Resting heart rate variability and heart rate recovery after - Springer Link

Clin Auton Res (2014) 24:53–61 DOI 10.1007/s10286-014-0225-2

RESEARCH ARTICLE

Resting heart rate variability and heart rate recovery after submaximal exercise Aljosˇa Danieli • Lara Lusa • Nejka Potocˇnik • Bernard Meglicˇ • Anton Grad • Fajko F. Bajrovic´

Received: 3 August 2013 / Accepted: 20 January 2014 / Published online: 8 February 2014 Ó Springer-Verlag Berlin Heidelberg 2014

Abstract Purpose Aerobic training accelerates Heart Rate Recovery after exercise in healthy subjects and in patients with coronary disease. As shown by pharmacological autonomic blockade, HRR early after exercise is dependent primarily on parasympathetic reactivation. Thus, accelerated HRR early after exercise in endurance-trained athletes may be attributed to augmented parasympathetic reactivation. In the present study, we tested the hypothesis that the HRR early after submaximal exercise is related to the pre-exercise parasympathetic modulation.

Electronic supplementary material The online version of this article (doi:10.1007/s10286-014-0225-2) contains supplementary material, which is available to authorized users. A. Danieli (&) B. Meglicˇ F. F. Bajrovic´ Department of Neurology, University Medical Centre, Zalosˇka 2, 1000 Ljubljana, Slovenia e-mail: [email protected] A. Danieli Barsos Medical Centre, Gregorcˇicˇeva 11, Ljubljana, Slovenia

Methods Thirty endurance-trained athletes (20 males, 50 ± 7 years) and thirty control subjects (20 males, 52 ± 6 years) performed a submaximal exercise on a cyclo-ergometer. Pre-exercise resting short-term heart rate variability (HRV) parameters in time and frequencydomains were correlated with HRR during the first 30 s, 1 and 2 min after cessation of exercise. Results We found that HRR was statistically significantly faster in athletes than in controls at all examination time points (p \ 0.05). HF, SDNN and RMSSD were statistically significantly higher in athletes than in controls (p \ 0.05), but other resting HRV parameters were not statistically different between groups. After 30 s, 1 and 2 min of recovery, HRR correlation with total power, HF, HFnu and RMSSD was positive, while the correlation with LF/HF was negative for small and positive for larger values. The opposite was true for SDNN. Conclusions These findings support the hypothesis that HRR early after submaximal exercise is related to resting parasympathetic modulation in the middle-aged subjects. In addition, they suggested an optimal range of HRV for maximal HRR after exercise.

L. Lusa Medical Faculty, Institute of Biostatistics and Medical Informatics, University of Ljubljana, Vrazov trg 2, Ljubljana, Slovenia

Keywords Autonomic nervous system Physical training Heart rate Aging

N. Potocˇnik Medical Faculty, Institute of Physiology, University of Ljubljana, Zalosˇka 4, Ljubljana, Slovenia

Introduction

A. Grad General Hospital Izola, Polje 40, Izola, Slovenia F. F. Bajrovic´ Medical Faculty, Institute of Pathophysiology, University of Ljubljana, Zalosˇka 4, Ljubljana, Slovenia

It has been recently shown that a delayed heart rate recovery (HRR) during the 1st or 2nd min after acute exercise is an independent predictor of overall mortality [1, 2]. Both cross-sectional and longitudinal studies have shown that aerobic endurance training accelerates HRR after exercise in healthy subjects [3, 4]. This is thought to

