Impact of High-intensity Intermittent and
Moderate-intensity Continuous Exercise on Autonomic
Modulation in Young Men
Authors C. Cabral-Santos1, 2, T. R. Giacon2, E. Z. Campos1, J. Gerosa-Neto1, B. Rodrigues3, L. C. M. Vanderlei2,
F. S. Lira1, 2
Affiliations Affiliation addresses are listed at the end of the article
Introduction
▼
Heart rate variability (HRV) is considered a non-invasive and well-accepted method for measur-ing cardiovascular autonomic modulation [20]. The HRV recovery has been widely studied, since its reduction during rest and exercise are identi-fied as a predictor of mortality and the risk of sudden death [13, 20]. More recently, HRV recov-ery has also been considered an important tool to evaluate stress after exercise sessions of different intensities [10, 24].
Seiler et al. [26] compared HRV recovery after different exercises in trained athletes. They found that iso-volume (i. e., 60 min) exercise performed between the aerobic and anaerobic thresholds, and above the anaerobic threshold, resulted in similar HRV recovery, indicating that high-inten-sity exercise (HIE) does not promote higher auto-nomic stress regarding return to baseline conditions compared to moderate-intensity exercise (MIE). According to Seiler et al. [26], this finding may explain why athletes prefer to train
at a higher intensity, triggering greater physical capacity adjustments, with an autonomic modu-lation recovery similar to that observed in mod-erate-intensity training [9, 26].
Recently, studies have verified the effects of high-intensity compared to moderate-high-intensity train-ing on body composition, physical fitness and molecular alterations; high-intensity training is usually more effective in improving these varia-bles [4, 5, 12]. However, data concerning whether HIE and MIE exert different HRV recovery in physically active subjects is conflicting, because HIE intensities are usually too low and the vol-ume is not matched [6, 16, 20, 23]. Thus, in con-trast to highly trained athletes [8, 26], physically active subjects exhibit a different autonomic bal-ance after iso-volume HIE and MIE, which is an important factor when selecting training distri-bution for this population.
Moreover, high-intensity training is often selected as it is time-efficient compared with moderate-intensity training [4, 12, 26]; however, the superiority of high-intensity training over
accepted after revision December 21, 2015 Bibliography DOI http://dx.doi.org/ 10.1055/s-0042-100292 Published online: March 7, 2016
Int J Sports Med 2016; 37: 431–435 © Georg Thieme Verlag KG Stuttgart · New York ISSN 0172-4622
Correspondence
Carolina Cabral Santos Eduardo Zapaterra Campos Exercise and Immunometabo-lism Reseach Group - Physical Education Department Universidade Estadual Paulista Rua Roberto Simonse, 305 Presidente Prudente Brazil 19061513 Tel.: + 55/183/2295 826 Fax: + 55/183/2295 826 [email protected] Key words ●▶ high-intensity exercise ●▶ moderate-intensity continu-ous exercise ●▶ autonomic modulation ●▶ heart rate variability
Abstract
▼
The aim of this study was to compare heart rate variability (HRV) recovery after two iso-volume (5 km) exercises performed at different intensi-ties. 14 subjects volunteered (25.17 ± 5.08 years; 74.7 ± 6.28 kg; 175 ± 0.05 cm; 59.56 ± 5.15 mL · kg − 1 ·
min − 1) and after determination of peak oxygen
uptake (VO2Peak) and the speed associated with
VO2Peak (sVO2Peak), the subjects completed 2
ran-dom experimental trials: high-intensity exercise (HIE – 1:1 at 100 % sVO2Peak), and
moderate-inten-sity continuous exercise (MIE – 70 % sVO2Peak). HRV
and RR intervals were monitored before, during and after the exercise sessions together with, the HRV analysis in the frequency domains (high-frequency – HF: 0.15 to 0.4 Hz and low-(high-frequency
– LF: 0.04 to 0.15 Hz components) and the ratio between them (LF/HF). Statistical analysis com-parisons between moments and between HIE and MIE were performed using a mixed model. Both exercise sessions modified LFlog, HFlog, and LF/HF (F = 16.54, F = 19.32 and F = 5.17, p < 0.05, respectively). A group effect was also found for LFlog (F = 23.91, p < 0.05), and HFlog (F = 57.55, p < 0.05). LF/HF returned to resting value 15 min after MIE exercise and 20 min after HIE exercise. This means that the heavy domain (aerobic and anaerobic threshold) induces dissimilar auto-nomic modification in physically active subjects. Both HIE and MIE modify HRV, and generally HIE delays parasympathetic autonomic modulation recovery after iso-volume exercise.
moderate-intensity training could be even higher if an iso- volume HIE is performed. On the other hand, this could lead to higher autonomic stress in HIE compared to MIE, both during exercise and recovery (i. e., retarding autonomic modulation recovery), as both intensity and session time are greater, reduc-ing the applicability of HIE for every-day trainreduc-ing. However, this hypothesis remains to be tested. Thus, the aim of the present study.
