hypothesis on heart rate and exercise

January 2, 2014

Sweaty Science: How Does Heart Rate Change with Exercise?

A physical pursuit from Science Buddies

By Science Buddies

Key concepts The heart Heart rate Health Exercise

Introduction Have you ever wondered how many times your heart beats in a day, a month, a year—or will beat in total throughout your life? Over an average lifetime, the human heart beats more than 2.5 billion times. For a person to keep their heart healthy, they should eat right, not smoke and get regular exercise. In this science activity, you'll measure your heart rate during different types of physical activities to find out which gives your heart the best workout to help keep it fit.

Background A 150-pound adult has about 5.5 liters of blood on average, which the heart circulates about three times every minute. A person's heart is continuously beating to keep the blood circulating. Heart health experts say that the best ways to keep our hearts healthy is through a balanced diet, avoiding smoking and regular exercise. 

On supporting science journalism

If you're enjoying this article, consider supporting our award-winning journalism by subscribing . By purchasing a subscription you are helping to ensure the future of impactful stories about the discoveries and ideas shaping our world today.

Exercise that is good for your heart should elevate your heart rate. But by how much, for how long and how often should your heart rate be elevated? This has to do with how fit you are and your maximum heart rate, which, for adults, is about 220 beats per minute (bpm) minus your age. For example, if you are 30 years old, your maximum heart rate would be 190 bpm. The American Heart Association (AHA) recommends doing exercise that increases a person's heart rate to between 50 to 85 percent of their maximum heart rate. This range is called the target heart rate zone. The AHA recommends a person gets at least 30 minutes of moderate to vigorous exercise—exercise that elevates their heart rate to the target heart rate zone—on most days of the week, or a total of about 150 minutes a week. 

Materials • Scrap piece of paper • Pen or pencil • Clock or timer that shows seconds or a helper with a watch • Comfortable exercise clothes (optional) • Simple and fun exercise equipment, such as a jump rope, bicycle, hula-hoop, two-pound weight, etc. Alternatively you can do exercises that do not require equipment, such as walking, doing jumping jacks, jogging in place, etc. You will want to do at least two different types of exercises, both of which you can sustain for 15 minutes. (Remember to always stop an exercise if you feel faint.)  • Calculator

Preparation • Practice finding your pulse. Use the first two fingers of one hand to feel your radial pulse on the opposite wrist. You should find your radial pulse on the "thumb side" of your wrist, just below the base of your hand. Practice finding your pulse until you can do it quickly. (You can alternatively take your carotid pulse to do this activity, but be sure you know how to safely take it and press on your neck only very lightly with your fingers.) • Measure your resting heart rate, which is your heart rate when you are awake but relaxed, such as when you have been lying still for several minutes. To do this, take your pulse when you have been resting and multiply the number of beats you count in 10 seconds by six. This will give you your resting heart rate in beats per minute (bpm). What is your resting heart rate? Write it on a scrap piece of paper. • You will be measuring your heart rate during different types of physical exercises over a period of 15 minutes. Choose at least two different exercises. Some examples include jumping rope, lifting a two-pound weight, riding a bike, hula-hooping, walking, etc. Gather any needed materials. (If you want to make a homemade hula-hoop, steps for doing this are given in the activity Swiveling Science: Applying Physics to Hula-Hooping .) Do you think the activities will affect your heart rate differently? How do you think doing each activity will affect your heart rate?

Procedure • Choose which exercise you want to do first. Before starting it, make sure you have been resting for a few minutes so that your heart is at its resting heart rate.  • Perform the first exercise for 15 minutes. While you do this, write down the number of beats you count in 10 seconds after one, two, five, 10 and 15 minutes of activity. (You want to quickly check your pulse because it can start to slow within 15 seconds of stopping exercising.) How do the number of beats you count change over time? How did you feel by the end of the exercise? • Calculate your heart rate after one, two, five, 10 and 15 minutes of exercise by multiplying the number of beats you counted (in 10 seconds) by six. How did your heart rate (in bpm) change over time? • Repeat this process for at least one other exercise. Leave enough time between the exercises so that your heart rate returns to around its normal resting level (this should only take a few minutes). How did you feel by the end of the second exercise? How did your heart rate change over time for this exercise?  • Take a look at the results you wrote down for this activity. Which exercise increased your heart rate the most? Which exercise increased your heart rate the fastest? Which exercise(s) elevated your heart rate to the target heart rate zone (50 to 85 percent of your maximum heart rate, where your maximum heart rate is 220 bpm minus your age)? Do you notice any consistent patterns in your results? • Extra: Try this activity again but test different physical exercises. How does your heart rate change when you do other exercises? How are the changes similar and how are they different? • Extra: Measure your heart rate while lying down, while sitting down, and while standing. How does your heart rate change with body position?  • Extra: Repeat this activity with other healthy volunteers. How does their heart rate compare to yours? How does their change in heart rate while exercising compare to how yours changed?  • Extra: Try this activity again but vary the intensity of your exercise. What intensity level elevates your heart rate to 50 percent of its maximum heart rate? What about nearly 85 percent of its maximum? Be sure not to exceed your recommended target heart rate zone while exercising! Observations and results After just a minute of exercise, did you see your heart rate reach its target heart rate zone? Did it initially jump higher for a more strenuous exercise, like hula-hooping, compared to a more moderately intense exercise, such as walking?

If you did a moderately intense exercise, such as walking, you may have seen an initial jump in your heart rate (where your heart rate falls within the lower end of your target heart rate zone within about one minute of exercise), but then your heart rate only slowly increased after that. After 15 minutes, you may have reached the middle of your target heart rate zone. To reach the upper end, people usually need to do a moderately intense exercise for a longer amount of time (such as for 30 minutes). If you did a more strenuous exercise—hula-hooping, for example—you may have seen a higher initial bump in your heart rate (such as reaching the middle of your target heart rate zone after just one minute of exercise), and then your heart rate stayed about the same for the remaining 14 minutes of exercise. Overall doing a more strenuous exercise generally raises a person's heart rate faster compared to doing an exercise that is only moderately intense.

More to explore Target Heart Rates , from the American Heart Association Cut to the Heart , from NOVA and PBS Life's Simple 7—Get Active , from the American Heart Association Heart Health: How Does Heart Rate Change with Exercise? , from Science Buddies

This activity brought to you in partnership with  Science Buddies

Thank you for visiting nature.com. You are using a browser version with limited support for CSS. To obtain the best experience, we recommend you use a more up to date browser (or turn off compatibility mode in Internet Explorer). In the meantime, to ensure continued support, we are displaying the site without styles and JavaScript.

• View all journals
• Explore content
• About the journal
• Publish with us
• Sign up for alerts
• Open access
• Published: 15 November 2023

Modeling personalized heart rate response to exercise and environmental factors with wearables data

• Achille Nazaret 1 ,
• Sana Tonekaboni 2 ,
• Gregory Darnell 3 ,
• Shirley You Ren 3 ,
• Guillermo Sapiro 3 &
• Andrew C. Miller   ORCID: orcid.org/0000-0003-0818-3705 3  

npj Digital Medicine volume  6 , Article number:  207 ( 2023 ) Cite this article

4320 Accesses

1 Citations

3 Altmetric

Metrics details

• Cardiovascular biology
• Data integration
• Machine learning

Heart rate (HR) response to workout intensity reflects fitness and cardiorespiratory health. Physiological models have been developed to describe such heart rate dynamics and characterize cardiorespiratory fitness. However, these models have been limited to small studies in controlled lab environments and are challenging to apply to noisy—but ubiquitous—data from wearables. We propose a hybrid approach that combines a physiological model with flexible neural network components to learn a personalized, multidimensional representation of fitness. The physiological model describes the evolution of heart rate during exercise using ordinary differential equations (ODEs). ODE parameters are dynamically derived via a neural network connecting personalized representations to external environmental factors, from area topography to weather and instantaneous workout intensity. Our approach efficiently fits the hybrid model to a large set of 270,707 workouts collected from wearables of 7465 users from the Apple Heart and Movement Study. The resulting model produces fitness representations that accurately predict full HR response to exercise intensity in future workouts, with a per-workout median error of 6.1 BPM [4.4–8.8 IQR]. We further demonstrate that the learned representations correlate with traditional metrics of cardiorespiratory fitness, such as VO 2 max (explained variance 0.81 ± 0.003). Lastly, we illustrate how our model is naturally interpretable and explicitly describes the effects of environmental factors such as temperature and humidity on heart rate, e.g., high temperatures can increase heart rate by 10%. Combining physiological ODEs with flexible neural networks can yield interpretable, robust, and expressive models for health applications.

Similar content being viewed by others

Longitudinal cardio-respiratory fitness prediction through wearables in free-living environments

Estimation of cardiorespiratory fitness using heart rate and step count data

Sensing leg movement enhances wearable monitoring of energy expenditure

Introduction.

The increasing availability of wearable technologies has empowered individuals to monitor their overall health and well-being throughout daily life—a recent research study found that 21% of Americans use wearable fitness trackers or smartwatches 1 . Such wearable data carry the potential for machine learning (ML) models to discover new correlates of human health from device signals 2 . Previous work has shown success in various applications, from clinical monitoring tools to fitness and activity planners 3 . In the clinical domain, modern ML methods have been shown to predict cardiovascular events from wearables data 4 , 5 , 6 , as well as overall health conditions indicated by different lab measures 7 . In the public health domain, wearables data and ML algorithms have been successfully deployed in disease surveillance 8 such as detecting influenza-like illness 9 , monitoring disease progression in the population 10 , 11 , and informing the design of new epidemiological studies 12 . This relatively recent area of research demonstrates that wearables data is a rich source of information that can provide insight into an individual’s overall health.

Heart rate response to activity or exercise reflects an individual’s cardiorespiratory fitness. In sports medicine, exercise stress testing is used to quantify cardiac health by measuring cardiac response to a controlled physical activity 13 . Physiological models of heart rate response to exercise intensity have been developed to measure cardiorespiratory in controlled settings in small studies 14 , 15 , 16 , 17 . Similar principles may be used to monitor cardiorespiratory health throughout daily activities using wearables data, as opposed to controlled-environment tests. However, there are two major challenges. First, these physiological models of heart rate response must be adapted to use data collected by wearables—e.g., step count, speed, elevation change, and weather. Second, personalizing these models to measure an individual’s cardiorespiratory fitness is both computationally and statistically challenging. Given the prevalence of wearables, such health monitoring could reach a broader population with an increased frequency, and without interruption to daily life.

In this work, we develop a scalable algorithm that predicts heart rate response to workout intensity as measured by wearables—step count, speed, and elevation change. We augment an expert model from the exercise physiology literature with machine learning components and inference techniques 14 . Our approach combines a physiological model of heart rate (HR) based on ordinary differential equations (ODEs) with neural networks and representation learning to estimate personalized, user-specific parameters. Our algorithm learns to map a subject’s recent workout history to a personalized representation that is predictive of HR response in future workouts under the ODE model. This is done with data from the Apple Heart and Movement Study (AHMS) 18 , rendering this work as one of the largest and in regular environments (outside of the lab) reported in the literature on this topic.