123

54

be due to enhanced intrinsic heart rate regulation, increased baroreceptor and metaboreceptor sensitivity, and improved autonomic balance after endurance training [5]. As shown by pharmacological autonomic blockade, HRR early after exercise is dependent primarily on parasympathetic reactivation [6]. Thus, accelerated HRR early after exercise in athletes may be attributed to augmented parasympathetic reactivation. Therefore, one can hypothesize that HRR early after exercise is related to the pre-exercise resting parasympathetic modulation [7]. A few available studies examining the correlation between HRR after exercise, with pre-exercise resting indices of heart rate variability (HRV), a non-invasive measure of the cardiac autonomic control, are not conclusive. HRR was positively correlated with resting HRV indices of parasympathetic modulation in middle-aged healthy subjects and in patients with coronary artery disease [8, 9]. However, HRR was not correlated with resting HRV in trained or untrained healthy young subjects [5, 10]. From these data, a clear relationship between the HRR after exercise and resting HRV cannot be elucidated and, therefore, further studies are needed. The aim of this cross-sectional study was to examine whether HRR early after exercise is related to pre-exercise resting HRV indices of parasympathetic modulation in middle-aged sedentary and endurance-trained subjects. We hypothesized first that HRR early after submaximal exercise and resting HRV parameters indicating parasympathetic activity are faster and higher, respectively, in athletes than in control subjects, and second, that HRR early after exercise is correlated with parasympathetic indices of HRV at rest.

Methods Subjects Sixty healthy subjects were recruited in this cross-sectional study, which was approved by the National Medical Ethics Committee of the Republic of Slovenia. All the subjects provided a written informed consent before participation. Their physical examination and history revealed no autonomic dysfunction, chronic diseases, medication usage or smoking. Their electrocardiogram (ECG) was normal. Subjects were assigned to two groups according to reported daily physical activities and Astrand nomogram, which is an adjusted calculation of maximal oxygen uptake (VO2max) from submaximal heart rate (HR), considering age, gender, height and maximal heart rate (HRmax) [11]. Individual HRmax was determined using the formula HRmax = 205.8 - (0.685 9 age) [12]. The group of athletes (ET; n = 30; 20 males; age range between 40 and 61 years) had sustained aerobic activities such as running

123

Clin Auton Res (2014) 24:53–61 Table 1 Basal characteristics of the group of endurance-trained athletes (ET) and the group of sedentary control subjects (C)

Age (years) Weight (kg)

ET group (n = 30; 20 males)

C group (n = 30; 20 males)

p value

50 (±7)

52 (±6)

0.38

83 (±9)

\0.001

70 (±10)

Height (cm)

175 (±8)

175 (±6)

HRrest (bpm)

63 (±7)

79 (±12)

0.76

SBP (mmHg)

120 (±16)

132 (±12)

DBP (mmHg)

80 (±9)

90 (±7)

\0.001

VO2max (ml/kg/min)

52 (±5)

33 (±6)

\0.001

\0.001 0.007

Values are expressed as mean ± SD VO2max calculated maximal oxygen consumption, HRrest heart rate at rest, SBP systolic blood pressure, DBP diastolic blood pressure

or cycling for at least 1 h more than 3 days per week, and calculated VO2max higher than 47 and 45 ml/kg/min for male and female, respectively. The group of control subjects (C; n = 30; 20 males age range between 39 and 62 years) had no regular exercise in the past 5 years, and calculated VO2max lower than 43 and 40 ml/kg/min for male and female, respectively. The basal characteristics of the subjects from ET and C groups are shown in Table 1. Study protocol The study was carried out in a laboratory room (22–24 °C) between 2 and 6 pm. The subjects refrained from the physical exertion 24 h and from consumption of alcohol, caffeine, tobacco and food for at least 5 h before the beginning of the exercise test. To lower their anxiety and enhance their HR vagal control, a psychiatrist trained in relaxation technique helped subjects to relax in supine position. After they lied down, ECG lead electrodes were attached to the chest and a cuff of the sphygmomanometer was wrapped around the right arm. Subsequently, the Polar wear link W.I.N.D, attached to the strap, was tied around the chest. Only when the subjects obtained a more relaxed state in supine position, heart rate with Polar RS800CX training computer (Polar Electro Oy Kempele Finland), ECG (Cardiosoft LP, Houston, Texas, USA) and blood pressure, were started to be continuously measured for five consecutive minutes. We repeated the exact protocol after 30 min of quiet rest in sitting position on the cycloergometer. Following the blood pressure (BP) measurement, the subjects performed a graded exercise test on the cyclo-ergometer, starting at 40 W and then increasing the work load at a rate of 50 W every 3 min until 85 % of their age predicted HRmax. Immediately afterwards, the subjects stopped exercising without the ‘‘cool-down’’ period and remained in the seated position. HR was measured during

Clin Auton Res (2014) 24:53–61

the first 2 min after stopping the exercise. Throughout the exercise test, the pedal frequency was kept at 60 revolutions per minute and the subjects breathed spontaneously without attempting to control the depth or frequency of the respiratory pattern.