Methods
▼
Subjects
14 male subjects volunteered for the present study. They pre-sented a health and neuromuscular status that demonstrated their ability to complete the study protocol. All procedures per-formed in the study were in accordance with the ethical stand-ards of the University Research Ethics Committee for studies involving human participants and met the ethical standards of this journal described for Harriss and Atkinson [14]. Written informed consent was obtained from all subjects after they had been informed about the purpose and risks of the study.
Experimental design
Subjects completed 3 experimental trials at the laboratory. The first visit aimed to determine peak oxygen uptake (VO2Peak) and
the speed associated with VO2Peak (sVO2Peak). During the
remain-ing 2 visits, all participants were submitted to 2 protocols of 5 km of treadmill running in randomized sequence [28]: high-intensity exercise (HIE), or moderate-high-intensity exercise (MIE), separated by at least 72 h. Due to the influence of time of day on HRV, all tests took place at the same time of the day, between 1:00 p.m. and 6:00 p.m., at a average temperature of between 20 ℃ and 24 ℃ [1]. The subjects were instructed to abstain from strenuous exercise for at least 24 h prior to each exercise session and were encouraged to maintain their usual nutritional and hydration routines. Moreover, they were also requested not to ingest stimulants (tea, coffee, soda, chocolate, chocolate powder) or alcoholic beverages during this period.
Incremental test
The participants were subjected to an incremental test on a treadmill (Inbramed MASTER CI, Inbrasport®, Porto Alegre,
Bra-zil). The initial speed was set at 8 km · h − 1, increasing by 1 km · h − 1
every 2 min until volitional exhaustion. Strong verbal encour-agement was given during the test. The oxygen uptake was measured (Quark PFT, Cosmed®, Rome, Italy) throughout the test
and the average of the last 30 s defined as VO2Peak. The sVO2Peak
was assumed as the final incremental test speed. When the sub-ject was unable to complete a stage, the speed was expressed according to the time in the final stage, determined as follows: sVO2Peak = speed of final complete stage + [(time, in seconds,
remaining at the final incomplete stage / 120 s) * 1 km.h − 1] [18].
High and moderate-intensity exercise sessions
For both exercise trials, the subjects performed a warm-up con-sisting of running at 50 % of sVO2Peak for 5 min at 1 % incline. The
HIE was performed intermittently with subjects running on a treadmill for 1 min at 100 % of sVO2Peak [23], interspersed by
1 min of passive recovery (without exercise) until they had com-pleted 5 km. The MIE consisted of a continuous 5-km run on the treadmill at 70 % of sVO2Peak [23].
Heart rate variability recovery
Heart rate variability was monitored before, during and after each exercise session. A recording strap was placed on the indi-vidual’s chest at the sternal angle and at the Polar S810i heart rate receiver (Polar Electro, Finland) on their wrist. After place-ment of the strap and monitor, the individuals remained at rest in the supine position, breathing spontaneously for 20 min. After this period, the individuals performed the exercise protocol before returning to the supine position, at rest, breathing spon-taneously, for 60 min. For HRV analysis, the behavior pattern was recorded beat-to-beat throughout the experimental protocol, with a sampling rate of 1 000 Hz.
In order to evaluate HRV, data on the intervals between heart beats (RR intervals) were sent to a microcomputer, from the pulse receptor’s data transmission port to Polar Precision
Perfor-mance software (Polar Electro, Finland), using an infrared signal
interface. Only series with less than 5 % errors were included in the study. The RR interval series passed initially through filter-ing usfilter-ing standard filter Polar Precision Performance software (Polar Electro, Finland) [17], with a moderate filter, after which visual inspection was performed of the temporal series of RR intervals on the computer monitor to ensure the absence of arti-facts that could interfere with HRV analysis. The HRV indices were calculated in the time and frequency domains using Kubios HRV software (Kubios, Biosignal Analysis and Medical Image Group, Department of Physics, University of Kuopio, Finland) [22].
