovery from submaximal exercise, corresponding to differences in HRR.
Methods Participants Twelve girls between 9 and 11 years of age were recruited to participate in this study. Additionally, 13 boys who participated in a previous study (Guilkey et al. 2014) were included for comparison purposes. All children who participated were apparently healthy and recreationally active, but not well trained, based on their responses on a health history questionnaire. Pubertal status was determined and based on parental assessment of pubic hair growth based on the criteria established by Tanner (1962). Participants were excluded if they had any cardiorespiratory or metabolic diseases or were taking medications that are known to alter cardiac function. Prior to participation in the study,
Eur J Appl Physiol
child participants gave written assent and parents gave written consent. This study was approved by the Institutional Review Board at Ball State University. Procedures The first visit to the laboratory was used to familiarize participants to all aspects of the study. Upon arrival to the laboratory, the children practiced paced breathing (0.25 Hz) that was utilized in subsequent visits to control for respiratory sinus arrhythmia, which can confound HRV analysis (Brown et al. 1993). Furthermore, the children were oriented to the OMNI rating of perceived exertion (RPE) scale. Participants also completed a three-stage stepwise incremental exercise test on an electronically braked cycle ergometer. Oxygen consumption (VO2) was measured continuously during the exercise test. The VO2 during the final minute of each stage was averaged to develop a VO2–work rate relationship, which was used to establish a work rate during the third visit. The second and third visits took place between 7:00 and 10:00 am. The participants were fasted for at least 10 h and abstained from vigorous physical activity for at least 12 h, as confirmed verbally. The second visit consisted of a 10-min supine resting period to determine the participants resting HR and HRV. Following the resting period, participants completed a graded exercise test to maximal effort. This test started at 20 W for 2 min and the work rate increased by 10 W every minute until a maximal voluntary effort was achieved. HR, using a Polar monitor, and pulmonary gas exchange, using a mouthpiece breathing valve and nose clip, were measured continuously throughout the test. RPE was recorded after the first 2 min and every minute thereafter. In lieu of a plateau in VO2, secondary criteria were used to establish the achievement of a maximal effort. The criteria used were as follows: (1) failure to maintain a pedal rate >50 rpm; (2) respiratory exchange ratio (RER) ≥1.00; (3) OMNI RPE scale ≥8 and (4) peak HR at the termination of test ≥95 % of age predicted maximal (208 − 0.7 × age). If two of the four criteria were met, maximal exercise was considered to have been achieved (Guilkey et al. 2014; Rogowski et al. 2012). Within 30 s, following the conclusion of the graded exercise test, participants assumed a supine position on an examination table. The participants remained supine for 10 min, avoiding talking and excessive movement during this time period. On the third visit, participants arrived at the laboratory at approximately the same time of day as visit two and completed a 12-min exercise bout at an intensity equivalent to the VO2 (±5 %) at ventilatory threshold (VT). Prior to the exercise bout, participants warmed up for 3 min at a light intensity followed by exercise at the desired intensity (work load estimated from VO2–work rate relationship). If
necessary, the work load was adjusted in the first 6 min of the bout, in order to produce a VO2 within the acceptable range. HR and pulmonary gas exchange were measured continuously after the second minute of exercise and RPE was assessed at the sixth and twelfth minute of exercise. Following the 12 min of exercise, the recovery period followed the same protocol as the previous visit. VT determination Pulmonary ventilation (VE) was plotted against VO2 and a third-order polynomial trend line was fitted. The graphs were then analyzed by three independent investigators. VT was defined as the point during exercise at which there was a non-linear increase in VE in relation to VO2. The results of the three investigators were pooled and the coefficient of variation between the values was determined. If the coefficient of variation was less than 5 %, all three values were averaged; if the coefficient of variation was greater than 5 %, the two closest values were averaged (Anderson and Mahon 2007; Guilkey et al. 2014). HRR HRR from maximal exercise was determined as the absolute and percent difference in HR from the peak HR (greatest 30 s average) at the end of exercise to the HR at the first minute of recovery (HRR = peak HR − HR after 1 min recovery; HRR % Δ = absolute HRR/peak HR × 100). HRR from submaximal exercise was determined as the absolute and percent difference in HR from the end of exercise (average HR over final minute) and the HR at the first minute of recovery (HRR = HR last 60 s − HR