Heart Rate Variability-Derived Thresholds for Exercise Intensity Prescription in Endurance Sports: A Systematic Review of Interrelations and Agreement with Different Ventilatory and Blood Lactate Thresholds

Morning versus Nocturnal Heart Rate and Heart Rate Variability Responses to Intensified Training in Recreational Runners

BACKGROUND

Resting heart rate (HR) and HR variability (HRV) are widely used parameters to assess cardiac autonomic nervous system function noninvasively. While resting assessments can be performed during sleep or after awakening, it would be relevant to know how interchangeable the results of these measurements are. This study aimed at examining the alignment between nocturnal and morning assessments during regular endurance training and in response to intensive training. A total of 24 recreational runners performed a 3-week baseline period (BL) and a 2-week overload (OL) period (Lucia's training impulse + 80%). Their running performance was assessed with a 3000-m running test after the BL and OL. The participants recorded daily their nocturnal HR and HRV (the natural logarithm of the root mean square of successive differences; LnRMSSD) with a photoplethysmography-based wrist device and performed an orthostatic test (2-min supine, 2-min standing) every morning with a chest-strap HR sensor. The HR and LnRMSSD segments that were analyzed from the nocturnal recordings included start value (SleepStart), end value (SleepEnd), first 4-h segment 30 min after detected sleep onset (Sleep4h), and full sleep time (SleepFull). The morning segments consisted of the last-minute average in both body positions. All segments were compared at BL and in response to the 3000-m test and OL.

RESULTS

All nocturnal HR and LnRMSSD segments correlated with supine and standing segments at BL (r = 0.42 to 0.91, p < 0.05). After the 3000-m test, the HR increased and LnRMSSD decreased only in the SleepStart, Sleep4h, and SleepFull segments (p < 0.05). In response to the OL, the standing HR decreased (p < 0.01), while the LnRMSSD increased (p < 0.05) in all segments except for SleepStart. The Pearson correlations between relative changes in nocturnal and morning segments were - 0.11 to 0.72 (3000-m) and - 0.25 to 0.79 (OL). The OL response in Sleep4h HR and LnRMSSD correlated with the relative change in 3000-m time (r = 0.63, p = 0.001 and r=-0.50, p = 0.013, respectively).

CONCLUSIONS

Nocturnal and morning HR and LnRMSSD correlated moderately or highly in the majority of cases during the BL, but their responses to intensive training were not similarly aligned, especially in LnRMSSD. The nocturnal segments seemed to be sensitive to physical loading, and their responses were associated with the performance-related training responses.

Exercise intensity measurement using fractal analysis of heart rate variability: Reliability, agreement and influence of sex and cardiorespiratory fitness

The study aimed to establish the test-retest reliability of detrended fluctuation analysis of heart rate variability (DFA-α1) based exercise intensity thresholds, assess its agreement with ventilatory- and lactate-derived thresholds and the moderating effect of sex and cardiorespiratory fitness (CRF) on the agreement. Intensity thresholds for thirty-seven participants (17 females) based on blood lactate (LT1/LT2), gas-exchange (VT1/VT2) and DFA-α1 (αTh1/αTh2) were assessed. Heart rate (HR) at αTh1 and αTh2 showed good test-retest reliability (coefficient of variation [CV] < 6%), and moderate to high agreement with LTs (r = 0.40 - 0.57) and VTs (r = 0.61 - 0.66) respectively. Mixed effects models indicated bias magnitude depended on CRF, with DFA-α1 overestimating thresholds versus VTs for lower fitness levels (speed at VT1 <8.5 km⋅hr-1), while underestimating for higher fitness levels (speed at VT2 >15 km⋅hr-1; VO2max >55 mL·kg-1·min-1). Controlling for CRF, sex significantly affected bias magnitude only at first threshold, with males having higher mean bias (+2.41 bpm) than females (-1.26 bpm). DFA-α1 thresholds are practical and reliable intensity measures, however it is unclear if they accurately represent LTs/VTs from the observed limits of agreement and unexplained variance. To optimise DFA-α1 threshold estimation across different populations, bias should be corrected based on sex and CRF.