55

results were presented graphically. The association between HRR values at different times was assessed using Spearman’s correlation and its 95 % confidence intervals (CI). The difference between the groups was considered statistically significant at p \ 0.05.

Heart rate and heart rate variability analysis Results Resting HRV was analyzed from 5-min-long ECG recordings obtained before exercise was examined. Data were imported into a software program (Kubios HRV version 2.0, Department of physics, University of Kuopio, Finland), where artifacts and premature beats were corrected. The selected time domain parameters, i.e., standard deviation of normal RR intervals (SDNN), depending on parasympathetic and sympathetic activities, and root mean square of sequential deviations (RMSSD), a marker of parasympathetic modulation, were computed from RR intervals. The frequency domain parameters were calculated with Welch’s periodogram and fast Fourier transform. The power of integrals over the high-frequency band (HF; 0.15–0.40 Hz), a marker of parasympathetic modulation, and the low-frequency band (F; 0.04–0.15 Hz), depending on parasympathetic and sympathetic activities, was calculated for each time series. In addition, the LF/HF ratio, which is considered as a marker of sympathovagal balance, was calculated [7]. LF and HF spectral components were expressed in absolute values and in normalized units (LF or HF/total power (TP)––VLF). The heart rate was recorded at rest (HRrest), at exercise termination (HRpeak), at the 30 s (HR30), at the 1st min (HR1) and at the 2nd min (HR2) after cessation of the exercise. The heart rate recovery at the 30 s (HRR30), 1st min (HRR1) and 2nd min (HRR2) after cessation of the exercise was defined as the differences between the HRpeak and the HR30, HR1 and HR2, respectively. Apart from expression of HRR in absolute values, the relative decline from HRpeak to the HR30, HR1, HR2 after exercise termination was also calculated using the formula: %HRR = HRR/HRpeak 9 100. Statistical analysis Statistical analyses were performed with Statistica software (Stat Soft Inc., USA) and R statistical language [13]. The results were summarized as mean (±SD). The groups of control subjects and of athletes were compared using Student’s t test. To avoid false-positive results, the p values were adjusted using a multivariate permutation procedure [14]. The association between HRR and the resting HRV indices was estimated using the univariate linear regression model, where HRR was the outcome and the HRV indices were modeled using restricted cubic splines (rcs) [15]; the

Basal characteristics The basal characteristics of the subjects from ET and C groups are summarized in Table 1. Resting HR, blood pressure and weight were significantly lower (p \ 0.007), while VO2max was significantly higher (p \ 0.001) in the ET group than in the C group. The differences in age and height between groups were not significant (p [ 0.05). Exercise test responses and heart rate recovery At peak exercise, systolic blood pressure increased to 190 ± 27 and 200 ± 30 mmHg and diastolic blood pressure to 85 ± 12 and 90 ± 10 in ET group and C group, respectively. The differences between the groups were not significant (p [ 0.05). The heart rate at peak exercise and during 2 min of recovery is shown in Table 2. At 30 s, 1 min and 2 min of recovery, HR was significantly lower in the ET group than in the C group (p \ 0.001). Both absolute and relative values of HRR during the first 2 min of recovery were significantly higher in the ET group than in the C group (p \ 0.001). All the associations remained significant also after the adjustment for multiple testing using a multivariate permutation procedure. Heart rate variability at rest Resting HRV parameters are shown in Table 3. Absolute values of resting parameters of frequency domain (TP, HF, LF) as well as time domain (SDNN, RMSSD) were significantly higher in the ET group than in the C group (p \ 0.005). LFnu and LF/HF ratios were significantly lower (p \ 0.05) and HFnu was higher (p = 0.036) in the ET group than in the C group. Only HF, SDNN and RMSSD remained significantly associated with the group membership after adjusting the analyses for multiple testing using a multivariate permutation procedure. Relationship between HRR early after exercise with resting HRV The relationship between HRR early after exercise with resting HRV in seated position was analyzed after the data from both groups were pooled together. The association

