Heart Rate Variability-Established Thresholds to Determine the Ventilatory and Lactate Thresholds of Endurance Athletes

his study aimed to determine the use of heart rate variability (HRV)-established thresholds to accurately estimate endurance athletes’ ventilatory threshold 1 (VT1), respiratory compensation point (RCP) and lactate threshold 2 (LT2). Eleven cyclists (aged: 23.7 ± 3.1 years) from the African Continental Development Cycling team and ten middle- and long-distance athletes (age: 21.2 ± 1.8 years) from a South African university participated in this study. Before the start of an incremental cycling or running maximal oxygen consumption (V̇O2max) test each participant was fitted with a Fixed Polar HR Transmitter Belt and Monitor to determine the R-R intervals of the last 60 seconds of each stage. The Kubios HRV Premium software package was used to analyze the R-R-intervals. Blood samples were taken 30 seconds before the end of each stage and analyzed for blood lactate. Ventilatory threshold points were identified using the criteria of an increase in V̇ E/V̇O2 with no increase in V̇ E/V̇CO2 (VT1) and an increase in both V̇E/V̇O2 and V̇ E/V̇CO2 (RCP). The LT2 was

Premium software package was used to analyze the R-R-intervals. Blood samples were taken 30 seconds before the end of each stage and analyzed for blood lactate. Ventilatory threshold points were identified using the criteria of an increase in V̇ E/V̇O2 with no increase in V̇ E/V̇CO2 (VT1) and an increase in both V̇E/V̇O2 and V̇ E/V̇CO2 (RCP). The LT2 was identified as the second increase in blood lactate concentration from one increment to the next. Concerning HRV thresholds, VT1 was determined at a DFAα1 value of 0.75, the first breakpoint in the standard deviation of the instantaneous (SD1) and continuous long-term RR interval (SD2) curves, the visual deflection in the squared root of the mean squared differences between successive R-R intervals (RMSSD) curve, and the first abrupt increase in high-frequency (HF) power x HF frequency. The RCP was detected at a DFAα1 value of 0.5 and the final abrupt increase in HF. The Bland-Altman plots revealed that the absolute power outputs at VT1SD1 and VT1RMSSD showed significant agreements with the absolute power outputs at VT1. Large paired-sample correlations were also found between the absolute power outputs at VT1SD1, VT1RMSSD and VT1. In conclusion, coaches, sport scientists and other professionals are encouraged to use SD1 and RMSSD to determine the VT1-related training program workloads of endurance-trained athletes, especially in areas where laboratories are not available. Keywords Ventilatory Threshold, Lactate Threshold, Heart Rate Variability, Heart Rate Variability Thresholds 1. Introduction The estimation of the ventilatory and lactate thresholds is important as it allows sport-related practitioners and scientists to adapt training intensities and monitor and assess endurance performance [1,2]. Last-mentioned thresholds serve as important indicators of individual exercise tolerance which refers to the ability of an individual to perform higher intensities of exercise for longer durations [3]. For this reason, threshold-based

International Journal of Human Movement and Sports Sciences 11(2): 398-410, 2023 399 training models allow practitioners to set individualized training intensities for endurance sport as it accounts for individual metabolic responses to training [4]. The most recognized and used methods for the determination of these thresholds are indirect calorimetry, open-circuit spirometry, and computerized instrumentation as well as blood lactate sampling during a graded exercise test [5]. The ventilatory threshold 1 or aerobic threshold (VT1) is regarded to be the point where the ventilatory equivalent for oxygen (V̇ E/V̇O2) increases nonlinearly while the ventilatory equivalent for carbon dioxide (V̇ E/V̇CO2) remains constant [6]. The VT1 can also be defined as the point where the first rise in the fraction of expired oxygen (PETO2) occurs, while the fraction of expired carbon dioxide (PETCO2) either increases or remains constant [7]. In addition, the intersection between V̇CO2 and V̇O2 (V-slope) graphs where the slope changes from less than one to equal or greater than one, is also used to identify the VT1 [7]. The ventilatory threshold 2 (VT2), anaerobic threshold (AT) or respiratory compensation point (RCP) is the point where an increase in both V̇ E/V̇O2 and V̇ E/V̇CO2 occur and a nonlinear increase in the exhaled carbon dioxide volume (V̇CO2) compared to the inhaled V̇O2 is observed [6,8]. The deflection point of the end-tidal PETCO2 is also used to identify the VT2 [7]. The secondary increase in arterial blood lactate concentration (BLa) is known as the second LT point (LT2) [9]. Despite proof of the reliability and validity of the last-mentioned techniques to determine the thresholds of different populations of athletes, these techniques require expensive equipment which needs to be calibrated frequently, are invasive and need expert personnel to operate [10-15]. Given these difficulties, alternative quick-to-determine, practical, inexpensive-to-obtain, and easy-to-interpret measures are rather needed to determine the thresholds of athletes. In this regard, heart rate variability (HRV) may serve as an easier, non-invasive, and cost-effective measure to determine athletes’ thresholds [5,16,17]. Heart rate variability (HRV) serves as a parameter of the autonomic nervous system (ANS) functioning and reflects the sympathetic (SNS) and parasympathetic