Agreement Between Heart Rate Variability - Derived vs. Ventilatory and Lactate Thresholds: A Systematic Review with Meta-Analyses

BACKGROUND

Determining thresholds by measuring blood lactate levels (lactate thresholds) or gas exchange (ventilatory thresholds) that delineate the different exercise intensity domains is crucial for training prescription. This systematic review with meta-analyses aims to assess the overall validity of the first and second heart rate variability - derived threshold (HRVT1 and HRVT2, respectively) by computing global effect sizes for agreement and correlation between HRVTs and reference - lactate and ventilatory (LT-VTs) - thresholds. Furthermore, this review aims to assess the impact of subjects' characteristics, HRV methods, and study protocols on the agreement and correlation between LT-VTs and HRVTs.

METHODS

Systematic computerised searches for studies determining HRVTs during incremental exercise in humans were conducted. The agreements and correlations meta-analyses were conducted using a random-effect model. Causes of heterogeneity were explored by subgroup analysis and meta-regression with subjects' characteristics, incremental exercise protocols, and HRV methods variables. The methodological quality was assessed using QUADAS-2 and STARDHRV tools. The risk of bias was assessed by funnel plots, fail-safe N test, Egger's test of the intercept, and the Begg and Mazumdar rank correlation test.

RESULTS

Fifty included studies (1160 subjects) assessed 314 agreements (95 for HRVT1, 219 for HRVT2) and 246 correlations (82 for HRVT1, 164 for HRVT2) between LT-VTs and HRVTs. The standardized mean differences were trivial between HRVT1 and LT1-VT1 (SMD = 0.08, 95% CI -0.04-0.19, n = 22) and between HRVT2 and LT2-VT2 (SMD = -0.06, 95% CI -0.15-0.03, n = 42). The correlations were very strong between HRVT1 and LT1-VT1 (r = 0.85, 95% CI 0.75-0.91, n = 22), and between HRVT2 and LT2-VT2 (r = 0.85, 95% CI 0.80-0.89, n = 41). Moreover, subjects' characteristics, type of ergometer, or initial and incremental workload had no impact on HRVTs determination.

CONCLUSION

HRVTs showed trivial differences and very strong correlations with LT-VTs and might thus serve as surrogates. These results emphasize the usefulness of HRVTs as promising, accessible, and cost-effective means for exercise and clinical prescription purposes.

Detrended fluctuation analysis to determine physiologic thresholds, investigation and evidence from incremental cycling test

PURPOSE

Training zones are generally assessed by gas-exchange thresholds (GET). Several mathematical analyses of heart rate variability (HRV) are proposed for indirect GET determination. Our study aimed to investigate the accordance of the detrend fluctuation analysis (DFA α1) for determining GET with first (VT1) and second ventilatory (VT2) thresholds in well-trained subjects.

METHODS

Eighteen female and 38 male sub-elite cyclists performed a maximal incremental cycling test of 2-min stage duration with continuous gas exchange and HR measurements. Power output (PO), Oxygen uptake ( V ˙ O2) and HR at VT1 and VT2 were compared with DFA α1 0.75 (HRVT1) and 0.50 (HRVT2). Agreements between PO, V ˙ O2 and HR values were analyzed using Bland-Altman analysis.

RESULTS

Large limits of agreement between VT1 and HRVT1 were observed for measures of V ˙ O2 expressed in mL.min-1.kg-1 [- 21.3; + 14.1], HR [ 39.2; + 26.9] bpm and PO [- 118; + 83] watts. Indeed, agreements were also low between VT2 and HRVT2 for measures of V ˙ O2 [- 26.7; + 4.3] mL.min-1.kg-1, HR [- 45.5; + 10.6] bpm and PO [- 157; + 35] watts. Our results also showed a sex effect: women obtained worst predictions based on DFA α1 than men for HR (p = 0.014), PO (p = 0.044) at VT1 andV ˙ O 2 (p = 0.045), HR (p = 0.003) and PO (p = 0.004) at VT2.

CONCLUSION

There was unsatisfactory agreement between the GET and DFA α1 methods for VT1 and VT2 determination in both sex well-trained cyclists. Trial registration number 2233534 on 2024/03/05 retrospectively registered.