123

56


Table 2 Exercise test responses and heart rate recovery in the group of endurance-trained athletes (ET) and in the group of sedentary control subjects (C)

Table 3 Resting heart rate variability parameters in the group of endurance-trained athletes (ET) and in the group of sedentary control subjects (C)



p value

Adjusted p value

HRpeak (bpm)

150 (±5)

148 (±4)

0.089

0.95

TP (ms2)

1,297 (±1,054)

HR30 (bpm)

119 (±9)

127 (±8)

\0.001

0.01

HF (ms2)

353 (±436)

91 (±81)

HRR30 (bpm)

31 (±9)

21 (±9)

\0.001 \0.001

LF (ms2)

828 (±704)

%HRR30

20 (±6)

14 (±6)

\0.001 \0.001

HF nu

HR1(bpm)

98 (±10)

115 (±9)

\0.001 \0.001

LF/HF

HRR1(bpm)

52 (±11)

33 (±9)

\0.001 \0.001

Time domain

Heart rate



p value

Adjusted p value

Frequency domain 665 (±744) \0.001

0.14

\0.001

0.04

522 (±654)

0.004

0.78

27 (±16)

17 (±9)

0.036

0.13

4.7 (±4)

6.2 (±4)

0.034

0.94 0.02

%HRR1

34 (±7)

21 (±6)

\0.001 \0.001

SDNN (ms)

35 (±11)

25 (±11)

\0.001

HR2 (bpm)

83 (±10)

105 (±9)

\0.001 \0.001

RMSSD (ms)

31 (±13)

16 (±7)

\0.001 \0.001

HRR2 (bpm) %HRR2

66 (±11) 44 (±7)

43 (±3) 29 (±6)

\0.001 \0.001 \0.001 \0.001

190 (±27)

200 (±30)

SBPpeak (mmHg)

0.71

0.99

0.035

0.13

DBPpeak (mmHg)

85 (±14)

90 (±10)

Maximum power (W)

215 (±50)

165 (±50)

Duration of exercise (s)

740 (±239)

587 (±157) \0.001 \0.001

\0.001 \0.001

Values are expressed as mean ± SD HRpeak Heart rate at exercise termination, HR30 heart rate at 30 s after cessation of exercise, HRR30 absolute heart rate decline 30 s after cessation of exercise, %HRR30 relative decline in heart rate 30 s after cessation of exercise, HR1 heart rate at 1st min after cessation of exercise, HRR1 absolute heart rate decline 1 min after cessation of exercise, %HRR1 relative decline in heart rate 1st min after cessation of exercise, HR2 heart rate at 2nd min after cessation of exercise, HRR2 absolute heart rate decline 2 min after cessation of exercise, %HRR2 relative decline in heart rate 2 min after cessation of exercise, SBPpeak systolic blood pressure at exercise termination, DBPpeak diastolic blood pressure at exercise termination, Maximum power maximum power reached during exercise

between HRR30, HRR1 and HRR2 with the resting HRV indices in seated position is presented graphically in Figs. 1, 2, 3. The associations of HRR30, HRR1 and HRR2 with TP, LF, HF, HFnu and RMSSD were positive. The associations with LF/HF were negative for small values and positive for larger values, while the associations with SDNN were positive for small values and negative for larger values. HRR30 was significantly correlated with all HRV parameters, except for LF, HF and TP, while HRR1 and HRR2 were significantly correlated with all HRV parameters except for LF. Due to the similarity of the regression shapes at 30 s to those at 1 and 2 min of recovery, where the correlation was strongly significant, we could conclude that HRR30 was also correlated with HF and TP. The differences were probably the consequence of the high variability and low number of data at higher predictor values.