difficulties, alternative quick-to-determine, practical, inexpensive-to-obtain, and easy-to-interpret measures are rather needed to determine the thresholds of athletes. In this regard, heart rate variability (HRV) may serve as an easier, non-invasive, and cost-effective measure to determine athletes’ thresholds [5,16,17]. Heart rate variability (HRV) serves as a parameter of the autonomic nervous system (ANS) functioning and reflects the sympathetic (SNS) and parasympathetic nervous system (PNS) activity of the ANS [18]. Heart rate variability (HRV) refers to the time between heart beats or the R-R intervals [12,16,18]. The progressive workloads of a graded exercise test will lead to a gradual and linear increase in sympathetic activity in relation to the imposed workload with concurrent and gradual parasympathetic withdrawal [19]. At more or less 60 to 80% of maximal oxygen consumption (V̇O2max) total parasympathetic withdrawal occurs followed by a drastic increase in sympathetic activity [20]. Therefore, the workloads at the last-mentioned intensities are characterized by a significant reduction in HRV, which is termed the HRV threshold [21,22]. Consequently, ventilatory thresholds may be under the same nervous system control as HRV, which means that the different thresholds can possibly be determined by examining the changes in HRV during exercise [12,23,24]. Proof that the nervous system control is similar for HRV and the ventilatory thresholds was provided by several researchers who observed strong relationships between HRV and the different threshold points [5,14,23]. For instance, a significant correlation was found for different HRV-threshold indexes (SD1 and RMSSD) of healthy male participants and the visual inspection of the workload (r = 0.58-0.96) and V̇O2 (ml/kg/min) (r = 0.70-0.98) [25]. The Bland-Altman technique also showed a good agreement between VT1 and the standard deviation (perpendicular to the line of identity within the Poincaré plot) of correlations between R-R intervals denoted as SD1 (-0.7 ± 7.98 ml/kg/min), as well as between VT1 and the squared root of the mean squared differences between successive R-R intervals (RMSSD) (-3.1 ± 7.54 ml/kg/min) [25]. In addition, the relationships between the gas analysis- and HRV-related methods for determining the VT1 in Spanish professional basketball players indicated significant correlations between VT1 V̇O2 (r = 0.54,

denoted as SD1 (-0.7 ± 7.98 ml/kg/min), as well as between VT1 and the squared root of the mean squared differences between successive R-R intervals (RMSSD) (-3.1 ± 7.54 ml/kg/min) [25]. In addition, the relationships between the gas analysis- and HRV-related methods for determining the VT1 in Spanish professional basketball players indicated significant correlations between VT1 V̇O2 (r = 0.54, p ˂ 0.005), VT1 HR (r = 0.57 p ˂ 0.005), VT1 speed (r = 0.47, p ˂ 0.005) and SD1 [12]. However, stronger correlations were observed between VT2 V̇O2 (r = 0.91, p ˂ 0.005), VT2 HR (r = 0.90, p ˂ 0.005), VT2 speed (r = 0.93, p ˂ 0.005) and peak frequency in the high-frequency range (ƒHF) [12]. On the other hand, Cassirame et al. [1] found no significant correlations (p > 0.05) between the ƒHF of elite ski-mountaineers, the HR and the speed at the VT1, but significant strong correlations between ƒHF, the HR (r = 0.91, p ˂ 0.001) and the speed at the VT2 (r = 0.92, p ˂ 0.001). Another HRV parameter that can be used as an indicator of VT1 or LT1 is the short-term scaling exponent alpha 1 of the nonlinear detrended fluctuation analysis method (DFA1) with a value of ~0.75 [21]. In this regard, results from a group of elite endurance cyclists showed significant correlations between power output/HR at LT1 and power output/HR at DFA 1 (~0.75) (r = 0.85, p < 0.001) as well as between power outputs/HRs at LT2 and DFA1 (~0.5) (r = 0.93, p < 0.001) but no significant correlations were found between HRs at LT1 and DFA1 (~0.75) [22]. Although researchers also use ~ 0.5 as the cut-off value for the identification of RCP, the accuracy of this value is still being debated [21]. Other researchers did not only report significant strong correlations between the VT1 and the heart rate ventilatory threshold (HRVT) of healthy participants (r = 0.92, p ˂ 0.001) but also between LT2 and HRVT (r = 0.85, p ˂ 0.001) [14]. These results are very similar to the significant strong correlation