We show that our personalized representations and model can accurately estimate the heart rate profile given workout data sequences by simply using an individual’s workout history. Unlike most existing work in the literature that performs short-term HR prediction 19 , 20 , our method can predict the entire HR trend of a completely new workout of up to 2 h. These learned health representations of individuals could be used in a variety of applications, such as personalized workout planning and estimating HR zones or calories burned during a workout.

The ability to accurately estimate HR response to any workout shows that the representations summarize meaningful information about an individual’s health. We investigate this by showing that, after being learned only for predicting heart rate, the learned representations correlate well with traditional metrics of cardiorespiratory fitness, e.g., VO 2 max. Additionally, we explicitly show evidence of an effect of weather on heart rate during workouts across the study population; this is done by augmenting the biophysical ODE mode to consider environmental factors.

Study design and participants

Our study uses workout measurements contributed to the Apple Heart and Movement Study (AHMS) between November 2019 and July 2022 18 . The Apple Heart and Movement Study is a prospective, single-group, open-label, siteless, pragmatic observational study conducted in collaboration with the American Heart Association and Brigham and Women’s Hospital to investigate the relationship between physical activity, mobility, and heart health. The study was approved by the Advarra Central Institutional Review Board (PRO00036784) and registered to ClinicalTrials.gov (ClinicalTrials.gov Identifier: NCT04198194). There is no compensation for participation.

The AHMS allows participants to securely share information collected by their Apple Watch. To be eligible, the participants must own an Apple Watch paired with their iPhone, be at least 18 years old (at least 19 years old in Alabama and Nebraska; at least 21 years old in Puerto Rico), live in the United States of America, have installed the Apple Research app on their iPhone, do not share their iCloud account, iPhone, or Apple Watch with anyone else, and are willing and able to provide informed consent to participate in the study. The study app was used to verify eligibility, obtain participants’ consent, provide study education, and direct participants through the study procedures.

Within the AHMS, we selected participants who chose to use the Workout app on their Apple Watch to record their outdoor runs. The AHMS logs the outdoor weather information W (temperature and humidity) at the time of each workout and provides the participant’s heart rate during each workout along with four measurements of exercise intensity: the instantaneous speed from the pedometer sensor, the instantaneous speed from the GPS, the step cadence from the pedometer, and the elevation gain from the altimeter. The sensor measurements are interpolated on a 10-s grid to form, for each workout w , a heart rate time-series \(\widehat{{{{\rm{HR}}}}}\in {{\mathbb{R}}}^{d}\) and a multivariate time-series of exercise intensity \(I\in {{\mathbb{R}}}^{4\times d}\) ; where d is the duration of the workout. We finally only selected workouts with a duration between 15 and 120 min long that had no missing data and further filtered out participants with less than 10 valid runs. We obtained 270,707 outdoor runs from 7465 distinct subjects (see Fig. 1 for cohort summary and inclusion criteria).

a Subject and workout inclusion diagram. b Description of study data and summary statistics of included participants. Intervals reflect the 2.5 and 97.5 quantile range.

Our hybrid approach blends a physiological model of heart rate dynamics with machine learning components (e.g., deep neural networks) to adapt it to wearable data and personalize it to individual subjects. We first detail the physiological model and then describe our ML-based augmentations and learning algorithm. A diagram of the end-to-end system is detailed in Fig. 2 .

Overview of the method for modeling heart rate response to exercise with wearable data. The top panel describes the study population data and details of the wearable workout measurements used. The two bottom panels describe the training procedure and how the model components are trained to learn the personalized representation and ODE parameters using only heart rate and workout data. Learned representations can then be used downstream to predict various physiological traits, e.g., Fitness, BMI, and age.

Physiological HR dynamics model

Several works in the sports physiology literature have studied heart rate dynamics in response to exercise using ordinary differential equations (ODEs) 14 , 15 , 16 , 17 . These approaches translate the physical mechanisms of the human body into differential equations in order to incorporate domain (physiology) knowledge in the modeling. This is an appealing method to build interpretable and ultimately trustworthy models of fitness and health.

A common approach for modeling changes in heart rate, HR, due to exercise intensity, t   ↦   I ( t ), is to introduce the oxygen demand, D , as an intermediary quantity through coupled ODEs 15

In this dynamical system, the function f translates the instantaneous activity intensity I into the necessary oxygen demand for I . The top equation attempts to match the current body oxygen demand D with the instantaneous demand f ( I ) (also known as the drive function ). Parameter B controls how fast D adapts to f ( I ). At the same time, the second equation drives the heart rate HR toward the pace required to deliver the demand D . Parameter A controls how fast the heart can adapt while the terms with HR min , HR max , and α ,  β control how difficult it is to reach the maximal heart rate or to rest down to the minimal heart rate.

To learn the parameters of this ODE, previous studies of such models have limited their data collection to controlled laboratory environments with small samples—typically fewer than ten individuals 14 , 15 , 16 , 17 . Workout intensity data I was typically the power output or cadence from a stationary bicycle, and the drive function f ( I ) was modeled using low-order polynomials 14 , 15 . Here, we extend the ODE method to large-scale uncontrolled environments and use it to model workout data from wearable devices; we do this by learning some parts of the ODE as neural networks and some parameters of the ODE as user-specific variables. Furthermore, we augment the ODE to incorporate environmental factors, such as temperature and fatigue.

Modeling heart rate in uncontrolled environments

In uncontrolled environments, accurately measuring the exercise intensity I becomes challenging. Instead, wearables rely on sensors like GPS and pedometers to measure proxy variables for activity intensity (such as speed, elevation, and number of steps). However, the relationship between I and f ( I ) becomes unclear when dealing with such generally noisy measurements. To tackle this challenge, we use a flexible neural network to model and learn the drive function f .

Uncontrolled environments might also have external factors that can affect heart rate. For instance, higher temperatures induce a higher oxygen demand 21 . In contrast to controlled environments, workouts in uncontrolled settings may also vary in duration, leading to potential changes in heart rate dynamics over time due to fatigue. Hence, we refine the heart rate equations by modeling how they are affected by the weather W , which includes temperature and humidity measurements at the time of workout, and add the effect of fatigue incurred over time t during the workout. We parameterize these effects by neural networks g ( W ) and h ( t ), respectively, and we incorporate them into the demand equation. The term ( f ( I ( t )) −  D ( t )) becomes ( f ( I ( t ))  ⋅   g(W) ⋅ h(t)  −  D ( t )). For instance, g ( W ) > 1 indicates an increase in oxygen demand for the weather W .

Learning a personalized large-scale heart rate model

Each individual possesses their own set of personalized parameters (A, B, α , β , as well as the drive function f and HR min , HR max ) that capture their unique heart rate dynamics in response to exercise. Inferring these parameters for each subject and understanding how they might evolve over time can reveal important insight into their health status 17 . However, learning these parameters for every subject and each new workout is computationally expensive. Instead of directly learning a set of parameters for each subject, we assume that an individual’s health state at a given time can be represented by a low-dimensional latent vector \(z\in {{\mathbb{R}}}^{\ell }\) . Then, we turn each ODE parameter into a function of this health representation. For instance, the parameter α becomes α ( z ), and f ( I ) becomes f ( z ,  I ). All these “functions of z ” are parametrized as neural networks, and our goal will be to learn these health representations.

With these changes, the ODE in Equation (1) becomes

Fitting our proposed ODE model to large-scale wearable data involves learning both the global shared neural networks and the subject-time specific health representation z . In order to efficiently learn the health representation z , we finally introduce one last component inspired by deep learning methodologies. We posit that at any time, the subject’s workout history up to this time contains all of the information to characterize z . To model the complex interactions that define z based on this history, we utilize a convolutional neural network (CNN) encoder architecture. This allows us to learn the subject-specific health representation as a function of their past workout data. Hence, we exchangeably refer to z as the health representation or the history embedding . Figure 2 summarizes the full pipeline of our method: at any given time and for any subject, the model takes the workout history of this subject up to this time and feeds it to an encoder function to obtain a health representation z . Subsequently, the representation z is transformed into ODE parameters that are used to solve the ODE for new incoming workouts. Training the model end-to-end is done with standard gradient descent to identify the best neural network weights that best predict workout heart rate sequences.

For training and evaluation, we divided the data into a training set and a testing set. The training set comprises the first 80% of workouts for each subject, while the remaining 20% of workouts form the test set and are held out during training. We selected a few hyperparameters using the best training loss. Additional model details and a description of the implementation, neural network hyperparameters, and training procedure using ODE solvers can be found in Supplementary Methods .

Heart rate profile forecasting

The representation z estimated using an individual’s workout history can be used to predict the heart rate in future workouts. We measured the accuracy of heart rate prediction on workouts that were held out for each subject. Figure 3 shows two examples comparing the true heart rate to the heart rate estimated using our model (additional predictions in Supplementary Fig. 7) . Note that for predicting HR for workout w happening at date T , our model only uses the workout intensity measures of that sample I and the personalized health representation z —coming from encoding the previous workouts; i.e., the model does not observe any HR measurements for making predictions.

a Heart rate reconstruction performance. We compare the average workout mean absolute error (MAE) (and 95 % CI), as well as the median workout MAE and interquartile range (IQR). Additionally, we compare the mean absolute percent error (MAPE) (and IQR) and the prediction sequence correlation (and IQR). We observe the history embedding z used in the hybrid ODE model and seq-to-seq baseline improves predictions, and the hybrid ODE model consistently outperforms the strong seq-to-seq baseline. b Example HR predictions for two separate workouts. The x -axis indicates time since the beginning of the workout and the y -axis shows the subject’s instantaneous heart rate (beats per minute). The measured heart rate sequence is in gray, and the predicted sequence is in red. Uncertainty bands about the observation reflect a standard deviation of ± 5 beats per minute in the heart rate measurement. Additional predictions in Supplemental Fig. 7 .

We compare the prediction performance of our model to seq-to-seq deep models with and without the embedding z with similar modeling capacity, as well as a simple heuristic that uses the per-subject mean HR to form the prediction. Figure 3 reports the performance across these baselines, with our model outperforming the context-free (i.e., no z ) and the strong seq-to-seq baseline. We also measure the performance of our model in estimating the HR after the first 2 min of the workout. Indeed, it is difficult, if not impossible, for a model to predict the heart rate at the beginning of a workout. The initial heart rate depends on the user’s activity prior to the workout, which is unobserved and unpredictable. Conversely, we hypothesize that the heart rate after 2 min can be explained by the user’s activity in the first 2 min that we do observe. Again, we see that the hybrid ODE model outperforms all baselines.

Oxygen demand inference

One of the key benefits of our hybrid model lies in the interpretability of its latent variables. Supplementary Fig. 8 depicts the inferred demand curve D for a set of random workouts. We observe that the demand is highly correlated with HR, but typically at a lag—changes in HR tend to follow changes in demand, and the speed of those changes is described by the ODE parameters.

Calories burned estimation

The number of calories burned during exercise can be approximated using heart rate measurements during the workout with a linear formula 22 . This, which is only a first-order approximation often augmented with other movement measurements, is useful for planning workouts based on calorie burn goals and even more useful in cases where individuals are not wearing a wearable device that records heart rate. Our method can reliably estimate the number of calories burned with a 5% relative error (the same relative error as the heart rate reconstruction), only using workout metrics that can be measured using a smartphone.

