Abstract
Introduction: Heart rate variability (HRV), expressed by the beat-to-beat variation in heart rate, offers a noninvasive indicator of autonomic nervous system (ANS) activity. Measurement of the ANS response is increasingly used to evaluate the effect of training load on the organism. Most authors compared only the impact of different types of running training sessions (TS) (low-intensity training (LIT), high-intensity interval training (HIIT)), or separately plyometric TS on HRV. In this study, HRV was used to clarify how different types of running TS and plyometric TS influence post-exercise ANS response. Methods: 12 highly trained runners participated in this study. Each subject completed three types of TS – LIT running, HIIT running and plyometric. 5-min pre-exercise ECGs were recorded just before TS and 5-min post- exercise ECGs were recorded 10 min after TS. Altogether 13 time-domain and frequency-domain HRV features were calculated. Finally, the changes between pre- and post- exercise values of HRV features were computed. Results: From 13 tested features, 9 and 10 features were statistically significant for distinguishing between plyometrics and LIT, and HIIT and LIT, respectively. There are no statistically significant differences in HRV changes between plyometrics and HIIT. It could be assumed these two TS affect ANS similarly. 1. Introduction One of the possibilities how to monitor body response to sport training is the evaluation of heart rate variability (HRV). HRV features reflect the autonomic nervous system (ANS) function and thus provide information about reaction of body to sport training. Based on the HRV analysis, it is possible to adjust the training plan and increase the effectiveness of the training. [1] The autonomous regulation of heart rate (HR) and its acute and
is the evaluation of heart rate variability (HRV). HRV features reflect the autonomic nervous system (ANS) function and thus provide information about reaction of body to sport training. Based on the HRV analysis, it is possible to adjust the training plan and increase the effectiveness of the training. [1] The autonomous regulation of heart rate (HR) and its acute and chronic adaptation to exercise has been very discussed topics in sport training for the last 20 years [2,3]. The understanding of individual athlete’s reactions to different types of training is crucial to remove physiological factors limiting endurance performance [4]. From the essential training variables, exercise intensity and its distribution are probably the most critical and the most heavily debated [3]. If the athlete’s reaction to activities is not known, the training schedule cannot be adjusted. Thus, HRV analysis can be very helpful to solve this task. The application of HRV analysis in sports became easier and affordable thanks to the smart devices such as sport smartwatches or a combination of smartwatch and chest belt. These devices enable electrocardiographic (ECG) and photoplethysmographic (PPG) signal sensing. From these data, HRV features can be calculated. Increased resting HRV is usually associated with the good fitness of the athlete and positive adaptation to endurance training [5]. It manifests subjectively (wellbeing) as well as objectively (e.g. by measuring VO2max). Decreased HRV corresponds with fatigue or illness [5]. Extensive research has been carried out regarding the general aspects and mechanisms of autonomic cardiovascular regulation during exercise and recovery [3,5,6] in different sports. However, there are only a few studies devoted to comparing ANS responses for different types of sport activities. This is probably because elite athletes primarily train only one specific sport (e.g. long- distance runners train primarily running) to increase their performance. Other sport activities are considered more as active recovery (cycling) or rehabilitation (yoga) or as a targeted increase in muscular strength (strength training, plyometric training) than as the main pillar of the training. That is the reason why researchers do not take into consideration more and various training activities. This means there is