The RR interval series were analyzed at the following moments: M1 (after 5 min at rest), M2 (immediately after the exercise until the 5th min of recovery), after the 5th min, the recovery period
was divided into 11 excerpts (M3 to M13) of 5 min each. All the excerpts obtained contain at least 256 consecutive RR intervals. In the time domain, the following indices were used for HRV analysis: RMSSD and SDNN. The RMSSD index is defined as the root mean square of successive differences between adjacent normal RR intervals in a given time interval in milliseconds, and the SDNN represents the standard deviation of normal to normal RR intervals in milliseconds [30].
For the HRV analysis in the frequency domain, the high-fre-quency (HF, 0.15 to 0.4 Hz) and low-frehigh-fre-quency components (LF, 0.04 to 0.15 Hz), as well as the ratio between them (LF/HF), were analyzed. The spectral analysis was calculated using the Fourier Transform algorithm [30] after which the HRV parameters were logarithmically transformed (Log) to control for skewed distri-butions.
Statistical analysis
Statistical analysis was performed using SPSS 17.0. Prior to the analysis, the normality of data was tested and confirmed using the Shapiro-Wilk test. Comparison of HRV recovery between moments, and between HIE and MIE was performed using a mixed model. Exercise and moment were specified as fixed fac-tors, and the subjects as a repeated factor. If a significant main effect or interaction existed, this was further explored through multiple comparison analyses with the Sidak adjustment for multiple comparisons. The comparison between both exercise characteristics was assessed using Student’s t test, with signifi-cance level set at 5 %.
Results
▼
The subjects’ characteristics, anthropometry measures, sum-mary of the incremental test and baseline HRV are shown in ●▶ Table 1.
The exercise characteristics are presented in ●▶ Table 2.
Signifi-cant differences were found between speed, total session time and total exercise time.
Resting HRV was similar between each exercise session. Both exercise sessions significantly modified SDNN and RMSSD (F = 18.45, F = 14.42, p < 0.05; respectively; ●▶ Fig. 1), as well as LF
log, HF log and LF/HF (F = 16.54, F = 19.32, and F = 5.17, p < 0.05; respectively, ●▶ Fig. 2). A group effect was also found for SDNN
(F = 12.16, p < 0.05), RMSSD (F = 43.18, p < 0.05), LF log (F = 23.91, p < 0.05) and HF log (F = 57.55, p < 0.05). Nevertheless, no interac-tion (group vs. moment) was observed.
For MIE, the SDNN returned to rest values after 20 min of recov-ery, while the RMSSD returned after 30 min. Both the LF log and HF log returned to rest values at the 20th min of recovery. After
the HIE, the SDNN returned to the rest values only after 45 min, while the RMSSD did not recover over the 60 min. Moreover, the LF log and HF log recovered after 20 and 35 min, respectively. The LF/HF returned to rest values 15 min after MIE exercise and 20 min after HIE exercise.
Discussion
▼
The aim of the present study was to compare HRV recovery after 2 iso-volume (5 km) exercises performed at different intensities. The main finding of the present study was that both iso-volume HIE and MIE modified HRV, and that HIE delayed HRV recovery compared with MIE. These results mean that from the heavy domain (i. e., between the aerobic and anaerobic threshold) onward, iso-volume exercise induces dissimilar autonomic modification in physically active subjects.
During exercise, general HRV decreases (i. e., parasympathetic withdraw and sympathetic excitation) with the aim of increas-ing heart rate [13]. After exercise, the return of HRV to rest level delays according to exercise intensity, duration and modality [6, 20, 26]. Furthermore, HRV recovery is also used to character-ize the acute stress response to training sessions in trained indi-viduals [3, 26], which contributes to training prescription. Thus, since HIE and MIE training were compared in physically active subjects [12, 28], understanding the behavior of HRV recovery after these training sessions may also contribute to training pre-scriptions for physically active subjects or for optimizing exer-cise training sessions for individuals who present cardiovascular disease. Stewart et al. [29] have verified that HRV recovery is delayed after another range of exercise intensity (at anaerobic threshold), and duration (2 h), evidencing the applicability of HRV recovery to differentiate autonomic recovery after different stimuli. Indeed, HRV obtained by both cardiac monitor and elec-trocardiography is also suggested to be used in different view-points [19].
It is important to note that SDNN increased immediately after both exercises. SDNN index, which represents the standard devi-ation of normal to normal RR intervals and reflects the overall variability [30], had a significant increase immediately after exercise until the 5th min of recovery compared to baseline.