Heart rate zone prediction

Exercise heart rate zones are the percentage of an individual’s age-related maximum heart rate reached throughout the course of exercise, where maximum heart rate is derived using the common 220 bpm − age heuristic 23 (distinct from the ODE model parameter). Using our physiological model, we can predict heart rate zones using the workout data. This can help individuals plan personalized exercise routines to achieve their fitness goals more effectively. We define six zones (% intervals [0, 50, 60, 70, 80, 90, 100]) of maximum heart rate, and Fig. 4 a shows the performance of our method in predicting the HR zone for the whole population, as well as different subgroups of the population. We find that the model can predict heart rate zones with an accuracy of around 67%. To provide a comparison, we computed the marginal distribution of true heart rate zones across all workouts and found that the most frequent zone ([80,90]) occurs about 38% of the time. So the best possible constant predictor would be correct about 38% of the time—significantly lower than the accuracy achieved by our model.

a Predictive performance of heart rate zones (mean and standard deviation). The most frequent zone occurred 38% of the time on average, which corresponds to the accuracy the best possible baseline predictor can achieve. b VO 2 max prediction performance in mean squared error (MSE) and mean absolute error (MAE) (mean and standard error).

Quantifying the impact of the weather on heart rate

Leveraging the interpretability of our ODE model, we analyze the learned neural network g and quantify the relative effect of weather on the body’s oxygen demand. This constitutes one of the largest studies of this kind (over 270,000 workouts). The neural network g is a global function shared between all subjects and workouts. Figure 5 shows an increase in body oxygen demand by up to 10% in high temperatures and humidity. Moreover, we found that personalizing g as W   ↦   g ( z ,  W ) in the same way that the personalized exercise intensity function f is parameterized by the representation z did not result in significant improvements in heart rate predictions. We kept a shared g for simplicity and interpretability.

Impact of weather temperature and humidity on the body’s oxygen demand g ( W ), as modeled by the differential equation. As temperature and humidity rise, we see an increase in oxygen demand, amounting to a 3-6% increase near 100 degrees (or about 4.5–9 BPM at 150 BPM effort).

Learning about cardiorespiratory health

To check that our representations summarize information about cardiorespiratory health, we use a summary of cardio fitness, VO 2 max, estimated by wearable devices. VO 2 max is the maximum rate of oxygen the body can consume during exercise, normalized by body mass. While a cardiopulmonary exercise test (CPET) is the gold standard for measuring VO 2 max, such tests can be prohibitively expensive and even infeasible for certain populations. Instead, we can approximate VO 2 max from sub-maximal exercise bouts using measurements collected in our study 24 , including heart rate, GPS, and user informaiton 25 .

Using the health representations z , we predict the estimated VO 2 max with a simple linear regression model and achieve a mean absolute error of ± 2.16 mL/(kg  ⋅  min), which is about 5% of the average VO 2 max in the data (42.5 mL/(kg  ⋅  min)). Figure 4 b reports the performance of a linear regression model on the ODE representations only, on demographics only, or on both. The demographics include subject height, weight, biological sex, and age. Supplementary Fig. 11 shows a 2D projection of the health representation for different workouts where we can see the separation of higher and lower values of VO 2 max. We also examined the association between the z representations and subject age and body mass index (BMI). With a linear regression model, we find that the z representations explain 33% ( ± 0.3 %) of the variance in age and 16% ( ± 0.7%) of the variance of BMI in our cohort.

Interpreting hybrid ODE model inferences

We also investigate the relationship between the inferred ODE parameters ( A , B , α , and β ) and age, sex, and fitness. Supplementary Fig. 9 illustrates the variation of these four parameters as a function of age (on the x -axis), stratified by VO 2 max tertiles. As previously mentioned, a higher value of α suggests that individuals can more easily approach their resting heart rate. We observe that the highest VO 2 max tertile exhibits a significantly higher inferred α , although the gap disappears as the cohort ages. Similarly, a higher value of β indicates that individuals can more readily reach their maximum heart rate. Among the youngest cohort, we observe that the fittest group can reach the inferred maximum heart rate more quickly. The parameter A characterizes the overall sensitivity of heart rate changes. In the younger age range, the least fit (lowest VO 2 max cohort) displays a significantly higher average A value, which diminishes as age increases. Lastly, parameter B signifies the sensitivity of the demand sequence D to changes in exercise intensity. We note that this value exhibits less variability across ages and VO 2 max strata.

The increasing availability of wearable devices is enabling individuals to track their health and fitness. We developed a method that predicts heart rate response to workout intensity using data from a wearable device. We learn representations that summarize the dynamics of the HR response, by combining machine learning techniques with an expert model from the exercise physiology literature. All results are derived from one of the largest studies in the general population (outside of the lab), illustrating the power of wearables in real, everyday scenarios.

We show that this hybrid model can accurately predict heart rate sequences for new workouts given a user-specific history of recent workouts. Beyond heart rate predictions, we show that representations from this algorithm can serve as a measure of cardiorespiratory fitness, which can help track fitness levels over time and aid personalized workout planning. Additionally, we show evidence of the effect of weather on heart rate demand across the study population.

Methodologically, we demonstrate how machine learning techniques can be used to translate an expert model—originally developed for a controlled setting—to noisy, real-world signals collected by wearables throughout a subject’s daily life. We use ML techniques to both augment the model (e.g., using a neural network to model the intensity-demand function) and to scale the algorithm to a large dataset of users and workouts (e.g., via learned subject-specific history embeddings).

There remain multiple avenues for future work. The first is further developing applications of this model for planning new workouts and tracking fitness changes over time. While predictions of heart rate response describe user fitness, it remains an open question how to translate insights from such a model to improvements in fitness over time. With respect to behavior change, the studied effects of incentivizing exercise adherence remain unclear 26 , 27 , and the extent to which personalized insights may help is unknown.

Additionally, we aim to better understand how these learned representations are associated with or predict changes in cardiovascular function and adverse cardiovascular outcomes. Cardiovascular health is strongly associated with exercise and fitness level 28 , and physical activity can be a preventative activity and prognostic indicator for heart failure 29 . And while we focus on running workouts, lower-intensity walking workouts may also carry rich information about an individual’s cardiovascular health—adaptations of our approach could be applied to walking workouts (or both walking and running workouts).

Lastly, this work proposes a new methodology to combine expert models of physiology alongside machine learning components. While we observe that this expert model can provide an inductive bias that makes predictions more accurate, it remains to be studied how accurately physiologically interpretable parameters can be identified and inferred from wearables data. Furthermore, more complex models of heart rate—e.g., directly parameterizing VO 2 max or exercise thresholds—remain to be studied within this hybrid framework.

A limitation of the study design is that participants must have access to an iPhone with the Research app and an Apple Watch to be eligible for participation. Furthermore, we select individuals who have running workouts to train and evaluate our models. Further study is required to understand if our approach is accurate under a less active population—such selection bias may confound our interpretation of inferred physiological parameters. Another limitation is that this study focuses only on outdoor running workouts. In theory, this approach could be extended to other activities, such as cycling or indoor runs, provided the analogous workout intensity is sufficiently rich. Additionally, our study is limited to questions of cardiorespiratory fitness, and not subsequent health outcomes. Follow-up on more detailed longitudinal data is necessary to establish a strong link between representations learned by our algorithm and subsequent health-related events.

Data availability

Due to privacy and consent considerations, data from the AHMS cannot be shared. We do provide a detailed algorithmic implementation of our approach, and readers interested in the tools can contact the authors for further support.

Code availability

The code used to run our model will be hosted at https://github.com/apple/ml-heart-rate-models upon acceptance. This repository includes methods for preprocessing data, fitting models, and making predictions.

Vogels, E. A. About One-in-five Americans Use a Smart Watch or Fitness Tracker . https://www.pewresearch.org/fact-tank/2020/01/09/about-one-in-five-americans-use-a-smart-watch-or-fitness-tracker/ . (Pew Research Center, Washington, D.C., 2020).

Ferguson, T. et al. Effectiveness of wearable activity trackers to increase physical activity and improve health: a systematic review of systematic reviews and meta-analyses. Lancet Digit. Health 4 , e615–e626 (2022).

Article   CAS   PubMed   Google Scholar  

Tang, M. S. S., Moore, K., McGavigan, A., Clark, R. A. & Ganesan, A. N. Effectiveness of wearable trackers on physical activity in healthy adults: systematic review and meta-analysis of randomized controlled trials. JMIR mHealth uHealth 8 , e15576 (2020).

Article   PubMed   PubMed Central   Google Scholar  

Ballinger, B. et al. Deepheart: semi-supervised sequence learning for cardiovascular risk prediction. In: AAAI Conference on Artificial Intelligence (AAAI Press, Palo Alto, California USA, 2018).

Nemati, S. et al. Monitoring and detecting atrial fibrillation using wearable technology. In: 2016 38th Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC) , 3394–3397 (IEEE, 2016).

Suzuki, T., Kameyama, K.-i. & Tamura, T. Development of the irregular pulse detection method in daily life using wearable photoplethysmographic sensor. In: 2009 Annual International Conference of the IEEE Engineering in Medicine and Biology Society , 6080–6083 (IEEE, 2009).

Dunn, J. et al. Wearable sensors enable personalized predictions of clinical laboratory measurements. Nat. Med. 27 , 1105–1112 (2021).

Ates, H. C., Yetisen, A. K., Güder, F. & Dincer, C. Wearable devices for the detection of covid-19. Nat. Electron. 4 , 13–14 (2021).

Article   CAS   Google Scholar  

Radin, J. M., Wineinger, N. E., Topol, E. J. & Steinhubl, S. R. Harnessing wearable device data to improve state-level real-time surveillance of influenza-like illness in the usa: a population-based study. Lancet Digit. Health 2 , e85–e93 (2020).

Avram, R. et al. Predicting diabetes from photoplethysmography using deep learning. J. Am. Coll. Cardiol. 73 , 16–16 (2019).

Article   Google Scholar  

Mitratza, M. et al. The performance of wearable sensors in the detection of SARS-CoV-2 infection: a systematic review. Lancet Digit. Health 4 , e370–e383 (2022).

Article   CAS   PubMed   PubMed Central   Google Scholar  

Golbus, J. R., Pescatore, N. A., Nallamothu, B. K., Shah, N. & Kheterpal, S. Wearable device signals and home blood pressure data across age, sex, race, ethnicity, and clinical phenotypes in the michigan predictive activity & clinical trajectories in health (mipact) study: a prospective, community-based observational study. Lancet Digit. Health 3 , e707–e715 (2021).

Lear, S. A., Brozic, A., Myers, J. N. & Ignaszewski, A. Exercise stress testing. Sports Med. 27 , 285–312 (1999).

Mazzoleni, M. J. et al. A dynamical systems approach for the submaximal prediction of maximum heart rate and maximal oxygen uptake. Sports Eng. 21 , 31–41 (2018).

Mazzoleni, M. J., Battaglini, C. L., Martin, K. J., Coffman, E. M. & Mann, B. P. Modeling and predicting heart rate dynamics across a broad range of transient exercise intensities during cycling. Sports Eng. 19 , 117–127 (2016).

Stirling, J. R., Zakynthinaki, M., Refoyo, I. & Sampedro, J. A model of heart rate kinetics in response to exercise. J Nonlinear Math. Phys. 15 , 426–436 (2008).