after 1 min recovery; HRR % Δ = absolute HRR/HR last 60 s × 100). Percent change was utilized to account for differences in HR observed at maximal and submaximal exercise between boys and girls (Turley 1997). HRV An electrocardiogram was continuously recorded during the entire 10 min of the resting period and for 10 min following both exercise bouts at a sampling frequency of 1000 Hz using a Biopac MP35 system. Participants adhered to a breathing frequency of 15 breaths per minute (0.25 Hz) for the entire resting period and the final 6 min of each recovery period to control for respiratory sinus arrhythmia (Fleming et al. 2011; Guilkey et al. 2014). The final 5 min of the electrocardiogram recording during the resting and recovery periods were used for the HRV analysis, in order to ensure stationarity of the signal (Camm et al. 1996). This approach has been done in previous studies in adults (Javorka et al. 2002; Mendonca et al. 2010) and children (Goulopoulou et al. 2006) during
13
recovery from exercise with no stated violations of stationarity in the HR data. The electrocardiogram recording was imported into WinCPRS software for analysis of HRV in the time and frequency domain. R-waves were automatically identified. In addition, R-waves were manually inspected to ensure proper identification and inclusion in the analysis. Moreover, ectopic heart beats were identified and corrected using the interpolation method (Lippman et al. 1994). Standard deviation of NN intervals (SDNN), RMSSD, and pNN50 were determined in the time domain. SDNN has been shown to be an index of overall variation of HR, but it has been suggested during short-term signals the variations between heart beats is primarily due to the PNS (Goldberger et al. 2001; Kleiger et al. 1992; Shaffer et al. 2014). However, this view is not universally accepted and others view SDNN as an indicator of global autonomic modulation (Camm et al. 1996). In the present study, SDNN will be used to give insight into PNS modulation at rest and during recovery from exercise, as it has been used in previous studies of this nature. RMSSD and pNN50 will be used as measures of PNS modulation at rest and during recovery from exercise (Camm et al. 1996; Kleiger et al. 1992). In the frequency domain, the fast Fourier transform was used to determine the power spectrum of HRV. The power spectrum was then analyzed for total power (TP) (0.0–0.4 Hz), which give insights in global autonomic modulation (Camm et al. 1996). Additionally, HF (0.15– 0.4 Hz) was determined and is agreed upon as a measure of PNS modulation. HF was normalized to TP minus very low frequency (HFnu), which also represents PNS modulation. The low-frequency components of HRV and the ratio of low-frequency power and HF were excluded from this study because these variables do not give insight into PNS modulation (Camm et al. 1996).
Eur J Appl Physiol Table 1 Subject characteristics
Age (years) Height (cm) Weight (kg)
Girls
Boys
(n = 12)
(n = 13)
10.1 ± 0.7 142.5 ± 8.5
10.1 ± 0.8 144.2 ± 6.9
38.7 ± 12.2
36.9 ± 9.0
Values presented as mean ± SD
responses. At both exercise intensities, the Wilks’ Lambda showed significant effects (p 0.05). Ventilatory threshold VT was determined on 24 of the 25 participants and indeterminate on one boy, who was excluded from the
Eur J Appl Physiol Table 2 Physiological and perceptual responses at maximal exercise Girls (n = 12)
Boys (n = 13)
VO2 (L min−1)
1.51 ± 0.32
1.63 ± 0.32
VO2 (ml kg−1 min−1) HR (beats min−1) RER
40.1 ± 4.8
45.6 ± 4.8a
193 ± 9 1.05 ± 0.06
191 ± 8 1.07 ± 0.05
9.3 ± 1.7
9.2 ± 0.9
RPE Values presented as mean ± SD
HR heart rate, RER respiratory exchange ratio, RPE ratings of perceived exertion
Heart rate recovery Following maximal exercise, HRR was similar between sexes (p > 0.05) (Fig. 1). The percent decline of HR during the first minute of recovery in boys was 30.3 ± 5.0 % and in girls was 28.2 ± 4.2 %, which also was similar (p > 0.05). Additionally, there was no significant difference between the groups in HRR following submaximal exercise (Fig. 1). When expressed as percent decline of HR, the boys recovered by 38.0 ± 6.0 %. Whereas, the girls recovered by 34.3 ± 11.2 %. These differences were not significant.
a
Different from girls (p 0.05). The HR was similar between the boys and girls (154 ± 13 vs. 151 ± 20 beats min−1) (p > 0.05). Likewise, the RER and RPE were not significantly different between boys (0.90 ± 0.04 and 4.8 ± 1.7, respectively) and girls (0.91 ± 0.04 and 4.5 ± 2.1, respectively).
Time and frequency domain variables at rest and during recovery from maximal and submaximal exercise are shown in Tables 4 and 5, respectively. For the maximal exercise responses, there were no significant interactions between sex and exercise for any of the time and frequency domain variables. In addition, there were no main effects of sex for any of the variables (p > 0.05). However, there were significant main effects for exercise for each variable. Specifically, lnSDNN, lnRMSSD, pNN50, lnTP, lnHF, and lnHFnu were all greater at rest than during recovery from maximal exercise (p recovery (main effect of exercise, p