Dur-ing the recovery from the exercise period, the initial return of heart rate to baseline occurs primarily due to parasympathetic reactivation [7, 15]. With cessation of exercise, there is loss of central command, and the baroreflex activation and other mechanisms contribute to the parasympathetic activity increase [15]. This parasympathetic reactivation promotes increase in RR intervals variation, which may be associated with SDNN index increase immediately after exercise.
Table 1 Mean, standard deviation and 95 % of confidence interval (95 % CI) for subjects characteristics and heart rate variability indices.
Variable Subjects (n = 14) 95 % CI Age (years) 25.17 ± 5.08 22.50–27.83 Body Mass (kg) 74.70 ± 6.28 71.41–77.98 Height (m) 1.75 ± 0.05 1.72–1.77 BMI (kg/m2) 24.42 ± 1.91 23.41–25.42 VO2peak (ml · kg · min − 1) 59.56 ± 5.15 56.86–62.25 Rest HR (bpm) 61.50 ± 8.07 57.27–65.72 RMSSD (ms) 58.15 ± 18.92 48.23–68.06 SDNN (ms) 68.50 ± 16.10 60.06–76.93 LF Log 3.04 ± 0.30 2.88–3.19 HF Log 2.99 ± 0.29 2.83–3.14 LF:HF (a.u.) 1.45 ± 1.03 0.91–1.98
Values are mean ± standard deviation. BMI = body mass index; VO2peak = peak oxygen uptake; HR = heart rate; RMSSD = root mean square of successive differences; SDNN = standard deviation of NN intervals; LF = low frequency; HF = high frequency; LF:HF = LF/HF ratio
Table 2 Mean and standard deviation (SD) of high-intensity exercise (HIE) and moderate-intensity exercise (MIE) characteristics (n = 14).
HIE MIE
Speed (km · h − 1) 14.51 ± 1.19 10.16 ± 0.83 *
Total session duration (minutes) 40.72 ± 3.39 29.72 ± 2.42 * Total exercise duration (minutes) 20.86 ± 1.70 29.72 ± 2.42 *
* significantly different from HIE
250 200 150 100 50 0 150 100 50 0 0 20 40 60 0 20 40 60 MIE MIE MIE MIE SDNN (% of rest value) RMSS D (% of rest value) * *
Time (minutes) Time (minutes)
ϕ ϕ
a b Fig. 1 Heart rate variability recovery of time
domain indices (a SDNN and b RMSSD) after MIE (●), and HIE (■). * significantly different from rest value for MIE; Φ significantly different from rest value for HIE.
Martinmäki and Rusko analyzed short-term HRV recovery (10 min) after high-intensity and low-intensity exercise and found that the intensity delayed parasympathetic recovery after high-intensity exercise [20]. However, the high-intensity exer-cise was performed at 61 % of maximal power, which is lower than the MIE in the present study. Mourot et al. compared a con-stant and interval exercise with similar physical workloads [21]. They found slower parasympathetic recovery (i. e., HF) up to 60 min after the interval exercise. Compared with the present study, these studies presented either lower intensity of exercise [20] or session volume [21]. Aiming to understand the effect of both intensity and volume, Kaikkonen et al. compared HRV recovery after 3 iso-volume different constant-speed exercises: low, moderate and high-intensity [16]. The authors found no difference in HRV recovery after moderate and high-intensity exercise; however, the Kaikkonen et al. study also presented an intensity lower than the present study (74 % vs. 100 % of sVO2Peak) [16].
When comparing the effects of HIE and MIE, HIE seems to induce better or similar physical fitness adaptations than MIE, with sig-nificantly lower volume [5, 12]. Therefore, HIE training would induce even higher adaptation, if performed with the same MIE volume; however, this would likely induce greater autonomic stress after the HIE. The present study demonstrates that both iso-volume HIE and MIE likely reduced parasympathetic control (RMSSD) and increased sympathetic predominance (high LF/HF ratio) [3]. However, the group effects for all variables indicate that, in general, HIE delays parasympathetic autonomic modula-tion recovery when compared with iso-volume MIE. These results contradict Seiler et al.'s study, in which similar HRV responses were verified after 2 exercise sessions performed at and above anaerobic threshold intensity with the same total ses-sion time [26]. 2 factors may contribute to the differences: (i) the subject's training status, since they evaluated highly trained ath-letes, and (ii) we equalized exercise session volume by distance covered instead of session time.