Zakynthinaki, M. S. Modelling heart rate kinetics. PLoS ONE 10 , e0118263 (2015).

Apple. Apple Heart & Movement Study. https://clinicaltrials.gov/ct2/show/NCT04198194 ClinicalTrials.gov Identifier: NCT04198194 (Apple, 2019).

Ludwig, M., Hoffmann, K., Endler, S., Asteroth, A. & Wiemeyer, J. Measurement, prediction, and control of individual heart rate responses to exercise-basics and options for wearable devices. Front. Physiol. 9 , 778 (2018).

Ni, J., Muhlstein, L. & McAuley, J. Modeling heart rate and activity data for personalized fitness recommendation. In: The World Wide Web Conference . Association for Computing Machinery. New York, NY, United States.1343–1353 (2019).

Galloway, S. & Maughan, R. J. Effects of ambient temperature on the capacity to perform prolonged cycle exercise in man. Med. Sci. Sports Exerc. 29 , 1240–1249 (1997).

Keytel, L. et al. Prediction of energy expenditure from heart rate monitoring during submaximal exercise. J. Sports Sci. 23 , 289–297 (2005).

Center for Disease Control and Prevention. Target heart rate and estimated maximum heart rate. https://www.cdc.gov/physicalactivity/basics/measuring/heartrate.htm (Center for Disease Control and Prevention, 2022).

Bennett, H., Parfitt, G., Davison, K. & Eston, R. Validity of submaximal step tests to estimate maximal oxygen uptake in healthy adults. Sports Med. 46 , 737–750 (2016).

Article   PubMed   Google Scholar  

Apple. Using apple watch to estimate cardio fitness with Vo 2 max. https://www.apple.com/healthcare/docs/site/Using_Apple_Watch_to_Estimate_Cardio_Fitness_with_VO 2 _max.pdf (Apple, 2021).

Mitchell, M. S. et al. Financial incentives for exercise adherence in adults: systematic review and meta-analysis. Am. J. Prev. Med. 45 , 658–667 (2013).

Strohacker, K., Galarraga, O. & Williams, D. M. The impact of incentives on exercise behavior: a systematic review of randomized controlled trials. Ann. Behav. Med. 48 , 92–99 (2014).

Nystoriak, M. A. & Bhatnagar, A. Cardiovascular effects and benefits of exercise. Front. Cardiovasc. Med. 5 , 135 (2018).

Cattadori, G., Segurini, C., Picozzi, A., Padeletti, L. & Anzà, C. Exercise and heart failure: an update. ESC Heart Failure 5 , 222–232 (2018).

Download references

Acknowledgements

We thank the participants of the AHMS for their generosity in donating data to advance science toward health and well-being improvements. AN was supported in part by funding from the Eric and Wendy Schmidt Center at the Broad Institute of MIT and Harvard. We also thank Calum MacRae and Joe Futoma for their discussion and feedback.

Author information

Authors and affiliations.

Columbia University, New York, NY, USA

Achille Nazaret

University of Toronto, Toronto, ON, Canada

Sana Tonekaboni

Apple, Cupertino, CA, USA

Gregory Darnell, Shirley You Ren, Guillermo Sapiro & Andrew C. Miller

You can also search for this author in PubMed   Google Scholar

Contributions

A.N. and S.T. contributed to project conceptualization, data curation and analysis, methodological development, software implementation, validation, visualization, and manuscript writing and editing. A.C.M. contributed to project conceptualization, data curation, methodological development, project administration, writing and editing the manuscript, and general project supervision. G.D., S.R., and G.S. contributed to methodological development and analysis, provided supervision, and contributed to the writing and editing of the manuscript.

Corresponding author

Correspondence to Andrew C. Miller .

Ethics declarations

Competing interests.

All authors were employed by Apple during this research. Apple supported travel for A.N., S.T., and A.C.M. A.C.M., G.D., S.R., and G.S. own Apple stock. Study design and data collected in this study focus on signals generated by Apple devices (e.g., iPhone and Watch), as well as surveys. The data Apple Heart and Movement Study is sponsored by Apple. A patent application on related technology has been filed. Beyond employment, Apple was not involved in conducting this research, writing this report, or the decision to submit this work for publication.

Additional information

Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Supplementary information

Supplemental material, rights and permissions.

Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons license, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons license and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this license, visit http://creativecommons.org/licenses/by/4.0/ .

Reprints and permissions

About this article

Cite this article.

Nazaret, A., Tonekaboni, S., Darnell, G. et al. Modeling personalized heart rate response to exercise and environmental factors with wearables data. npj Digit. Med. 6 , 207 (2023). https://doi.org/10.1038/s41746-023-00926-4

Download citation

Received : 31 March 2023

Accepted : 20 September 2023

Published : 15 November 2023

DOI : https://doi.org/10.1038/s41746-023-00926-4

Share this article

Anyone you share the following link with will be able to read this content:

Sorry, a shareable link is not currently available for this article.

Provided by the Springer Nature SharedIt content-sharing initiative

Quick links

• Explore articles by subject
• Guide to authors
• Editorial policies

Sign up for the Nature Briefing: Translational Research newsletter — top stories in biotechnology, drug discovery and pharma.

Advertisement

Genetics and the heart rate response to exercise

• Open access
• Published: 27 March 2019
• Volume 76 , pages 2391–2409, ( 2019 )

Cite this article

You have full access to this open access article

• Yordi J. van de Vegte 1   na1 ,
• Balewgizie S. Tegegne 3   na1 ,
• Niek Verweij 1 ,
• Harold Snieder 3 &
• Pim van der Harst 1 , 2 , 4  

13k Accesses

29 Citations

7 Altmetric

Explore all metrics

The acute heart rate response to exercise, i.e., heart rate increase during and heart rate recovery after exercise, has often been associated with all-cause and cardiovascular mortality. The long-term response of heart rate to exercise results in favourable changes in chronotropic function, including decreased resting and submaximal heart rate as well as increased heart rate recovery. Both the acute and long-term heart rate response to exercise have been shown to be heritable. Advances in genetic analysis enable researchers to investigate this hereditary component to gain insights in possible molecular mechanisms underlying interindividual differences in the heart rate response to exercise. In this review, we comprehensively searched candidate gene, linkage, and genome-wide association studies that investigated the heart rate response to exercise. A total of ten genes were associated with the acute heart rate response to exercise in candidate gene studies. Only one gene ( CHRM2 ), related to heart rate recovery, was replicated in recent genome-wide association studies (GWASs). Additional 17 candidate causal genes were identified for heart rate increase and 26 for heart rate recovery in these GWASs. Nine of these genes were associated with both acute increase and recovery of the heart rate during exercise. These genes can be broadly categorized into four categories: (1) development of the nervous system ( CCDC141 , PAX2 , SOX5, and CAV2 ); (2) prolongation of neuronal life span ( SYT10 ); (3) cardiac development ( RNF220 and MCTP2 ); (4) cardiac rhythm ( SCN10A and RGS6 ). Additional 10 genes were linked to long-term modification of the heart rate response to exercise, nine with heart rate increase and one with heart rate recovery. Follow-up will be essential to get functional insights in how candidate causal genes affect the heart rate response to exercise. Future work will be required to translate these findings to preventive and therapeutic applications.

Similar content being viewed by others

Thirty loci identified for heart rate response to exercise and recovery implicate autonomic nervous system

Genetic study links components of the autonomous nervous system to heart-rate profile during exercise

Inter-individual variation in adaptations to endurance and resistance exercise training: genetic approaches towards understanding a complex phenotype

Avoid common mistakes on your manuscript.

Introduction

The regulation of resting heart rate is complex; autonomic tone, central and peripheral reflexes, hormonal influences, and factors intrinsic to the heart are all important determinants [ 1 , 2 ]. Despite recent developments in the understanding of the complex interplay of the plethora of biological mechanisms influencing resting heart rate [ 3 ], our understanding is still incomplete.

The acute heart rate response to exercise, heart rate increase during and heart rate recovery after exercise, offers unique insights into cardiac physiology compared to heart rate in rest and can therefore be exploited to obtain additional information on cardiac function [ 4 ]. Impaired increase of heart rate during exercise (chronotropic incompetence) and an attenuated heart rate recovery have been associated with all-cause mortality and sudden cardiac death in healthy individuals [ 5 , 6 , 7 ] and in those with cardiac disease, including individuals with heart failure [ 8 ] and coronary artery disease [ 9 ]. Regular endurance exercise training has been proven to shift the cardiac autonomic balance towards vagal dominance [ 10 ]. The long-term response of heart rate to exercise results in favourable changes in chronotropic function, including decreased resting and submaximal heart rate as well as increased heart rate recovery [ 11 ].

Both the acute and long-term responses of heart rate to exercise have been shown to have a large heritable component [ 12 , 13 , 14 , 15 , 16 , 17 ]. Development in the understanding of the human genome and genetic analysis enables researchers to investigate the possible molecular mechanisms underlying interindividual differences in the acute and long-term heart rate response to exercise [ 18 ]. In this review, we summarize the current knowledge of the acute and long-term heart rate response to exercise, with a focus on the genetic contribution. In addition, we identify gaps in our knowledge and discuss possible future directions that might be of interest to enhance the understanding of the heart rate response to exercise and consider its potential clinical applications.

Acute response

• Heart rate increase

In general, the regulation of the circulatory system during exercise involves several adaptations. These adaptations include dilatation of resistance vessels in the active muscles, a decrease in vagal outflow to the heart, followed by an increase of sympathetic outflow. If exercise is intense, the cholinergic fibers to the adrenal medulla are also activated, resulting in release of epinephrine into the circulation [ 19 ]. Under normal physiological conditions, this results in increased venous return, contractility, and heart rate [ 20 ]. In turn, ejection fraction increases due to a greater ejection of blood at the end of systole and increased diastolic filling of the ventricles as the duration of the systole decreases with increased heart rate [ 20 ].

The increase of heart rate during exercise is for a major part attributable to the decrease in vagal tone followed by an increase in sympathetic outflow and an increase in levels of circulating catecholamines [ 19 ]. It has been shown that a substantial component of interindividual differences in the heart rate increase during exercise is genetically determined, with heritability estimates ranging from 0.17 to 0.32 (Table  1 ) [ 12 , 14 , 15 ]. This suggests that genetic analyses may identify novel biological mechanisms involved in the regulation of heart rate response to exercise.