train primarily running) to increase their performance. Other sport activities are considered more as active recovery (cycling) or rehabilitation (yoga) or as a targeted increase in muscular strength (strength training, plyometric training) than as the main pillar of the training. That is the reason why researchers do not take into consideration more and various training activities. This means there is no knowledge and experience about the influence of ANS through HRV between different sport activities. Comparing different studies cannot provide correct information, because athletes from specific sport specializations will react to other sport activities differently (e.g. reaction of cyclist on running TS at defined intensity and reaction of runner on the same TS will be different). Computing in Cardiology 2022; Vol 49 Page 1 ISSN: 2325-887X DOI: 10.22489/CinC.2022.265
It is important to know how different types of activities (running TS, special TS for a targeted increase of strength, etc.) affect athletes. In this study, HRV was used to clarify how different types of running TS – low-intensity (LIT), high-intensity interval training (HIIT) – and plyometric TS influence post-exercise ANS response. The evaluation of training response to TS is usually addressed in four ways. First, values of HRV features measured 10 min after TS are calculated [6]. Second, values of HRV features measured the morning after TS (after waking up) are compared with values measured in the morning of the TS day [7]. Third, lnRMSSD is calculated after performing the orthostatic test the morning after TS [8]. Fourth, the change between pre-exercise and post-exercise values of HRV features is calculated [9]. In this study, the fourth method was used. Using this method, it can be accurately evaluated how TS influences ANS. Using any other method, various factors (such as work, psychological stress or sleep) may affect ANS and, thus, distort the results. 2. Methods 2.1. Data collection Altogether, 12 highly trained runners (18-35 years) participated in this study. Each subject completed some of these three types of TS – LIT running, HIIT running and plyometrics. Specifically, 10 LIT TS (2 athletes performed this TS twice), 12 HIIT TS, 9 plyometric TS (2 athletes performed this TS twice) were performed. 5-min pre-exercise ECGs were taken just before TS and 5-min post-exercise ECGs were taken 10 min after TS. ECG signal was sensed by Bittium Faros 180 ECG device with sampling frequency of 1,000 Hz. From these signals, HRV features were computed. 2.2. Training sessions description Three types of TS were performed in this study – plyometric TS, LIT TS, and HIIT TS. The plyometric TS was defined as 10 min warm-up (5 min easy jogging and 5 min stretching) followed by 4 series of 6× basic box jump, 6× squat jump, 6× left leg vertical jump, 6× right leg vertical jump, 6× incline jump and 6× push up with a clap. Finally, the 5 min shakeout was performed. Plyometrics increases
HIIT TS. The plyometric TS was defined as 10 min warm-up (5 min easy jogging and 5 min stretching) followed by 4 series of 6× basic box jump, 6× squat jump, 6× left leg vertical jump, 6× right leg vertical jump, 6× incline jump and 6× push up with a clap. Finally, the 5 min shakeout was performed. Plyometrics increases neuro-muscular coordination by training the nervous system and making movements more automatic during activity [10,11]. Numerous authors have described increased jump height, reduced sprint time, improved running economy, coordination, and postural control as a result of plyometric training [10,11]. LIT TS was defined as 50 min of easy continuous running at 60-75% of maximal HR (MHR). LIT TS improves mainly cardiovascular performance and endurance and enables better delivery of oxygen to muscles. HIIT TS was defined as 15 min warm-up (easy running at 60-75% of MHR) followed by 6×2.5 min running at 90-95% of MHR with 2 min rest between intervals, ended by 10 min easy running at 60-75% of MHR. HIIT TS improves mainly the ability to run longer and faster at a higher intensity. 2.3. HRV features extraction Thirteen different pre- and post-exercise HRV features often used in other studies were calculated [12,13]. The features are described in Table 1 in detail. Table 1. Description of HRV features used in this study. Feature Units Description SDNN ms The standard deviation of NN intervals. It reflects sympathetic and parasympathetic activities and describes overall variability or total power. RMSSD ms Root mean square of successive NN intervals differences. It describes parasympathetic activity. lnRMSSD ms Natural logarithm of RMSSD. It describes parasympathetic activity. NN50 count Number of successive NN intervals that differ by more than 50 ms. It reflects mainly parasympathetic activity. pNN50 % Percentage of successive NN intervals that differ by more than 50 ms. LFpower n.u. Low frequency power. Power in frequency band 0.04–0.15 Hz. It reflects sympathetic activity. HFpower n.u. High frequency power. Power in frequency band 0.15–0.4 Hz. It reflects mainly parasympathetic activity. LFperc % Relative LFpower – percentage of LF power to total power. HFperc %