In contrast to highly trained athletes [8, 25], physically active subjects do exhibit different HRV recovery after iso-volume HIE and MIE; this is important for selecting training distributions for this population. Given the increasing number of investigations comparing different aspects of high and moderate-intensity training (physiological, morphological, and performance), it would also be interesting to understand whether higher auto-nomic stress after HIE leads to different HRV adaptations. More-over, it is not known whether HIE accumulation in training interferes with training commitment.
It is important to highlight that even though the exercise volume was the same, the training load was not. Banister et al.
devel-oped the concept of training impulse – based on heart rate response to training – as a method for integrating training vari-ables [2], and Foster et al. proposed the calculation of training impulse through the rating of perceived effort (RPE) [11]. Both methods help to understand external loads (volume: minutes, kilometers, and repetitions) and internal loads (heart rate, lac-tate, and rating of perceived effort) [3]. It is likely that the HIE training impulse is higher than MIE, which would influence HRV response. This issue is a limitation of the present study. Moreo-ver, we did not control breathing frequency during the HRV analysis, which influences HFlog values [25]; however, even after 60 min of recovery, we found modifications in HRV related to the parasympathetic branch.
In summary, both HIE and MIE modify HRV, and generally HIE delays parasympathetic autonomic modulation recovery after an iso-volume exercise. This delay may be taken into account when prescribing training for physically active subjects, espe-cially since RMSSD – indicating parasympathetic control – did not recover after 1 h. Future studies should verify whether iso-training loads HIE and MIE also present different HRV recovery values.
Acknowledgements
▼
Eduardo Zapaterra Campos thanks CNPq for their support (401676/2014-5). Fabio Santos Lira thanks Fapesp for their sup-port (2013/25310-2).
Conflict of interest: None.
Affiliations
1 Exercise and Immunometabolism Group, Physical Education Department,
Universidade Estadual Paulista, Presidente Prudente, Brazil
2 Physiotherapy Graduate Program, Physiotherapy Department, Universidade
Estadual Paulista, Presidente Prudente, Brazil
3 Faculty of Physical Education, University of Campinas, Campinas, Brazil
References
1 Ammar A, Chtourou H, Trabelsi K, Padulo J, Turki M, Abed KE,
Hoekel-mann A, Hakim A. Temporal specificity of training: intra-day effects
on biochemical responses and Olympic-Weightlifting performances. J Sports Sci 2015; 33: 358–368
2 Banister EW, Clavert WT, Savage MV, Bach T. A systems model of train-ing for athletic performance. Aust J Sports Med 1975; 7: 57–61 3 Borresen J, Lambert MI. Autonomic control of heart rate during and
after exercise: measurements and implications for monitoring train-ing status. Sports Med 2008; 38: 633–646 Review
150 100 50 0 150 100 50 0 0 20 40 60 0 20 40 60 MIE
MIE MIEMIE
LF L og (% of rest value) HF L og (% of rest value) * *
Time (minutes) Time (minutes)
ϕ ϕ
a b Fig. 2 Heart rate variability recovery of frequency
domain indices (a LF log and b HF log) after MIE (●), and HIE (■). * significantly different from rest value for MIE; Φ significantly different from rest value for HIE.
4 Burgomaster KA, Hughes SC, Heigenhauser GJ, Bradwell SN, Gibala MJ. Six sessions of sprint interval training increases muscle oxidative potential and cycle endurance capacity in humans. J Appl Physiol 2005; 98: 1985–1990
5 Burgomaster KA, Heigenhauser GJ, Gibala MJ. Effect of short-term sprint interval training on human skeletal muscle carbohydrate metabolism during exercise and time-trial performance. J Appl Physiol 2006; 100: 2041–2047
6 Cipryan L, Vala R. Cardiac autonomic regulation after continuous and intermittent maximal exercise interventions. J Sports Med Phys Fit-ness 2015; 55: 495–505
7 Esco MR, Olson MS, Williford HN, Blessing DL, Shannon D, Grandjean P. The relationship between resting heart rate variability and heart rate recovery. Clin Auton Res 2010; 20: 33–38
8 Esteve-Lanao J, San Juan AF, Earnest CP, Foster C, Lucia A. How do endurance runners actually train? Relationship with competition per-formance. MedSci Sports Exerc 2005; 37: 496–504
9 Esteve-Lanao J, Lucia A, deKoning JJ, Foster C. How do humans control physiological strain during strenuous endurance exercise? PLoS One 2008; 3: e2943
10 Finkelstein J, Jeong IC. Using heart rate variability for automated iden-tification of exercise exertion levels. Stud Health Technol Inform 2015; 208: 137–141
11 Foster C, Florhaug JA, Franklin J, Gottschall L, Hrovatin LA, Parker S,
Doleshal P, Dodge C. A new approach to monitoring exercise training.