Several studies have focussed on identifying genetic determinants that explain interindividual differences in heart rate increase during exercise. Genes investigated in these studies are summarized in Table  2 , shown in Fig.  1 , and are further discussed here. The ACE gene was one of the first candidate genes thoroughly investigated for its possible relationship with the heart rate response to exercise [ 21 ]. Genetic association studies focusing on the effect of the ACE gene on heart rate increase during exercise reported many conflicting results [ 12 , 22 , 23 , 24 , 25 ]. Some studies tested genes for their indirect effect on the sympathetic nervous system. One study observed that the NOS3 gene, which produces nitric oxide, was associated with heart rate increase during exercise [ 26 ]. Although nitric oxide is mostly known for its vasodilatory effects, it is also thought to have a modulating effect on the parasympathetic and sympathetic nervous system [ 27 ]. GNAS1 was found to be associated with heart rate increase during exercise as well [ 28 ]. This gene encodes the G protein α-subunit that influences the sympathetic nervous system as it enables the coupling between adenylyl cyclase and β1-adrenergic receptors. On the other hand, many studies brought forward genes based on their direct involvement in the sympathetic nervous system, and associations were found with the ADRB1 [ 29 ] and ADRB2 [ 30 ] genes, which both encode for β-adrenergic receptors. Interestingly, many previous findings could not be replicated in the Framingham Offspring study, which investigated multiple genes instead of focusing on a single gene. In this study, associations were found with the ADRA1A and ADRA1D [ 12 ]. These genes encode for α-adrenergic receptors that are mainly involved in smooth muscle cell contraction during sympathetic stimulation [ 12 ]. However, associations with the ADRB1 and ADRB2 genes could not be re-established [ 12 ].

Graphical representation of genes (shown in italic) involved in acute heart rate increase during exercise grouped by working mechanism (shown in bold). The left and left upper part of the figure shows the nervous system. The middle upper part zooms in on a peripheral sympathetic neuron and its synapse. The heart is displayed on the right; the upper right of the figure shows the aorta with next to it a pacemaker cell in the cardiac sinus node. In the middle of the figure, below, we zoom in on cardiac tissue and receptors. Adrenergic receptors are shown in red. Sodium, potassium, and calcium channels are shown in red, pink, and green, respectively

Although these studies were important for laying the foundation of our knowledge on the genetic determinants of heart rate increase during exercise, they failed to yield a comprehensive view by focusing on one or only a few genes. The Framingham Offspring study was the first to address these issues by conducting an early genetic linkage analysis on heart rate increase and recovery. However, not one genetic signal reached the appropriate significance level, which can possibly be attributed to the relatively low sample size of this study ( n   = 2982) [ 12 ]. In addition, linkage analyses have been shown to be less successful when applied to polygenic traits such as heart rate response to exercise [ 31 ], in part because of their limited power to detect the effect of common alleles with modest effects on disease [ 32 ].

More recently, genome-wide association studies (GWASs) were introduced. GWASs do have the potential to detect common alleles with modest effects on disease, since this method allows an unbiased and comprehensive search across the genome for single nucleotide polymorphisms (SNPs) [ 33 ]. The first GWAS on heart rate increase during exercise found GAR1 and RYR2 genes to be associated [ 34 ]. GAR1 is required for ribosome biogenesis and telomere maintenance. However, its specific function and how it possibly interacts with heart rate increase during exercise is unknown. RYR2 encodes a calcium channel that mediates calcium release from the sarcoplasmic reticulum into the cytoplasm and is therefore essential in triggering cardiac muscle contraction (Table  2 , Fig.  1 ). RYR2 mutations in humans are associated with arrhythmogenic right-ventricular dysplasia and catecholaminergic polymorphic ventricular tachycardia. Interestingly, although caused by a different mutation in the RYR2 gene, both diseases are known to cause exercise-induced tachycardia [ 35 , 36 , 37 ]. However, these associations did not reach genome-wide significance, which might be due to the low sample size ( n  = 1238) [ 34 ].

Increasing the sample size for GWASs has been simplified by the development of inexpensive SNP arrays. Two GWASs were recently conducted on the acute heart rate response to exercise in the same cohort of the UK Biobank [ 14 , 15 ]. The discussion of methodological differences between these studies has been published previously [ 38 ] and is beyond the scope of the current review. However, a summary of important differences is necessary to understand different genes found between the two studies. One difference is that the first study by Verweij et al. had a slightly lower sample size, since they used only echocardiography (ECG) measurements and did not include heart rate measurements derived by the UK Biobank itself. Another difference is that the study of Verweij et al. applied a more stringent threshold to claim a genome-wide significant level to be true (strategy to reduce the risk of type-1 errors) compared to the study published later by Ramirez et al. ( p  < 8.3 × 10 −9 vs p  < 5.0 × 10 −8 , respectively).

Of special interest are three genes that were found to be associated with heart rate increase during exercise in both studies, which are SNCAIP, MCTP2, and POP4 [ 14 , 15 ]. The exact mechanism of SCNAIP is not known so far; however, studies in mice have shown that SCNAIP plays a role in neuronal degeneration (Table  2 , Fig.  1 ) [ 39 , 40 ]. POP4 is involved in the processing of precursor RNAs [ 41 ] and in the DNA damage response [ 42 ], thus preventing accumulation of deleterious mutations and DNA lesions and therefore potentially preventing genomic instabilities and carcinogenesis and prolonging neuronal life span. The MCTP2 gene is more specific to cardiac tissue. A mutation in the MCTP2 is known to cause left-ventricular outflow tract malformations in humans, which may alter the pressure within the ventricular outflow tract. Baroreceptors are densely located in this region and altered blood pressure could therefore lead to altered autonomic feedback on heart rate (Table  2 , Fig.  1 ) [ 43 ]. Several other candidate genes found in these studies already provide a biological hypothesis to account for the associations with heart rate response to exercise. These genes can be broadly categorized into four categories, that is: (1) development of the nervous system, including the CCDC141 [ 44 , 45 ], TCF4 [ 46 , 47 ] , PAX2 [ 48 ] , SOX5 [ 49 , 50 ], and CAV2 [ 51 ] genes; (2) prolongation of neuronal life span, including the SYT10 [ 52 ] gene; (3) cardiac development and disease, including RNF220 [ 53 , 54 ] gene; and, finally, (4) genes involved in cardiac rhythm, including SCN10A [ 55 ] and RGS6 [ 56 , 57 ]. Of these, CCDC141, CAV2 , SYT10, RNF220, and SCN10A were more strongly associated with heart rate recovery after exercise (Tables  2 , 3 ) and will be therefore discussed later. TCF4  is involved in the initiation of neuronal differentiation. Clinically, a mutation in TCF4  is known to cause Pitt-Hopkins syndrome, a severe congenital encephalopathy characterized by intellectual disability, developmental problems, seizures, breathing problems, and typical facial features [ 46 , 47 ]. PAX2 encodes paired box gene 2 and is important in the early embryonic development as well. It is mostly known for its involvement in development of the kidney and urinary tract, since it is linked to papillorenal syndrome [ 58 ] and focal segmental glomerulosclerosis [ 59 ]. However, downstream target effectors of PAX2 have been hypothesized to be involved in neuronal development because of their supposed effect on the CHARGE syndrome [ 48 ]. SOX5 is involved in the regulation of chondrogenesis and the development of the nervous system [ 50 ]. In mice, it was found that loss of SOX5 resulted in decreased neuronal differentiation and secondary migrational abnormalities [ 49 ]. Mutations of the SOX5 gene in humans are known to cause the Lamb–Shaffer syndrome, which is characterized by speech delay, behavioural problems, and nonspecific dysmorphic features [ 50 ]. RGS6 is part of the regulation mechanism of the parasympathetic nervous system in the heart [ 56 , 57 ]. It decreases muscarinic type 2 receptor (M2R) signalling in the sinoatrial node by rapidly terminating Gβγ signalling [ 56 , 57 ]. In mice, it was shown that RGS6 knockdown removes the negative regulation of Gβγ leading to enhanced G protein-coupled inwardly rectifying potassium channel (GIRK)-induced sinoatrial and atrioventricular node hyperpolarization [ 56 , 57 ]. It was therefore concluded that normal function of RGS6 is important for preventing parasympathetic override and severe bradycardia [ 56 ]. Its involvement in the parasympathetic nervous system was recently established in another GWAS in which it was found to be associated with heart rate variability [ 60 ], which is known to reflect parasympathetic activity [ 61 ]. Concerning heart rate increase during exercise, normal function of RGS6 probably facilitates parasympathetic withdrawal leading to the possibility to increase heart rate (Fig.  1 ).

Interestingly, none of the genes investigated in candidate gene studies were found to be associated with heart rate increase in any of the three GWASs. This is in line with the previous work in which early candidate gene studies were difficult to replicate [ 62 , 63 ]. Two genes, HMGA2 and PPIL1 , shown in Table  2 have not been discussed so far. PPIL1 is a gene that was recently found to be associated with heart rate variability as well [ 60 ]. However, to our knowledge, there is no current biological hypothesis to explain the association between PPIL1 or HMGA2 and heart rate increase during exercise.

• Heart rate recovery

Heart rate recovery is characterized by increased parasympathetic tone followed by sympathetic withdrawal, which follows an inversed gradient pattern compared to heart rate increase [ 19 ]. It was elegantly shown in a dual-blockade study that especially parasympathetic reactivation is essential for interindividual differences in heart rate recovery [ 64 ]. However, the exact mechanisms underlying these differences remain to be determined. Twin, family, and GWA studies estimated the genetic component to interindividual differences of heart rate recovery after one minute to range between 0.12 and 0.60 (Table  1 ). Therefore, genetic studies may yield novel insights into heart rate recovery. All genetic determinants investigated for their potential causal role in interindividual differences in heart rate recovery are summarized in Table  3 and are discussed below. An illustration of possible causal genes and how they are supposed to influence acute heart rate recovery after exercise is shown in Fig.  2 .

Graphical representation of genes (shown in italic) involved in acute heart rate recovery after exercise grouped by working mechanism (shown in bold). The left and left upper part of the figure shows the nervous system. The middle upper part zooms in on a parasympathetic neuron of the vagus nerve (twice) and its synapse. Note that although we zoom in on the brain stem (which is the main location of parasympathetic nuclei that innervate the vagus nerve), we actually show a peripheral parasympathetic neuron of the vagus nerve. The heart is displayed on the right; the upper right of the figure shows the aorta with next to it a pacemaker cell in the cardiac sinus node. In the middle of the figure, below, we zoom in on cardiac tissue and receptors. Cholinergic receptors and enzymes are shown in light blue and glutamate receptors in yellow. Sodium and potassium channels are shown in red and pink, respectively

Initially, the same candidate genes were proposed for heart rate recovery as for heart rate increase. For example, the ACE gene was found to be related to heart rate recovery in one candidate gene study as well [ 22 ]. Another study found ADRA1B and ADRA2B to be associated with heart rate recovery (Table  3 ) [ 12 ]. The association between ADRA2B gene and heart rate recovery was also found in another candidate gene study [ 65 ]. Other studies focused primarily on the parasympathetic nervous system represented by the CHRM2 gene. The minor alleles of the rs324640 and rs8191992 SNPs found in CHRM2 region were found to be associated with a lower heart rate recovery in the general population [ 66 ] and in patients with a history of myocardial infarction [ 67 ]. In addition, these minor alleles increased chances of death to coronary artery disease in the latter group [ 67 ].

The problem of biased selection of candidate genes has been solved by conducting GWASs as previously stated. The first GWAS on the acute heart rate response to exercise found heart rate recovery measured 3 min post-exercise to be associated with PRKAG2, though this association did not reach genome-wide significance. PRKAG2 is involved in the regulation of ATP restoration after periods of ATP depletion and therefore might influence the return of heart rate to its initial state (Table  3 , Fig.  2 ).