Quantifying exercise intensity with fractal correlation properties of heart rate variability: a study on incremental and constant-speed running

The short-term scaling exponent of detrended fluctuation analysis (DFAα1) applied to interbeat intervals may provide a method to identify ventilatory thresholds and indicate systemic perturbation during prolonged exercise. The purposes of this study were to (i) identify the gas exchange threshold (GET) and respiratory compensation point (RCP) using DFAα1 values of 0.75 and 0.5 from incremental exercise, (ii) compare DFAα1 thresholds with DFAα1 measures during constant-speed running near the maximal lactate steady state (MLSS), and (iii) assess the repeatability of DFAα1 between MLSS trials. Twelve runners performed an incremental running test and constant-speed running 5% below, at, and 5% above the MLSS, plus a repeat trial at MLSS. During 30-min running trials near MLSS, DFAα1 responses were variable (i.e., 0.27-1.24) and affected by intensity (p = 0.031) and duration (p = 0.003). No difference in DFAα1 was detected between MLSS trials (p = 0.597). In the early phase (~ 8 min), DFAα1 measures at MLSS (0.71 [0.13]) remained higher than the DFAα1 identified at RCP from the incremental test (0.57 [0.13]; p = 0.024). In addition, following ~ 18 min of constant speed running at MLSS, DFAα1 measures (0.64 [0.14]) remained higher than 0.5 (p = 0.011)-the value thought to demarcate the boundaries between heavy and severe exercise intensities. Accordingly, using fixed DFAα1 values associated with the RCP from incremental exercise to guide constant-speed exercise training may produce a greater than expected exercise intensity, however; the dependency of DFAα1 on intensity and duration suggest its potential utility to quantify systemic perturbations imposed by continuous exercise.

From data to decision: Machine learning determination of aerobic and anaerobic thresholds in athletes

Lactate analysis plays an important role in sports science and training decisions for optimising performance, endurance, and overall success in sports. Two parameters are widely used for these goals: aerobic (AeT) and anaerobic (AnT) thresholds. However, determining AeT proves more challenging than AnT threshold due to both physiological intricacies and practical considerations. Thus, the aim of this study was to determine AeT and AnT thresholds using machine learning modelling (ML) and to compare ML-obtained results with the parameters' values determined using conventional methods. ML seems to be highly useful due to its ability to handle complex, personalised data, identify nonlinear relationships, and provide accurate predictions. The 183 results of CardioPulmonary Exercise Test (CPET) accompanied by lactate and heart ratio analyses from amateur athletes were enrolled to the study and ML models using the following algorithms: Random Forest, XGBoost (Extreme Gradient Boosting), and LightGBM (Light Gradient Boosting Machine) and metrics: R2, mean absolute error (MAE), mean squared error (MSE) and root mean square error (RMSE). The regressors used belong to the group of ensemble learning algorithms that combine the predictions of multiple base models to improve overall performance and counteract overfitting to training data. Based on evaluation metrics, the following models give the best predictions: for AeT: Random Forest has an R2 value of 0.645, MAE of 4.630, MSE of 44.450, RMSE of 6.667; and for AnT: LightGBM has an R2 of 0.803, the highest among the models, MAE of 3.439, the lowest among the models, MSE of 20.953, and RMSE of 4.577. Outlined research experiments, a comprehensive review of existing literature in the field, and obtained results suggest that ML models can be trained to make personalised predictions based on an individual athlete's unique physiological response to exercise. Athletes exhibit significant variation in their AeT and AT, and ML can capture these individual differences, allowing for tailored training recommendations and performance optimization.

Metabolic flexibility and resting autonomic function in active menopausal women

PURPOSE

The present study aims to analyze the relationship between cardiac autonomic control at rest-i.e., baseline Heart Rate Variability (HRV)-and metabolic flexibility assessed by means of the FATox and CHOox oxidation rates at the intensities of maximum fat and carbohydrate oxidation (MFO and MCO, respectively).

METHODS

Twenty-four active over-60 women (66.8 ± 4.4 years) had their HRV assessed with 10 min recordings under resting conditions, and this was analyzed with Kubios Scientific software. After this, an incremental submaximal cycling test, starting at 30 watts, with increments of 10 watts every 3 min 15 s was performed. FATox and CHOox were calculated in the last 60 s at each step, using Frayn's equation. MFO and MCO were further obtained.

RESULTS

Nonlinear SampEn and 1-DFAα1 (Detrending Fluctuation Analysis score) at rest were both moderate and significantly (p < 0.05) related to FATox (r = 0.43, r = -0.40) and CHOox (r = -0.59, r = 0.41), as well as RER (r = -0.43, r = 0.43) at FATmax intensity. At the MCO intensity, no association was observed between HRV and oxidation rates. However, DFAα1 (r = -0.63, p < 0.05), the frequency ratio LF/HF (r = -0.63, p < 0.05), and the Poincaré ratio SD1/SD2 (r = 0.48, p < 0.05) were correlated with blood lactate concentration.