accuracy of this value is still being debated [21]. Other researchers did not only report significant strong correlations between the VT1 and the heart rate ventilatory threshold (HRVT) of healthy participants (r = 0.92, p ˂ 0.001) but also between LT2 and HRVT (r = 0.85, p ˂ 0.001) [14]. These results are very similar to the significant strong correlation that was found between the LT2 (182.0 ± 8.1 b·min -1 ) and power HF (Hz) (181.1 ± 8.2 b·min -1 ) of high-level swimmers when researchers assessed HR (r = 0.93, p ˂ 0.05) and velocity (r = 0.98, p < 0.05) [16]. In another study of recreational long-distance male runners, a significant correlation was observed between the second HRV threshold (HRVT2) and

400 Heart Rate Variability-Established Thresholds to Determine the Ventilatory and Lactate Thresholds of Endurance Athletes the LT2 (r = 0.86, p ˂ 0.01) [13]. These results gave rise to the conclusion that good agreements exist between the LT and HRV threshold points and that HRV has the potential to be used as a non-invasive measure to determine LT during a maximal incremental running test [13]. The above-mentioned findings suggest that HRV-related parameters, which serve as indicators of cardiovascular ANS function, may serve as indicators of the VT and LT points. However, up until now, researchers narrowed their focus on the use of very specific HRV-related variables when investigating the usefulness and accuracy of HRV-related parameters to determine different thresholds in a wide range of non- and sport-participating populations [1,5,11,13,14,16,22,24,25]. No researchers used a variety of available HRV-related measures to examine their accuracy in determining different thresholds in high-level endurance athletes. Therefore, the purpose of this study was to determine whether the HRV-established threshold absolute power outputs can be used to accurately estimate the absolute power outputs at VT1, RCP and LT2 of endurance athletes. We hypothesized that the HRV-established thresholds of endurance athletes as determined through changes in DFAα1, SD1, RMSSD, and HF power can be used to accurately estimate VT1, RCP, and LT2. Study results may allow coaches, sport scientists and sport-related professionals to determine the precise threshold-related training program workloads of athletes for the continuous monitoring of training adaptations in places where expensive and sophisticated testing equipment and laboratories are not available. 2. Materials and Methods 2.1. Study Design The hypothesis of the study was tested by making use of a selected group, quasi-experimental research design. 2.2. Participants Ten middle- and long-distance athletes from a university in South Africa (M ± SD = age: 23.7 ± 3.1 years; height: 175.8 ± 6.8 cm; mass: 66.3 ± 8.8 kg), as well as eleven cyclists from the African Continental Development Cycling team (M ± SD = age: 21.2 ± 1.8 years; height: 182.1 ± 7.8 cm; mass: 73.4 ± 5.3 kg) volunteered to participate in this study. According to the