As previously mentioned, sample size was drastically increased in the two recent studies in the UK Biobank [ 14 , 15 ]. Some differences between both studies have been discussed earlier (i.e., sample size and genome-wide significant threshold). Concerning heart rate recovery, it is worth mentioning that the phenotype definition was not equal between both studies. The study of Ramirez et al. [ 15 ] determined heart rate recovery traditionally as the difference between maximum heart rate and heart rate approximately 1 min after cessation of exercise. The study of Verweij et al. defined heart rate recovery at five time points, which included the differences between maximum heart rate and heart rate after 50, 40, 30, 20, and 10 s after exercise. This includes heart rate recovery at earlier time points (i.e., 10 s), which was recently established to be a superior predictor of outcome of all-cause mortality and death by coronary artery disease [ 6 , 7 ].

Interestingly, both studies found the previously investigated candidate gene CHRM2 to be associated with heart rate recovery [ 14 , 15 ]. CHRM2 encodes M2R, the main muscarinic cholinergic receptor in the heart. This receptor is known for both its negative chronotropic and inotropic effects after binding with acetylcholine released by postganglionic parasympathetic nerves (Table  3 , Fig.  2 ) [ 68 ]. The role of the parasympathetic nervous system in interindividual differences in heart rate recovery is additionally highlighted by the ACHE gene that was found in both studies. ACHE encodes for acetylcholinesterase, an enzyme which breaks down acetylcholine in the synaptic cleft of postganglionic parasympathetic nerves [ 69 ]. An increase of acetylcholinesterase would therefore cause an attenuated heart rate recovery by decreasing parasympathetic reactivation. Other genes that were found in both studies were SYT10, CNTN3, PAX2, CAV2, MED13L, RNF220, and NDUFA11 (Table  3 , Fig.  2 ) . SYT10 encodes a Ca 2+  sensor synaptotagmin 10 that triggers IGF-1 exocytosis, which, in turn, protects neurons from degeneration. SYT10 might play an important role in the regulation of heart rate, as it was found to be associated with resting heart rate [ 3 , 23 ], heart rate increase [ 15 ], and heart rate variability [ 60 ] as well. CNTN3 belongs to a group of glycosylphosphatidyl-anchored cell adhesion molecules that are mostly found in neurons [ 70 , 71 ]. Because of its similarity with TAG - 1 , it is thought to have an important function in neuronal outgrowth and wiring of the nervous system [ 70 , 71 , 72 ]. In the study of Ramirez et al. it was found that the allele of one SNP decreased heart rate recovery and increased CNTN3 expression levels in the nucleus accumbens [ 15 ]. Since heart rate recovery is mainly influenced by the parasympathetic nervous system [ 64 ], it was hypothesized that CNTN3 may also be relevant to cardiac parasympathetic modulation [ 15 ]. However, it is more likely to be associated with cardiac sympathetic modulation, since morphology of the nucleus accumbens has been shown to be correlated with cardiac sympathetic index [ 73 ]. PAX2 is known to be the first gene to be expressed in the mid- and hindbrain during embryonal developments in mice [ 74 ] and can be found in the hindbrain in the early stages of embryo development in humans as well [ 48 ]. The hindbrain includes the nucleus tractus solitarius, nucleus ambiguous, and dorsal nucleus of the vagus, which are known to mainly influence cardiac parasympathetic innervation of the heart through vagus nerve stimulation [ 75 ]. Less is known about CAV2 , which was found to be associated with heart rate response to exercise as well. However, one study pointed out that CAV2 is necessary for differentiation of dorsal root ganglion cells during the early differentiating programs [ 51 ]. The function of MED13L is unclear as well, but knockdown in zebrafish caused abnormal neural-crest cell migration [ 76 ]. This is supported by clinical characteristics in humans with MED13L mutations, which can be characterized by intellectual disabilities, developmental delay, and craniofacial anomalies [ 77 ]. RNF220 functions as an E3 ubiquitin ligase, which determines protein target specificity during posttranslational ubiquitination [ 53 ]. A possible link with heart rate recovery originates from the involvement of RNF220 in the canonical WNT signalling cascade. In a knockdown study, RNF220 was shown to stabilize β-catenin by interacting with ubiquitin-specific peptidase Usp7 [ 54 ]. This stabilizing function is important, because the WNT/β-catenin signalling pathway is involved in embryonic cardiac development [ 78 ], the development of cardiac disease [ 79 , 80 , 81 ], and in cardiac repair [ 80 ]. NDUFA11 is an accessory subunit of the mitochondrial membrane respiratory chain NADH dehydrogenase complex I. In humans, a splice‐site mutation in this gene is known to cause mitochondrial complex I deficiency. This can cause a wide range of disorders, including encephalocardiomyopathy [ 82 ]. Recently, it was shown that downregulation of NDUFA11 by small interfering RNA reduced ATP production and increased mitochondria reactive oxygen species production in cardiac mitochondria of mice [ 83 ]. NDUFA11 was found to be associated with heart rate variability as well, suggesting that it is an important factor in causing differences between individuals’ heart rate response [ 60 ].

Other candidate genes found in one of the GWASs provide a biological hypothesis for their possible causal role in interindividual differences in heart rate recovery as well. These genes include CCDC141, BCL11A, KCNH8, ALG10B, GNG11, GRIK2, and NEGR1. CCDC141 is a gene that plays a central role in neuronal development [ 44 , 45 ]. In fact, in utero knockdown of CCDC141 in mice resulted in impaired radial migration in [ 44 ]. The same applies to BCL11A , which encodes a C2H2-type zinc-finger protein that is involved in neuronal development. Studies in mice have shown that slowed migration of neurons upon knockdown resulted in microcephaly with decreased brain volume [ 84 ], particularly affecting the limbic system [ 85 ]. Within the human brain, it is most highly expressed in the caudate nucleus followed by hippocampus [ 86 ]. In humans, different de novo heterozygous mutations have been found to cause developmental disorder with persistence of fetal haemoglobin [ 85 ]. KCNH8 encodes a voltage-gated potassium channel. It is mainly expressed in the central nervous system and is involved in the regulation of neuronal excitation (Table  3 , Fig.  2 ) [ 87 , 88 , 89 ]. ALG10B is involved in potassium regulation, as well, since it is a potassium channel regulator that couples to KCNH2 . However, it is more involved in cardiac tissue than neuronal tissue and is known for its influence on heart rhythm. Upon binding with KCNH2, it reduces sensitivity to classic proarrhythmic drug blockade [ 90 ]. GNG11 encodes the γ11 subunit of the heterotrimeric G protein complex Gαβγ [ 91 ]. GNG11 is just as RGS6 thought to be involved in GIRK activation and was found to be associated with heart rate variability [ 60 ] as well. In this study, it was hypothesized that variations in this gene lower the availability of the γ11 subunit, thereby reducing Gαβγ component-induced GIRK activation [ 60 ]. This would lead to decreased heart rate variability through attenuated response to changes in cardiac vagal activity [ 60 ]. If true, the same would apply for heart rate recovery; decreased response to cardiac vagal reactivation after exercise would translate to blunted heart rate recovery. In addition, another mutation in the RGS6 gene in humans was shown to decrease susceptibility to the long QT syndrome [ 92 ]. GRIK2 encodes a glutamate receptor that is mostly expressed in the human cerebral and cerebellar cortices [ 93 ]. Here, it is involved in neuronal excitation and plays an important role in a variety of normal neurophysiologic processes. Neuronal Growth Regulator 1 ( NEGR1 ) is essential for neuronal morphology and, just as CNTN3 , has been shown to regulate neurite outgrowth (Table  3 , Fig.  2 ) [ 94 ]. Perhaps because of this essential function, NEGR1 has been associated with many polygenetic traits, including body mass index, years of education, and physical activity.

Heart rate increase and recovery share a high genetic correlation and it is therefore likely that there is overlap in genes that were found for both aspects of the heart rate response to exercise [ 14 ]. SNCAIP, SOX5, RGS6, and MCTP2 genes were already discussed for heart rate increase during exercise because of their stronger association with this phenotype.

BCAT1, CLPB, PRDM6, SKAP, and SERINC2 are also shown in Table  3 , but have not been discussed yet. To our knowledge, these genes could not be linked to heart rate recovery after exercise on a biological basis so far.

Long-term modification of the heart rate response to exercise

Regular endurance exercise training is known to shift the cardiac autonomic balance towards vagal dominance [ 10 ] and, as a consequence, diminish submaximal heart rate when an individual cycles at the same intensity [ 11 ]. Large interindividual differences were observed for submaximal heart rate training response [ 95 ] and heritability analysis estimated a genetic component ranging between 0.34 and 0.36 (Table  1 ) [ 16 , 17 ]. Therefore, several studies were conducted to gain insights in the causes of these interindividual differences. The first study in the HERITAGE family cohort found a heritability of 0.34 for exercise heart rate changes to regular training, with the strongest linkage on chromosome 2q33.3-q34 [ 17 ]. Next, this region was fine-mapped and it was found that the CREB1 gene locus was strongly associated with submaximal exercise heart rate training response [ 96 ]. Nonetheless, it only explained 5.45% of the 34% heritability [ 96 ].

To gain further insights in the genes causing the remaining fraction of its heritability, a GWAS was performed in the HERITAGE family cohort. In this study, nine SNPs were identified and accounted for the total of 34% heritability of exercise-induced changes to heart rate increase [ 97 ]. The most significantly associated SNP was linked to the YWHAQ gene (Table  4 ). YWHAQ is mostly expressed in the brain, heart, and pancreas [ 98 ], and its main function is apoptosis and cell proliferation. It was shown that the cardiac-specific mutated YWHAQ gene leads to increased pathological ventricular remodelling with increased cardiomyocyte apoptosis after experimental myocardial infarction [ 99 ]. It can be hypothesized that mutations in the YWHAQ gene lead to similar pathological cardiac remodelling after exercise training, causing diminished exercise-induced changes to heart rate increase. However, a neurological causal pathway cannot be ruled out, since the same mechanism could apply to neuronal remodelling needed to attenuate heart rate increase after regular exercise training [ 11 ].The CREB1 gene (Table  4 ) was significantly associated with submaximal heart rate response to exercise training as well [ 97 ]. In this study, it was hypothesized that CREB1 altered the exercise-induced changes in heart rate increase due to its effect on either cardiac [ 100 ] or neuronal memory [ 101 ]. Cardiac memory is a phenomenon in which an altered T wave on electrocardiogram can be seen when sinus rhythm restarts after a period of abnormal rhythm, for example, after ventricular pacing or arrhythmia [ 100 ]. The other hypothesis involving neuronal memory fits in our current understanding that neuron biology is of great importance in the heart rate response to exercise. Neuronal memory or long-term potentiation is a form of synaptic plasticity in which there is a long-lasting increase of synaptic strength in case the synapse is highly active. It could be hypothesized that regular exercise causes an increase of synaptic strength of parasympathetic neurons, thus altering the heart rate increase during exercise. However, CREB1 encodes a transcription factor that regulates many mechanisms in the body and its association with memory does not imply causality. A recent editorial rightfully addressed the fact that the same allele in another study was found to increase the rise of temperature [ 102 ] and, therefore, might decrease subjective liking of exercise training, potentially diminishing motivation [ 103 ].