J Strength Cond Res 2001; 15: 109–115
12 Gibala MJ, Little JP, van Essen M, Wilkin GP, Burgomaster KA, Safdar A,
Raha S, Tarnopolsky MA. Short-term sprint interval versus traditional
endurance training: similar initial adaptations in human skeletal mus-cle and exercise performance. J Physiol 2006; 15: 901–911 13 Goldberger JJ, Le FK, Lahiri M, Kannankeril PJ, Ng J, Kadish AH.
Assess-ment of parasympathetic reactivation after exercise. Am J Physiol 2006; 290: H2446–H2452
14 Harriss DJ, Atkinson G. Ethical Standards in Sport and Exercise Science Research: 2016 Update. Int J Sports Med 2015; 36: 1121–1124 15 Javorka M, Zila I, Balhárek T, Javorka K. Heart rate recovery after
exer-cise: relations to heart rate variability and complexity. Braz J Med Biol Res 2002; 35: 991–1000
16 Kaikkonen P, Nummela A, Rusko H. Heart rate variability dynamics during early recovery after different endurance exercises. Eur J Appl Physiol 2007; 102: 79–86
17 Kiviniemi AM, Hautala AJ, Kinnunen H, Tulppo MP. Endurance training guided individually by daily heart rate variability measurements. Eur J ApplPhysiol 2007; 101: 743–751
18 Kuipers H, Verstappen FT, Keizer HA, Geurten P, van Kranenburg G. Vari-ability of aerobic performance in the laboratory and its physiologic correlates. Int J Sports Med 1985; 6: 197–201
19 Lewis MJ, Short AL. Exercise and cardiac regulation: what can elec-trocardiographic time series tell us. Scand J Med Sci Sports 2010; 20: 794–804
20 Martinmäki K, Rusko H. Time-frequency analysis of heart rate variabil-ity during immediate recovery from low and high intensvariabil-ity exercise. Eur J Appl Physiol 2008; 103: 377–378
21 Mourot L, Bouhaddi M, Tordi N, Rouillon JD, Regnard J. Short- and long-term effects of a single bout of exercise on heart rate variability: com-parison between constant and interval training exercises. Eur J Appl Physiol 2004; 92: 508–517
22 Niskanen JP, Tarvainen MP, Ranta-aho PO, Karjalainen PA. Software for advanced HRV analysis. Comput Meth Progr Biomed 2004; 76: 73–81 23 Padulo J, Chamari K, Ardigò LP. Walking and running on treadmill:
the standard criteria for kinematics studies. MLTJ 2014; 4: 159–162 24 Padulo J, Powell D, Milia R, Ardigò LP. A paradigm of uphill running.
Plos One 2013; 8: e69006
25 Seiler KS, Kjerland GØ. Quantifying training intensity distribution in elite endurance athletes: is there evidence for an "optimal" distribu-tion? Scand J Med Sci Sports 2006; 16: 49–56
26 Seiler S, Haugen O, Kuffel E. Autonomic recovery after exercise in trained athletes: intensity and duration effects. Med Sci Sports Exerc 2007; 39: 1366–1373
27 Schulz K, Grimes DA. Generation of allocation sequences in randomised trials: chance, not choice. Lancet 2002; 359: 515–519
28 Skelly LE, Andrews PC, Gillen JB, Martin BJ, Percival ME, Gibala MJ. High-intensity interval exercise induces 24-h energy expenditure similar to traditional endurance exercise despite reduced time commitment. Appl Physiol Nutr Metab 2014; 39: 845–848
29 Stewart GM 1, Kavanagh JJ, Koerbin G, Simmonds MJ, Sabapathy S. Car-diac electrical conduction, autonomic activity and biomarker release during recovery from prolonged strenuous exercise in trained male cyclists. Eur J Appl Physiol 2014; 114: 1–10
30 Vanderlei LCM, Pastre CM, Hoshi RA, Carvalho TD, Godoy MF. Basic notions of heart rate variability and its clinical applicability. Rev Bras Cir Cardiovasc 2009; 24: 205–221