CONCLUSION

These results support the autonomic resources hypothesis, suggesting that better autonomic function at rest is related to enhanced metabolic flexibility in postmenopausal women. They also underpin a comprehensive analysis of cardiovascular-autonomic health with aging. The results imply that non-linear DFAα1 and SampEn are appropriate to analyze this association in health of the aging cardiovascular-autonomic system.

Relationship of Cycling Power and Non-Linear Heart Rate Variability from Everyday Workout Data: Potential for Intensity Zone Estimation and Monitoring

The short-term scaling exponent alpha1 of detrended fluctuation analysis (DFA-a1) of heart rate variability (HRV) has been shown to be a sensitive marker for assessing global organismic demands. The wide dynamic range within the exercise intensity spectrum and the relationship to established physiologic threshold boundaries potentially allow in-field use and also open opportunities to provide real-time feedback. The present study expands the idea of using everyday workout data from the AI Endurance app to obtain the relationship between cycling power and DFA-a1. Collected data were imported between September 2021 and August 2023 with an initial pool of 3123 workouts across 21 male users. The aim of this analysis was to further apply a new method of implementing workout group data considering representative values of DFA-a1 segmentation compared to single workout data and including all data points to enhance the validity of the internal-to-external load relationship. The present data demonstrate a universal relationship between cycling power and DFA-a1 from everyday workout data that potentially allows accessible and regular tracking of intensity zone demarcation information. The analysis highlights the superior efficacy of the representative-based approach of included data in most cases. Validation data of the performance level and the up-to-date relationship are still pending.

Reliability and validity of a non-linear index of heart rate variability to determine intensity thresholds

Exercise intensity distribution is crucial for exercise individualization, prescription, and monitoring. As traditional methods to determine intensity thresholds present limitations, heart rate variability (HRV) using DFA a1 has been proposed as a biomarker for exercise intensity distribution. This index has been associated with ventilatory and lactate thresholds in previous literature. This study aims to assess DFA a1's reliability and validity in determining intensity thresholds during an incremental cycling test in untrained healthy adults. Sixteen volunteers (13 males and 3 females) performed two identical incremental cycling stage tests at least 1 week apart. First and second ventilatory thresholds, lactate thresholds, and HRV thresholds (DFA a1 values of 0.75 and 0.5 for HRVT1 and HRVT2, respectively) were determined in heart rate (HR), relative oxygen uptake (VO2rel), and power output (PO) values for both tests. We used intraclass correlation coefficient (ICC), change in mean, and typical error for the reliability analysis, and paired t-tests, correlation coefficients, ICC, and Bland-Altman analysis to assess the agreement between methods. Regarding reliability, HRV thresholds showed the best ICCs when measured in PO (HRVT1: ICC = .87; HRVT2: ICC = .97), comparable to ventilatory and lactate methods. HRVT1 showed the strongest agreement with LA 2.5 in PO (p = 0.09, r = .93, ICC = .93, bias = 9.9 ± 21.1), while HRVT2 reported it with VT2 in PO (p = 0.367, r = .92, ICC = .92, bias = 5.3 ± 21.9). DFA a1 method using 0.75 and 0.5 values is reliable and valid to determine HRV thresholds in this population, especially in PO values.

Exploring the Hidden Complexity: Entropy Analysis in Pulse Oximetry of Female Athletes

This study examines the relationship between physiological complexity, as measured by Approximate Entropy (ApEn) and Sample Entropy (SampEn), and fitness levels in female athletes. Our focus is on their association with maximal oxygen consumption (VO2,max). Our findings reveal a complex relationship between entropy metrics and fitness levels, indicating that higher fitness typically, though not invariably, correlates with greater entropy in physiological time series data; however, this is not consistent for all individuals. For Heart Rate (HR), entropy measures suggest stable patterns across fitness categories, while pulse oximetry (SpO2) data shows greater variability. For instance, the medium fitness group displayed an ApEn(HR) = 0.57±0.13 with a coefficient of variation (CV) of 22.17 and ApEn(SpO2) = 0.96±0.49 with a CV of 46.08%, compared to the excellent fitness group with ApEn(HR) = 0.60±0.09 with a CV of 15.19% and ApEn(SpO2) =0.85±0.42 with a CV of 49.46%, suggesting broader physiological responses among more fit individuals. The larger standard deviations and CVs for SpO2 entropy may indicate the body's proficient oxygen utilization at higher levels of physical demand. Our findings advocate for combining entropy metrics with wearable sensor technology for improved biomedical analysis and personalized healthcare.