On the other hand, heart rate recovery increases when the cardiac autonomic balance shifts towards vagal dominance after regular endurance training [ 10 ]. Little research has been performed on the genetics of training-induced changes to heart rate recovery, although a heritable component has been suggested [ 66 ]. To our knowledge, only one study has been conducted on this subject. In this candidate gene study, it was found that the CHRM2 gene (Table  4 ) is linked to long-term modification of heart rate recovery to exercise training as well [ 66 ]. Participants who had a the minor alleles of the rs324640 and rs8191992 SNPs were not only found to have a lower acute heart rate recovery, but also showed less increase in heart rate recovery after regular endurance training. As previously mentioned, CHRM2 encodes the muscarinic acetylcholine receptor M2R and, upon activation, causes a negative chronotropic and inotropic response. It therefore seems that genetic variation in CHRM2 not only causes interindividual differences in acute heart rate recovery [ 68 ], but also in long-term modifications. A full overview of the genes discussed for the long-term heart rate response to exercise can be found in Table  4 .

Association of heart rate response to exercise-related genes with other traits

We assessed the association of described genes with other traits in publicly available GWASs using the GWAS catalogue (Online Resource 1). In short, the candidate causal genes that were associated with both heart rate increase and recovery were also associated with resting heart rate ( CCDC141 , RGS6 , RNF220 , SCN10A , and SYT10 ), heart rate variability ( CCDC141 , RGS6 , RNF220 , and SYT10 ), blood pressure ( CCDC141 and PAX2 ), atrial fibrillation ( CAV2 and SCN10A ), coronary artery disease ( CAV2 and SCN10A ), and ECG traits including the PR interval ( CAV2 and SCN10A ), QRS duration, and the Brugada syndrome (both SN10A ).

Some genes that were only associated with heart rate increase during exercise were found to be associated with resting heart rate and heart rate variability ( PPIL1 ), blood pressure ( ADRB1 , ACE , NOS3 , and HMGA2 ), atrial fibrillation ( MCTP2 and NOS3 ), exercise treadmill test and lung function (both RYR2 ). Similarly, some of the heart rate recovery genes were also associated with resting heart rate ( ACHE and GNG11 ), heart rate variability ( GNG11 and NDUFA11 ), blood pressure ( PRDM6 , PRKAG2 , and CHRM2 ), QRS duration, ( PRDM6 ), atrial fibrillation and coronary artery disease ( BCL11A , PRDM6 ), and obesity and vigorous physical activity levels (both NEGR1 ).

Future directions

Improvement of prevention and treatment of disease in the human health sector is the ultimate application of novel knowledge found by genetic studies and future research should be performed to achieve this goal ( Fig.  3 ) [ 104 ]. Functional follow-up of findings obtained by GWAS will be necessary to gain insights in how likely causal genes affect the heart rate response to exercise [ 104 ]. Most genes that were prioritized so far have a plausible biological mechanism in which they influence the heart rate response to exercise. However, the exact effect of all genes on exercise-induced heart rate changes could be validated in an experimental setting (Fig.  3 ). One possible method is to perform functional experiments in cardiomyocytes obtained from embryonic stem cells [ 105 ]. In cardiomyocytes, human diseases and risk factors with their underlying genetic contribution can be created in vitro [ 105 ]. Since cardiomyocyte cell cultures can beat spontaneously [ 105 ], simulating the effect of this genetic contribution allows for investigation of the acute heart rate response to pacing from resting to exercise heart rate levels in small cell cultures. In addition, by simulating the effect of this genetic contribution, drugs can be screened against an individual’s full genetic backgrounds to discover information on cardiotoxicity for each individual. This could potentially give insights in the development of personalized medicine strategies for heart rate modification [ 106 ], which is an essential strategy in the treatment of coronary artery disease [ 107 ] and heart failure (Fig.  3 ) [ 108 , 109 ]. Genes known to affect cardiac de- and repolarisation ( RYR2 , ALG10B, and SCN10A ) or GIRK channels in the cardiac sinus node ( RGS6 and GNG11 ) could be of interest to study in this setting. Recent development in the generation of spinal human cord neural cells could provide the same opportunity for investigating neuronal cell longevity including genes such as SCNAIP , POP4, and SYT10 [ 110 ]. Complex neurological mechanisms at the interplay of the sympathetic and parasympathetic nervous system (i.e., KCNH8 and GRIK2 ) or neuronal development (i.e., SOX5 , PAX2, and BCL11A ) are more difficult to investigate using this method. This can be solved by investigating these genes using in vivo models of animals that share a high percentage of their genomic pattern with humans, including mice [ 111 , 112 ], fruit flies [ 113 ], and zebrafish [ 114 ] (Fig.  3 ). For example, knockdown of RGS6 [ 56 , 57 ] , MED13L [ 76 ], and BCL11A [ 85 ] has already provided insights in biological consequences of mutations in these genes.

Possible follow-up of GWAS on heart rate response to exercise. Cell models based on pluripotent stem cells provide a potential functional model to study GWAS findings using experimental manipulations that cannot be performed in vivo. Complex mechanisms of genetic interplay could be studied in animals that share a high percentage of their genomic sequence with humans, including mice, fruit flies, and zebrafishes. Tools such as gene knockdowns can be used to manipulate the genomes of these animal models. The ultimate application of knowledge initiated by GWAS findings in heart rate response to exercise lies in the improvement of primary and secondary prevention and personalized medicine to improve human health

Improvement of prevention of disease is another goal of genetic research. While accurate risk prediction might be relatively straightforward for mono- and oligogenic disease, this is more difficult for polygenic diseases such as coronary artery disease and heart failure. However, the knowledge on genetic variants obtained by GWAS can be used to construct genetic risk cores by summing the number of risk alleles weighted by the corresponding beta coefficients. Recently, it was shown that the polygenic risk score of coronary artery disease had the ability to identify 8.0% of the population at greater than threefold risk for coronary artery disease [ 115 ]. These individuals can subsequently be selected for encouragement of behavioural lifestyle changes as relative effects of poor lifestyle were shown to be comparable between genetic risk groups [ 116 ]. Similar to the traditional risk score models in which several traditional risk phenotypes are used to predict risk events, this could ultimately be performed for genetic risk score models as well. As previously stated, there is a large body of observational studies that links heart rate response to exercise to all-cause mortality and cardiovascular disease in healthy individuals and those with a history of cardiac disease [ 5 , 6 , 7 , 8 , 9 ]. In this light, it would be interesting to see whether adding the polygenetic risk scores for the acute heart rate response to exercise into a genetic risk score model that includes the polygenetic risk score for the cardiovascular disease itself could improve detection of individuals at high risk of disease. However, it should be noted that both recent GWAS on the acute heart rate response to exercise did not find support for a genetic association with cardiovascular mortality [ 14 , 15 ]. The lack of an association in both studies might originate from the fact that a small replication cohort consisting of a relatively young and healthy population was used. The study of Verweij et al. [ 14 ] did find a significant association between heart rate response to exercise and parental age as proxy for all-cause mortality. However, first, it is required to investigate whether there is a genetic association with cardiovascular disease and all-cause mortality, preferably in a larger independent cohort [ 117 ].

The evidence on long-term modification of the heart rate response to exercise is limited so far [ 97 ]. If the genetics of the acute heart rate response to exercise can be used to predict cardiovascular mortality, the combination with information on the genetics of the long-term modification of the heart rate response to exercise could one day inform the choice of prevention strategy. For example, a high genetic risk score for a diminished acute response to exercise combined with a genetic risk score that indicates high training-induced changes to heart rate response could be an indicator of early primary or secondary prevention strategies (Fig.  3 ). On the other hand, a high genetic risk score for a diminished acute response to exercise combined with a genetic risk score that indicates little training-induced changes could be an indication of early intervention through medication (Fig.  3 ).

In the current review, we found a total of 10 genes associated with the acute heart rate response to exercise in candidate gene studies. Only one gene ( CHRM2 ), related to heart rate recovery, was replicated in recent GWASs. Additional 17 candidate causal genes were identified for heart rate increase and 26 for heart rate recovery in these GWASs. Nine of these genes were associated with both acute heart rate increase and recovery during exercise. These genes can be broadly categorized into four categories: (1) development of the nervous system ( CCDC141 , PAX2 , SOX5, and CAV2 ); (2) prolongation of neuronal life span ( SYT10 ); (3) cardiac development ( RNF220 and MCTP2 ), and (4) cardiac rhythm ( SCN10A and RGS6 ). Of the total of 43 genes, nine showed overlap with resting heart rate and heart rate variability, six with atrial fibrillation and coronary artery disease, two with ECG traits, and nine with blood pressure. The current findings support the idea that the autonomic nervous system is a major player in the regulation of the acute heart rate response to exercise. Heart rate recovery is especially influenced by parasympathetic nervous system genes ( ACHE and CHRM2 ), in line with the previous research [ 64 ]. Regarding the long-term response to exercise, heart rate increase during exercise was found to be mainly associated with genes involved in either cardiac or neuronal remodelling. Little evidence has been found for the long-term response of heart rate recovery to exercise, except for parasympathetic involvement. Future work will be required to translate these findings to preventive and therapeutic applications.

Hammond HK, Froelicher VF (1985) Normal and abnormal heart rate responses to exercise. Prog Cardiovasc Dis 27:271–296

Article   CAS   PubMed   Google Scholar  

Bahrainy S, Levy WC, Busey JM et al (2016) Exercise training bradycardia is largely explained by reduced intrinsic heart rate. Int J Cardiol 222:213–216. https://doi.org/10.1016/j.ijcard.2016.07.203

Article   PubMed   PubMed Central   Google Scholar  

Eppinga RN, Hagemeijer Y, Burgess S et al (2016) Identification of genomic loci associated with resting heart rate and shared genetic predictors with all-cause mortality. Nat Genet 48:1557–1563. https://doi.org/10.1038/ng.3708

Fletcher GF, Balady GJ, Amsterdam EA et al (2001) Exercise standards for testing and training: a statement for healthcare professionals from the American Heart Association. Circulation 104:1694–1740

Jouven X, Empana J-P, Schwartz PJ et al (2005) Heart-rate profile during exercise as a predictor of sudden death. N Engl J Med 352:1951–1958. https://doi.org/10.1056/NEJMoa043012

McCrory C, Berkman LF, Nolan H et al (2016) Speed of heart rate recovery in response to orthostatic challenge novelty and significance. Circ Res 119:666–675. https://doi.org/10.1161/CIRCRESAHA.116.308577

van de Vegte YJ, van der Harst P, Verweij N (2018) Heart rate recovery 10 seconds after cessation of exercise predicts death. J Am Heart Assoc 7:e008341. https://doi.org/10.1161/JAHA.117.008341

Arena R, Myers J, Abella J et al (2010) The prognostic value of the heart rate response during exercise and recovery in patients with heart failure: influence of beta-blockade. Int J Cardiol 138:166–173. https://doi.org/10.1016/j.ijcard.2008.08.010

Article   PubMed   Google Scholar  

Dresing TJ, Blackstone EH, Pashkow FJ et al (2000) Usefulness of impaired chronotropic response to exercise as a predictor of mortality, independent of the severity of coronary artery disease. Am J Cardiol 86:602–609. https://doi.org/10.1016/S0002-9149(00)01036-5