Africa (M ± SD = age: 23.7 ± 3.1 years; height: 175.8 ± 6.8 cm; mass: 66.3 ± 8.8 kg), as well as eleven cyclists from the African Continental Development Cycling team (M ± SD = age: 21.2 ± 1.8 years; height: 182.1 ± 7.8 cm; mass: 73.4 ± 5.3 kg) volunteered to participate in this study. According to the classification of Swann et al. [26], cyclists could be classified as elite/expert cyclists whereas athletes could be classified into three categories, namely: competitive, successful, and world-class elite athletes. The study followed the ethical principles of the Declaration of Helsinki as well as the ethical guidelines of the National Health Research Ethics Council of South Africa. The Health Research Ethics Committee of the institution where the research was conducted approved the study with the following number: NWU-00005-17-A1. Participants gave written informed consent which thoroughly explained the benefits, risks as well as expectations of study participation. 2.3. Procedures Participants completed the demographic questionnaire and were familiarized with the HRV equipment and the maximal aerobic capacity (V̇O2max) test one week before testing commenced. A week later, participants completed the maximal incremental V̇O2max tests. Before the start of the tests, participants completed a recovery and hydration status questionnaire, followed by body mass and stature measurements. Next, each participant was fitted with a Fixed Polar HR Transmitter Belt and Monitor (Polar Team 2 Pro, Polar Electro, Kempele, Finland) to record HRs and HRV where after a fifteen-minute dynamic warm-up was performed followed by a V̇O2max test. 2.4. Instruments and Tests 2.4.1. General Recovery and Hydration Status Questionnaire The general recovery and hydration status questionnaire indicated participants’ quantity as well as the quality of sleep during the previous night, their hydration status, and their degree of muscle soreness. The sleep quantity was indicated by noting the sleep duration and the sleep quality and hydration status were indicated through a 5-point Likert scale, and muscle soreness through a 3-point Likert scale [27,28,29]. Results of the general recovery and hydration status questionnaire aided in evaluating the overall recovery status of participants to ensure that they were in the

muscle soreness. The sleep quantity was indicated by noting the sleep duration and the sleep quality and hydration status were indicated through a 5-point Likert scale, and muscle soreness through a 3-point Likert scale [27,28,29]. Results of the general recovery and hydration status questionnaire aided in evaluating the overall recovery status of participants to ensure that they were in the best possible physiological state to successfully execute the V̇O2max test. 2.4.2. Standard incremental maximal oxygen uptake (V̇O2max) tests To measure the metabolic parameters during the running and cycling V̇O2max tests the Oxycon Pro METALYZER @ 3B high-resolution spiroergometry system (Cortex Biophysik GmbH, Leipzig, Germany) with breath-by-breath technology was used. The MetaSoft @ Studio application software analysed the results of the spiroergometry system. The Oxycon Pro is regarded to be a valid, reliable, and accurate system for the measurement of athletes’ and healthy participants’ metabolic parameters during low- and maximal-intensity exercise [30-32]. Before the test commenced, the gas analyzer was calibrated with standard gases. The sampled V̇O2 and V̇CO2 from the Oxygen Pro gas analyzer apparatus were used to calculate the respiratory exchange ratio (RER). Before the start of the test, each participant was fitted with a face mask and a soft nose clip. Before the start of the tests, cyclists performed a ten-minute warm-up at a resistance of 100 W on the

International Journal of Human Movement and Sports Sciences 11(2): 398-410, 2023 401 Wattbike Pro Air and Magnetic Resistance Trainer (Wattbike Limited, Nottingham, England), followed by high-intensity cycling at 250 W for 30 s, stretching and mobility exercises. The incremental cycling test was completed on the Wattbike Pro according to the adapted maximal ramp test protocol of Wattbike Ltd (2015:33-41) exactly two minutes after the warm-up. Cyclists were instructed to start at 150 W using an air resistance setting of one and to increase the power output with 35 W every two minutes. The test was terminated when the following criteria were met: when a minimum cadence of 80 rpm could not be maintained, when the criteria for V̇O2max was met (e.g. a RER-value higher than 1.15, oxygen consumption ceased to rise and reached a plateau or began to fall with an increase in work rate, and the maximum age-specific HR as calculated through the formula: Maximum HR (208–0.7*age) was reached), or they wanted to stop [33,34]. The Wattbike Pro automatically determined the absolute (W) and relative power outputs (W·kg -1 ) at the end of each increment. Before the start of the test, athletes completed a ten-minute warm-up at a running speed of 13 km·h -1 on a Woodway Pro XL Treadmill (Woodway, W229 N591, Foster Ct, Waukesha, WI), followed by a speed increase to 16 km/h for 30 s after which stretching and mobility exercises were executed. The incremental test was performed according to the adapted method of Peserico et al. [35] exactly two minutes after the warm-up. The test commenced at an incline of 1˚ and a speed of 12 km·h -1 , increasing with 1 km·h -1 every two minutes until volitional exhaustion. The test was terminated when the participants indicated they wanted to stop, when they failed to maintain the speed of the incremental level for at least 30 s, or when the V̇O2max criteria were reached. The studio application software automatically determined the absolute (W) and relative power outputs (W·kg -1 ) at the end of each incremental stage. Two physiological gas exchange points were