activity. pNN50 % Percentage of successive NN intervals that differ by more than 50 ms. LFpower n.u. Low frequency power. Power in frequency band 0.04–0.15 Hz. It reflects sympathetic activity. HFpower n.u. High frequency power. Power in frequency band 0.15–0.4 Hz. It reflects mainly parasympathetic activity. LFperc % Relative LFpower – percentage of LF power to total power. HFperc % Relative HFpower – percentage of HF to total power. LHratio - LF to HF power ratio. totPower n.u. Total frequency power – power in whole frequency band 0–0.4 Hz. SD1 ms Standard deviation of the Poincaré plot (short-term variability). SD2 ms Standard deviation of the Poincaré plot (long-term variability). Page 2
2.3. Statistical analysis The changes between pre-exercise and post-exercise values of HRV features were computed. They are expressed in percentage of pre-exercise values. At first, the Shapiro-Wilk test was used to reveal data distribution. The data were not normally distributed. Then, non-parametric Kruskal-Wallis test (p < 0.05) was used to find features which are suitable for discrimination between various TS. Finally, a post-hoc analysis (p < 0.05) was performed to reveal which pairs of features (TS) showed statistically significant differences. 3. Results Statistical analysis showed which HRV features differ between TS types. The results (p-values) of post-hoc tests are shown in Table 2. Features which show statistical significance are highlighted in green and blue. From 13 tested features, 9 features were statistically significant for distinguishing between plyometrics and LIT TS. None of the features were statistically significant for distinguishing between plyometrics and HIIT TS. 10 features were statistically significant for distinguishing between HIIT and LIT TS. The boxplots representing the features values (percentage difference between pre-exercise and post- exercise) are shown in Figure 1. All these features (except for LFpower) showed the highest drop in values of such feature after plyometric TS, followed by HIIT TS and the lowest drop (or even increase) showed LIT TS. These features reflect parasympathetic activity or a combination of parasympathetic and sympathetic activity. In contrast, LFpower showed increase for all TS, the most at HIIT, then at plyometric TS, and the lowest increase was at LIT. LFpower reflects mainly sympathetic activity. Table 2. Table of post-hoc tests p-values. Differences of HRV features were tested between three types (pairs) of TS. Green color indicates statistically significant differences (p < 0.05) between Plyo and LIT TS and between HIIT and LIT TS. Blue color indicates statistically significant difference between HIIT and LIT TS only. Features Plyo-LIT Plyo-HIIT HIIT-LIT SDNN 0.0003 0.3038 0.0031 RMSSD 0.0041 0.4592 0.0173 NN50 0.0006 0.3798 0.0033 pNN50 0.0008 0.4043 0.0036 LFpower 0.4472 0.7883 0.0361 HFpower 0.0027 0.4611 0.0106 LFperc 0.5578 0.7612 0.8860 HFperc 0.4049 0.5396 0.9355 LHratio 0.4181 0.5868 0.9100 totPower 0.0023 0.6026 0.0033 lnRMSDD 0.0028 0.3993 0.0171 SD1 0.0041 0.4592 0.0173
significant difference between HIIT and LIT TS only. Features Plyo-LIT Plyo-HIIT HIIT-LIT SDNN 0.0003 0.3038 0.0031 RMSSD 0.0041 0.4592 0.0173 NN50 0.0006 0.3798 0.0033 pNN50 0.0008 0.4043 0.0036 LFpower 0.4472 0.7883 0.0361 HFpower 0.0027 0.4611 0.0106 LFperc 0.5578 0.7612 0.8860 HFperc 0.4049 0.5396 0.9355 LHratio 0.4181 0.5868 0.9100 totPower 0.0023 0.6026 0.0033 lnRMSDD 0.0028 0.3993 0.0171 SD1 0.0041 0.4592 0.0173 SD2 0.0002 0.2560 0.0029 Figure 1. Boxplots of HRV features with statistically significant differences between TS (Plyo – Plyometric, LIT – light-intensity running training, HIIT – high-intensity interval running training). On the y-axis there are differences between pre-exercise and post-exercise values of HRV features expressed in percentage of pre-exercise value. 4. Conclusion In this study, HRV analysis was used to clarify how different types of running TS (LIT and HIIT) and plyometric TS influence post-exercise ANS response. Firstly, 13 HRV features were computed. Then the difference between pre-exercise and post-exercise values was expressed in percentage of pre-exercise value. Statistical analysis was performed. Kruskal-Wallis test proved that 10 HRV features show statistically significant changes between different types of TS. More detailed post-hoc tests showed that there are not statistically significant differences in HRV changes Page 3