123

Values are expressed as mean ± SD TP Total power, HF high-frequency power, LF low-frequency power, LFnu low-frequency power––normalized units, HFnu high-frequency power––normalized units, LF/HF low-frequency/high-frequency ratio, SDNN standard deviation of normal RR intervals, RMSSD root mean square of sequential deviations

If the data for male and female subjects were analyzed separately, then the shapes of the correlation curves between HRR and HRV parameters were essentially the same as in the whole group of subjects. Graphical presentation of the analyzed data by gender is shown in Online Resource 1.

Discussion In this study, we examined resting HRV and HRR early after exercise in endurance-trained middle-aged athletes and in their controls. We found that resting HRV parameters HF, RMSSD and SDNN were significantly higher and that HRR at 30 s, one and two minutes after submaximal exercise was significantly faster in athletes than in the control subjects. The major finding of this study was that HRR at 30 s, 1 and 2 min after exercise was correlated with the resting HRV parameters indicating increased parasympathetic modulation (HFnu, HF and RMSSD). These observations support our hypothesis that HRR early after exercise is related to basal parasympathetic modulation. Heart rate recovery after exercise In the present study, HRR during the first 30 s, 1 and 2 min after exercise was faster in endurance-trained middle-aged athletes than in the controls. This is in accordance with the other studies which have shown that aerobic training improves HRR early after exercise in healthy subjects as


57

Fig. 1 Estimated association between heart rate recovery during the 30 s after cessation of exercise (HRR30) with resting heart rate variability using univariate linear regression model after the data from both groups were pooled together. The solid line indicates the estimated associations, and the dotted lines represent the pointwise 95 % confidence intervals. TP Total power, HF high-frequency

power, LF low-frequency power, LFnu low-frequency power–– normalized units, HFnu high-frequency power––normalized units, LF/HF low-frequency/high-frequency ratio, SDNN standard deviation of normal RR intervals, RMSSD root mean square of sequential deviations, HRR 30 s absolute heart rate decline 30 s after cessation of exercise

well as in patients with coronary artery disease or chronic heart failure irrespective of age [3, 4].

aerobic training [18]. However, others have found no changes in frequency-domain measures of resting HRV following 8 weeks of moderate or 5 years of light intensity training in sedentary middle-aged men [19, 20]. Similarly, no changes in time and frequency domain resting HRV measures were found after 12 weeks of high intensity training in healthy males and females, and after 16 weeks of moderate training in obese females with or without type 2 diabetes [21, 22]. Results from these studies could be related to differences in the population of participants and training characteristics. Nevertheless, a significantly higher resting absolute values of HF and RMSSD in athletes than in the control subjects in the present study support the hypothesis that endurance training can alter neuroregulatory control over the heart by increasing the resting parasympathetic modulation [16–18].

Resting heart rate variability According to our hypothesis, resting HF, RMSSD and SDNN were higher in middle-aged athletes than in the controls, while other HRV parameters were not significantly different between the groups. In the only published, to our knowledge, cross-sectional study examining the effect of endurance training on resting HRV in middleaged subjects, all time and frequency domain HRV parameters were higher in the group of athletes [16]. In that study, however, only a small number of female subjects were included and a large number of correlations were performed without adjusting the p values to avoid falsepositive results. Published longitudinal studies have provided some inconclusive results. Increases in absolute values of time and frequency-domain measures of resting HRV have been demonstrated following 5 months of light to moderate endurance training in sedentary men [17] and in postmenopausal women following 6 mounts of light

Correlation between HRR early after submaximal exercise and resting HRV Confirming our hypothesis, we found a positive correlation of HRR early after exercise with pre-exercise resting time