when the participants indicated they wanted to stop, when they failed to maintain the speed of the incremental level for at least 30 s, or when the V̇O2max criteria were reached. The studio application software automatically determined the absolute (W) and relative power outputs (W·kg -1 ) at the end of each incremental stage. Two physiological gas exchange points were also identified from the standard incremental V̇O2max test data. The ventilatory threshold 1 (VT1) was determined using the criteria of an increase in V̇ E/V̇O2 with no increase in V̇ E/V̇CO2 and departure from the linearity of V̇ E [5]. The ventilatory threshold 2 (VT2) or RCP was taken as the point which corresponds to an increase in both V̇ E/V̇O2 and V̇ E/V̇CO2 [8]. VT1 and RCP were visually detected by two independent researchers, with no discrepancies between the visually detected values of the two researchers. Throughout the V̇O2max tests, blood samples were taken 30 seconds before the end of each stage and analysed with a Simplified Lactate Pro LT-1710 Blood Lactate Test Meter (Arkray Factory Inc., KDK Corporation, Shiga, Japan). The LT2 was determined by the identification of the secondary increase in arterial blood lactate concentration from one increment level to the next [9,36]. To record the variations in beat-to-beat intervals (R-R intervals) during the standard incremental V̇O2max test a Fix Polar Heart Rate Monitor and Transmitter Belt were used. These monitors are valid devices for obtaining R-R intervals [37]. The Polar device-obtained R-R-intervals were analysed via the Kubios HRV Premium software package (Version 3.4.2, Biosignal Analysis and Medical Imaging Group at the Department of Applied Physics, University of Kuopio, Kuopio, Finland). The Polar Team software package was used to export the R-R-related data sets from the Polar Monitors and the “automatic artefact correction” function was selected to automatically filter out the erroneous values before it was exported as a raw R-R data set [38]. The R-R intervals of the last 60 seconds of each stage were used to measure HRV. A one-minute window from the R-R data was used to measure DFAα1 [5]. The window width of

Monitors and the “automatic artefact correction” function was selected to automatically filter out the erroneous values before it was exported as a raw R-R data set [38]. The R-R intervals of the last 60 seconds of each stage were used to measure HRV. A one-minute window from the R-R data was used to measure DFAα1 [5]. The window width of the DFAα1 was set to 4 < n < 16 beats. To detect the VT1 a DFA α1 value of 0.75 was selected as this is also the midpoint between a fractal behavior of the HR time series of 1.0 (for low-intensity exercise) and an uncorrelated value of 0.5 (for high-intensity exercise) representing random behavior of the HR time series [21]. The first breakpoint in the SD1 and SD2 curves of the Poincaré plot was used to detect VT1 [11,13]. To determine VT1 through RMSSD a visual deflection in the curve of intensity versus RMSSD indexes was identified [25]. For power HF, VT1 was calculated from the R-R interval series using a time-varying short-term Fourier transform with a 64-second moving window [1,23]. The first abrupt increase in Power HF x HF frequency after it has reached a minimum was used to determine VT1 [1,23]. A DFAα1 value of 0.5 was used to detect the second threshold (RCP) [21]. Thereafter, the breakpoints of DFAα1 were matched to the power output values obtained during the incremental test. RCP was determined from HF power by identifying the final abrupt increase in the HF band [11,23]. 2.5. Data Analysis The descriptive statistics of each variable were first calculated. Next, the Shapiro-Wilk test for data normality was performed with a p-value of less than 0.05 determined as the cut-off point for the rejection of the null hypothesis that the data are normally distributed. Thereafter, paired-sample T-tests and correlation analyses were performed to determine whether significant differences or associations exist between the different HRV-, ventilatory- and lactate-established threshold points. The strength of correlations was categorized based on the following criteria: <0.1 (trivial), <0.3 (small), <0.5 (moderate), <0.7 (large), <0.9 (very large) and <1 (nearly perfect) [39].

The study investigates HRV thresholds for estimating ventilatory and lactate thresholds in endurance athletes.