between plyometric and HIIT TS. It could be therefore assumed that these two TS affect ANS similarly. 9 and 10 HRV features showed statistically significant differences between plyometric and LIT TS and between HIIT and LIT TS, respectively. Nowadays, in training practice, it is usually considered that HIIT running sessions lead to the highest load. According to our findings, we assume that plyometric TS and HIIT TS cause similar load for the organism. This finding should be considered when preparing training plans for runners and thus increase the effectiveness of training and prevent injury. Acknowledgments This work has been funded by the United States Office of Naval Research (ONR) Global, award number N62909- 19-1-2006. The authors wish to thank LCDR Joshua Swift from ONR Code 342 and Dr. Martina Siwek from ONR Global Central and Eastern European Office for their support. References [1] K. Hottenrott and O. Hoos, Eds., “Heart Rate Variability Analysis in Exercise Physiology”, in ECG Time Series Variability Analysis, Boca Raton: CRC Press, 2017, pp. 245-276. [2] S. Michael, O. Jay, M. Halaki, K. Graham, and G. M. Davis, “Submaximal exercise intensity modulates acute post- exercise heart rate variability”, European Journal of Applied Physiology, vol. 116, no. 4, pp. 697-706, 2016. [3] G. Davies, B. L. Riemann, and R. Manske, “Current Concepts of Plyometric Exercise”, Int J Sports Phys Ther, vol. 10, no. 6, pp. 760-786, 2015. [4] A. M. Turner, M. Owings, and J. A. Schwane, “Improvement in running economy after 6 weeks of plyometric training”, J Strength Cond Res, vol. 17, no. 1, pp. 60-67, 2003. [5] P. Kaikkonen, H. Rusko, and K. Martinmäki, “Post-exercise heart rate variability of endurance athletes after different high-intensity exercise interventions”, Scandinavian Journal of Medicine & Science in Sports, vol. 18, no. 4, pp. 511-519, 2008. [6] V. Vesterinen, K. Häkkinen, T. Laine, E. Hynynen, J. Mikkola, and A. Nummela, “Predictors of individual adaptation to high-volume or high-intensity endurance training in recreational endurance runners”, Scandinavian Journal of Medicine & Science in Sports, vol. 26, no. 8, pp. 885-893, 2016. [7] R. Šlachta, “MySASY”, Variabilita srdeční frekvence a její monitoring prostřednictvím
18, no. 4, pp. 511-519, 2008. [6] V. Vesterinen, K. Häkkinen, T. Laine, E. Hynynen, J. Mikkola, and A. Nummela, “Predictors of individual adaptation to high-volume or high-intensity endurance training in recreational endurance runners”, Scandinavian Journal of Medicine & Science in Sports, vol. 26, no. 8, pp. 885-893, 2016. [7] R. Šlachta, “MySASY”, Variabilita srdeční frekvence a její monitoring prostřednictvím diagnostiky mySASY, 0c2022. [8] J. Hayano and E. Yuda, “Pitfalls of assessment of autonomic function by heart rate variability”, Journal of Physiological Anthropology, vol. 38, no. 1, 2019. [9] C. Schneider, T. Wiewelhove, C. Raeder, A. A. Flatt, O. Hoos, L. Hottenrott, O. Schumbera, M. Kellmann, T. Meyer, M. Pfeiffer, and A. Ferrauti, “Heart Rate Variability Monitoring During Strength and High-Intensity Interval Training Overload Microcycles”, Frontiers in Physiology, vol. 10, May 2019. [10] K. S. Seiler and G. O. Kjerland, “Quantifying training intensity distribution in elite endurance athletes: is there evidence for an "optimal" distribution?”, Scandinavian Journal of Medicine and Science in Sports, vol. 16, no. 1, pp. 49-56, 2006. [11] E. F. Coyle, “Integration of the Physiological Factors Determining Endurance Performance Ability”, Exercise and Sport Sciences Reviews, vol. 23, 1995. [12] K. Kiyono, J. Hayano, E. Watanabe, and Y. Yamamoto, “Heart Rate Variability (HRV) and Sympathetic Nerve Activity”, in Clinical Assessment of the Autonomic Nervous System, Tokyo: Springer, 2017, pp. 147-161. Address for correspondence: Ing. Lucie Šaclová Department of Biomedical Engineering Faculty of Electrical Engineering and Communication Brno University of Technology Technická 12 Brno 616 00 Czech Republic marsanova@vutbr.cz Page 4