123

58


Fig. 2 Estimated association between heart rate recovery during the 1st min after cessation of exercise (HRR1) with resting heart rate variability using univariate linear regression model after the data from both groups were pooled together. The solid line indicates the estimated associations, and the dotted lines represent the pointwise 95 % confidence intervals. TP Total power, HF high-frequency

power, LF low-frequency power, LFnu low-frequency power–– normalized units, HFnu high-frequency power––normalized units, LF/HF low-frequency/high-frequency ratio, SDNN standard deviation of normal RR intervals, RMSSD root mean square of sequential deviations, HRR1 min absolute heart rate decline 1 min after cessation of exercise

and frequency domain HRV parameters indicating parasympathetic modulation (HF nu, HF and RMSSD) if data were pooled from both groups together. Indeed, the correlation between HRR30 and HF did not reach a statistical significance. However, similarity of correlation pattern between resting HF and HRR at 30 s to those at 1 and 2 min after exercise, where the correlations were statistically significant, suggests that also this correlation in fact is significant. Interestingly, the correlation curves of HRR with HF and RMSSD were steep up to 500 ms2 and 30 ms, respectively, and thereafter flattened. Although these nonlinear correlations reached statistical significance only for the HRR at 2 min after exercise, they suggest a saturation effect of the resting parasympathetic fluctuations on the HRR after exercise and/or an importance of factors limiting the HRR after exercise. Accordingly, pyridostigmine significantly increased postexercise HRR compared with placebo only in sedentary subjects but not in trained athletes in an earlier study [23]. Another interesting finding of our study was the inversion of correlation between HRR and resting SDNN from positive for values up to 40 ms and

negative beyond 40 ms indicating that an optimal resting SDNN for a maximal HRR after exercise would be the values at the transition zone between positive and negative parts of the correlation curve. This is supported also by the inversion of correlation between HRR and resting LF/HF from negative for values up to 5 and positive beyond. If the data for woman and man subjects were analyzed separately, then the shapes of the correlation curves between HRR and HRV were essentially the same as when analyzed together. This might suggest no gender difference in the relationship between HRV at rest and HRR after exercise; however, it should be interpreted carefully due to the relatively small number of subjects within each individual group, which may not have enough statistical power. Our findings are in line with earlier pharmacological studies suggesting that HRR 2 min after submaximal exercise (80–85 % of VO2max) is almost entirely dependent on parasympathetic activation [4, 6]. The majority of earlier studies on the correlation between HRR and HRV parameters were performed using the maximal exercise test protocol [8, 10, 24, 25]. Factors, such as sympathetic

123


59

Fig. 3 Estimated association between heart rate recovery during the 2nd min after cessation of exercise (HRR2) with resting heart rate variability using univariate linear regression model after the data from both groups were pooled together. The solid line indicates the estimated associations, and the dotted lines represent the point wise 95 % confidence intervals. TP Total power, HF high-frequency

power, LF low-frequency power, LFnu low-frequency power–– normalized units, HFnu high-frequency power––normalized units, LF/HF low-frequency/high-frequency ratio, SDNN standard deviation of normal RR intervals, RMSSD root mean square of sequential deviations, HRR2 min absolute heart rate decline 2 min after cessation of exercise

activity or ventilation, that are known to interfere with parasympathetic outflow during the postexercise recovery phase, prevent us from comparing these studies to those performed at submaximal exercise [26]. There are only a few studies examining the correlation of HRR with HRV parameters at submaximal exercise. In two of these studies, HRR after 3 and 4 min of recovery was positively correlated with parasympathetic HRV parameters, and either negatively or positively with LF [8, 27]. Another study showed no significant relationship between any resting HRV parameters and HRR during the first or second minute after cessation of exercise in healthy 20-year-old sedentary men [5]. Two possible factors should be considered while comparing the results of these studies. First, the lack of correlation between the HRR and resting HRV in some studies [5] could be due to high homogeneity in aerobic fitness level of subjects. Namely, using very narrow selection criteria (e.g., young subjects with very high VO2max), only a small, highly homogeneous fraction of the entire population is tested. In such cases, it is not possible to find any correlation between HRR and resting HRV in

the existing ‘‘true’’ relationship between the two variables [28]. Second, a correlation of HRR with LF found in some studies might be the consequence of a different methodology. In two studies [8, 27], both HRV and HRR recordings were made in the supine position, where the balance between sympathetic and parasympathetic nervous system is different from the recordings in the sitting position. Both studies also implied workload on a treadmill, where obtained HRR values after the equivalent workloads differ from those obtained on a cyclo-ergometer [29]. Together these findings suggest that monitoring the parasympathetic HRV parameters at basal resting conditions could potentially provide an acceptable assessment of physical fitness at workloads around anaerobic threshold for an individual. Limitations of the study In this study, the subjects were assigned to two groups according to the daily physical activities and Astrand [11] nomogram, which is an adjusted calculation of VO2max

123

60

from submaximal heart rate, considering age, gender, height and maximal heart rate (HRmax). This indirect method is not as accurate as the direct testing with, for example, the Douglas bag method that requires sophisticated equipment to measure the volume and gas concentrations of inspired and expired air. However, we believe that the estimation of the VO2max with this indirect method was accurate enough for the purpose of our study. Accordingly, HRR measured up to 1 min after submaximal exercise is predominantly dependent on parasympathetic activity and independent on the intensity of exercise and cardiovascular condition [6, 26]. In addition, in our study, respiration was not controlled during the HR recording for the HRV analysis. It has been documented that the depth and frequency of breathing can modify vagal modulation and HRV [30]. However, ventilation has been reported to be similar between trained and untrained states during rest [31]. Therefore, we believe that uncontrolled respiration in our study did not significantly influence resting HRV. Conflict of interest On behalf of all authors, the corresponding author states that there is no conflict of interest.

References 1. Cole CR, Foody JM, Blackstone EH, Lauer MS (2000) Heart rate recovery after submaximal exercise testing as predictor of mortality in a cardiovascularly healthy cohort. Ann Intern Med 132:552–555 2. Lauer MS (2009) Autonomic function and prognosis. Cleve Clin J Med 76(Suppl 2):S18–S22 3. Darr KC, Bassett DR, Morgan BJ, Thomas DP (1988) Effects of age and training status on heart rate recovery after peak exercise. Am J Physiol 254(2 Pt 2):H340–H343 4. Sugawara J, Murakami H, Maeda S, Kuno S, Matsuda M (2001) Change in post-exercise vagal reactivation with exercise training and detraining in young men. Eur J Appl Physiol 85:259–263 5. Javorka M, Zˇila I, Balha´rek T, Javorka K (2002) Heart rate recovery after exercise: relations to heart rate variability and complexity. Braz J Med Biol Res 35:991–1000 6. Imai K, Sato H, Hori M, Kusuoka H, Ozaki H, Yokoyama H, Takeda H, Inoue M, Kamada T (1994) Vagally mediated heart rate recovery after exercise is accelerated in athletes but blunted in patients with chronic heart failure. J Am Coll Cardiol 24:1529–1535 7. Task Force of the European Society of Cardiology and the North American Society of Pacing and Electrophysiology (1996) Heart rate variability. Standards of measurement, physiological interpretation, and clinical use. Circulation 93:1043–1065 8. Evrengul H, Tanriverdi H, Kose S, Amasyali B, Kilic A, Celik T, Turhan H (2006) The relationship between heart rate recovery and heart rate variability in coronary artery disease. ANE 11:154–162 9. Nunan D, Jakovljevic´ DG, Donovan G, Singleton LD, Sandercock GRH, Brodie DA (2010) Resting autonomic modulations and the heart rate response to exercise. Clin Auton Res 20:213–221

123

Clin Auton Res (2014) 24:53–61 10. Bosquet L, Gamelin FX, Berthoin S (2007) Is aerobic endurance a determinant of cardiac autonomic regulation? Eur J Appl Physiol 100:363–369 11. Astrand PO, Kaare R, Dahl HA, Stromme SB (2003) Textbook of work physiology, 4th edn. Human kinetics, Champaign, p 281 12. Robergs RA, Landwehr R (2002) The surprising history of the ‘‘HRmax = 220-age’’ equation. J Exer Physiol 5(2):1–10 13. Development Core Team (2009) R: A language and environment for statistical computing. R Foundation for Statistical Computing, Vienna, Austria. ISBN 3-900051-07-0, URL. Available online at: http://www.R-project.org. Accessed February 19, 2013 14. Westfall PH, Young SS (1993) Resampling-based multiple testing. Wiley, New York 15. Harrell FE Jr, Lee KL, Pollock BG (1988) Regression models in clinical studies: determining relationships between predictors and response. J Natl Cancer Inst 80(15):1199–1202 16. Paschoal MA, Polessi EA, Simioni FC (2008) Evaluation of heart rate variability in trained and sedentary climacteric women. Arq Bras Cardiol 90:74–79 17. Cornelissen VA, Verheyden B, Aubert AE, Fagard RH (2009) Effects of aerobic training intensity on resting, exercise and postexercise blood pressure, heart rate and heart-rate variability. J Human Hypertens. doi:10.1038/jhh 18. Jurca R, Church TS, Morss GM, Jordan AN, Earnest CP (2004) 8 weeks of moderate-intensity exercise training increases heart rate variability in sedentary postmenopausal women. Am Heart J 147(5):e21 19. Boutcher SH, Stein P (1995) Association between heart rate variability and training response in sedentary middle-aged men. Eur J Appl Physiol Occup Physiol 70:75–80 20. Uusitalo AL, Laitinen T, Va¨isa¨nen SB, Jordan AN, Earnest CP (2004) Physical training and heart rate and blood pressure variability: a 5-year randomized trial. Am J Physiol Heart Circ Physiol 286:H1821–H1826 21. Carter JB, Banister EW, Blaber AP (2003) The effect of age and gender on heart rate variability after endurance training. Med Sci Sports Exerc 35:1333–1340 22. Figuera A, Baynard T, Fernhall B, Carhart R, Kanalay JA (2007) Endurance training improves post-exercise cardiac autonomic modulation in obese women with and without type 2 diabetes. Eur J Appl Physiol 100:437–444 23. Dewland TA, Androne AS, Lee FA, Lampert RJ, Katz SD (2007) Effect of acetylcholinesterase inhibition with pyridostigmine on cardiac parasympathetic function in sedentary adults and trained athletes. Am J Physiol 293:H86–H92. doi:10.1152/ajpheart. 01339.2006 24. Ohuchi H, Suzuki H, Yasuda K, Arakaki Y, Echigo S, Kamiya T (2000) Heart rate recovery after exercise and cardiac autonomic nervous activity in children. Pediatr Res 47:329–335 25. Lee M, Mendoza A (2012) Dissociation of heart rate variability and heart rate recovery in well-trained athletes. Eur J Appl Physiol 112:2757–2766 26. Buchheit M, Simon C, Charloux A, Doutreleau S, Piquard F, Brandenberger G (2006) Relationship between very high physical activity energy expenditure, heart rate variability and self-estimate of health status in middle-aged individuals. Int J Sports Med 27:697–701 27. Chen JY, Lee YL, Tsai WC, Lee CH, Chen PS, Li YH, Tsai LM, Chen JH, Lin LJ (2011) Cardiac autonomic functions derived from short term heart rate variability recordings associated with heart rate recovery after treadmill exercise test in young individuals. Heart Vess 26(3):282–288 28. Armstrong RA, Eperjesi F, Gilmartin B (2005) The use of correlation and regression methods in optometry. Clin Exp Optom 88(2):81–88

Clin Auton Res (2014) 24:53–61 29. Rahimi K, Thomas A, Adam M, Hayerizadeh BF, Schuler G, Secknus MA (2006) Implications of exercise test modality on modern prognostic markers in patients with known or suspected coronary artery disease: treadmill versus bicycle. Eur J Cardiovasc Prev Rehabil 13(1):45–50 30. Hayano J, Mukai S, Sakakibara M, Okada A, Takata K, Fujinami T (1994) Effects of respiratory interval on vagal modulation of heart rate. Am J Physiol 267(1 Pt 2):H33–H40

61 31. Hatfield BD, Spalding TW, Santa Maria DL, Porges SW, Potts JT, Byrne EA, Brody EB, Mahon AD (1998) Respiratory sinus arrhythmia during exercise in aerobically trained and untrained men. Med Sci Sports Exerc 30(2):206–214

123