Hautala AJ, Mäkikallio TH, Kiviniemi A et al (2004) Heart rate dynamics after controlled training followed by a home-based exercise program. Eur J Appl Physiol 92:289–297. https://doi.org/10.1007/s00421-004-1077-6

Brubaker PH, Kitzman DW (2011) Chronotropic incompetence: causes, consequences, and management. Circulation 123(9):1010–1020. https://doi.org/10.1161/CIRCULATIONAHA.110.940577

Ingelsson E, Larson MG, Vasan RS et al (2007) Heritability, linkage, and genetic associations of exercise treadmill test responses. Circulation 115:2917–2924. https://doi.org/10.1161/CIRCULATIONAHA.106.683821

Nederend I, Schutte NM, Bartels M et al (2016) Heritability of heart rate recovery and vagal rebound after exercise. Eur J Appl Physiol 116:2167–2176. https://doi.org/10.1007/s00421-016-3459-y

Verweij N, van de Vegte YJ, van der Harst P (2018) Genetic study links components of the autonomous nervous system to heart-rate profile during exercise. Nat Commun 9:898. https://doi.org/10.1038/s41467-018-03395-6

Article   CAS   PubMed   PubMed Central   Google Scholar  

Ramírez J, van Duijvenboden S, Ntalla I et al (2018) Thirty loci identified for heart rate response to exercise and recovery implicate autonomic nervous system. Nat Commun 9:1947. https://doi.org/10.1038/s41467-018-04148-1

Rice T, An P, Gagnon J et al (2002) Heritability of HR and BP response to exercise training in the HERITAGE Family Study. Med Sci Sports Exerc 34:972–979

An P, Pérusse L, Rankinen T et al (2003) Familial aggregation of exercise heart rate and blood pressure in response to 20 weeks of endurance training: the HERITAGE family study. Int J Sports Med 24:57–62. https://doi.org/10.1055/s-2003-37200

Visscher PM, Wray NR, Zhang Q et al (2017) 10 years of GWAS discovery: biology, function, and translation. Am J Hum Genet 101:5–22. https://doi.org/10.1016/j.ajhg.2017.06.005

Coote JH (2010) Recovery of heart rate following intense dynamic exercise. Exp Physiol 95:431–440. https://doi.org/10.1113/expphysiol.2009.047548

Saghiv M, Sagiv M (2017) Response of left ventricular volumes and ejection fraction during different modes of exercise in health and CAD patients. Int J Clin Cardiol Res 1:51–56

Google Scholar  

Montgomery HE, Marshall R, Hemingway H et al (1998) Human gene for physical performance. Nature 393:221–222. https://doi.org/10.1038/30374

Voroshin IN, Astratenkova IV (2008) Dependence of endurance performance on ACE gene polymorphism in athletes. Hum Physiol 34:117–119. https://doi.org/10.1007/s10747-008-1018-6

Article   CAS   Google Scholar  

Roltsch MH, Brown MD, Hand BD et al (2005) No association between ACE I/D polymorphism and cardiovascular hemodynamics during exercise in young women. Int J Sports Med 26:638–644. https://doi.org/10.1055/s-2004-830436

McCole SD, Brown MD, Moore GE et al (2002) Angiotensinogen M235T polymorphism associates with exercise hemodynamics in postmenopausal women. Physiol Genomics 10:63–69. https://doi.org/10.1152/physiolgenomics.00106.2001

Rankinen T, Pérusse L, Gagnon J et al (2000) Angiotensin-converting enzyme ID polymorphism and fitness phenotype in the HERITAGE Family Study. J Appl Physiol 88:1029–1035. https://doi.org/10.1152/jappl.2000.88.3.1029

Hand BD, McCole SD, Brown MD et al (2006) NOS3 gene polymorphisms and exercise hemodynamics in postmenopausal women. Int J Sports Med 27:951–958. https://doi.org/10.1055/s-2006-923901

Zanzinger J (1999) Role of nitric oxide in the neural control of cardiovascular function. Cardiovasc Res 43:639–649. https://doi.org/10.1016/s0008-6363(99)00085-1

Nieminen T, Lehtimäki T, Laiho J et al (2006) Effects of polymorphisms in beta1-adrenoceptor and alpha-subunit of G protein on heart rate and blood pressure during exercise test. The Finnish Cardiovascular Study. J Appl Physiol 100:507–511. https://doi.org/10.1152/japplphysiol.00899.2005

Defoor J, Martens K, Zielińska D et al (2006) The CAREGENE study: polymorphisms of the β1-adrenoceptor gene and aerobic power in coronary artery disease. Eur Heart J 27:808–816. https://doi.org/10.1093/eurheartj/ehi737

Eisenach JH, McGuire AM, Schwingler RM et al (2004) The Arg16/Gly β 2 -adrenergic receptor polymorphism is associated with altered cardiovascular responses to isometric exercise. Physiol Genomics 16:323–328. https://doi.org/10.1152/physiolgenomics.00152.2003

Altmüller J, Palmer LJ, Fischer G et al (2001) Genomewide scans of complex human diseases: true linkage is hard to find. Am J Hum Genet 69:936–950. https://doi.org/10.1086/324069

Hirschhorn JN, Daly MJ (2005) Genome-wide association studies for common diseases and complex traits. Nat Rev Genet 6:95–108. https://doi.org/10.1038/nrg1521

Hirschhorn JN (2005) Genetic approaches to studying common diseases and complex traits. Pediatr Res 57:74R–77R. https://doi.org/10.1203/01.PDR.0000159574.98964.87

Vasan RS, Larson MG, Aragam J et al (2007) Genome-wide association of echocardiographic dimensions, brachial artery endothelial function and treadmill exercise responses in the Framingham Heart Study. BMC Med Genet 8:S2. https://doi.org/10.1186/1471-2350-8-S1-S2

Tiso N, Stephan DA, Nava A et al (2001) Identification of mutations in the cardiac ryanodine receptor gene in families affected with arrhythmogenic right ventricular cardiomyopathy type 2 (ARVD2). Hum Mol Genet 10:189–194

Tsio N, Salamon M, Bagattin A et al (2002) The binding of the RyR2 calcium channel to its gating protein FKBP12.6 is oppositely affected by ARVD2 and VTSIP mutations. Biochem Biophys Res Commun 299:594–598. https://doi.org/10.1016/S0006-291X(02)02689-X

Priori SG, Napolitano C, Memmi M et al (2002) Clinical and molecular characterization of patients with catecholaminergic polymorphic ventricular tachycardia. Circulation 106:69–74

Huffman JE (2018) Examining the current standards for genetic discovery and replication in the era of mega-biobanks. Nat Commun 9:5054. https://doi.org/10.1038/s41467-018-07348-x

Nuber S, Franck T, Wolburg H et al (2010) Transgenic overexpression of the alpha-synuclein interacting protein synphilin-1 leads to behavioral and neuropathological alterations in mice. Neurogenetics 11:107–120. https://doi.org/10.1007/s10048-009-0212-2

Smith WW, Liu Z, Liang Y et al (2010) Synphilin-1 attenuates neuronal degeneration in the A53T -synuclein transgenic mouse model. Hum Mol Genet 19:2087–2098. https://doi.org/10.1093/hmg/ddq086

Jarrous N, Eder PS, Wesolowski D, Altman S (1999) Rpp14 and Rpp29, two protein subunits of human ribonuclease P. RNA 5:153–157

Abu-Zhayia ER, Khoury-Haddad H, Guttmann-Raviv N et al (2017) A role of human RNase P subunits, Rpp29 and Rpp21, in homology directed-repair of double-strand breaks. Sci Rep 7:1002. https://doi.org/10.1038/s41598-017-01185-6

Raven PB, Chapleau MW (2014) Blood pressure regulation XI: overview and future research directions. Eur J Appl Physiol 114:579–586. https://doi.org/10.1007/s00421-014-2823-z

Fukuda T, Sugita S, Inatome R, Yanagi S (2010) CAMDI, a novel disrupted in schizophrenia 1 (DISC1)-binding protein, is required for radial migration. J Biol Chem 285:40554–40561. https://doi.org/10.1074/jbc.M110.179481

Brandon NJ, Sawa A (2011) Linking neurodevelopmental and synaptic theories of mental illness through DISC1. Nat Rev Neurosci 12:707–722. https://doi.org/10.1038/nrn3120

De Pontual L, Mathieu Y, Golzio C et al (2009) Mutational, functional, and expression studies of the TCF4 gene in pitt-hopkins syndrome. Hum Mutat 30:669–676. https://doi.org/10.1002/humu.20935

Sepp M, Pruunsild P, Timmusk T (2012) Pitt-Hopkins syndrome-associated mutations in TCF4 lead to variable impairment of the transcription factor function ranging from hypomorphic to dominant-negative effects. Hum Mol Genet 21:2873–2888. https://doi.org/10.1093/hmg/dds112

Tellier A-L, Amiel J, Delezoide A-L et al (2000) Expression of thePAX2 gene in human embryos and exclusion in the CHARGE syndrome. Am J Med Genet 93:85–88. https://doi.org/10.1002/1096-8628(20000717)93:2%3c85:AID-AJMG1%3e3.0.CO;2-B

Lai T, Jabaudon D, Molyneaux BJ et al (2008) SOX5 controls the sequential generation of distinct corticofugal neuron subtypes. Neuron 57:232–247. https://doi.org/10.1016/J.NEURON.2007.12.023

Lamb AN, Rosenfeld JA, Neill NJ et al (2012) Haploinsufficiency of SOX5 at 12p12.1 is associated with developmental delays with prominent language delay, behavior problems, and mild dysmorphic features. Hum Mutat 33:728–740. https://doi.org/10.1002/humu.22037

Galbiati F, Volonte D, Gil O et al (1998) Expression of caveolin-1 and -2 in differentiating PC12 cells and dorsal root ganglion neurons: caveolin-2 is up-regulated in response to cell injury. Proc Natl Acad Sci USA 95:10257–10262

Woitecki AMH, Müller JA, van Loo KMJ et al (2016) Identification of synaptotagmin 10 as effector of NPAS4-mediated protection from excitotoxic neurodegeneration. J Neurosci 36:2561–2570. https://doi.org/10.1523/JNEUROSCI.2027-15.2016

Kong Q, Zeng W, Wu J et al (2010) RNF220, an E3 ubiquitin ligase that targets Sin3B for ubiquitination. Biochem Biophys Res Commun 393:708–713. https://doi.org/10.1016/j.bbrc.2010.02.066

Ma P, Yang X, Kong Q et al (2014) The ubiquitin ligase RNF220 enhances canonical Wnt signaling through USP7-mediated deubiquitination of β-catenin. Mol Cell Biol 34:4355–4366. https://doi.org/10.1128/MCB.00731-14

Facer P, Punjabi PP, Abrari A et al (2011) Localisation of SCN10A gene product Nav1.8 and novel pain-related ion channels in human heart. Int Heart J 52:146–152. https://doi.org/10.1536/ihj.52.146

Yang J, Huang J, Maity B et al (2010) RGS6, a modulator of parasympathetic activation in heart. Circ Res 107:1345–1349. https://doi.org/10.1161/CIRCRESAHA.110.224220

Wydeven N, Posokhova E, Xia Z et al (2014) RGS6, but not RGS4, is the dominant regulator of G protein signaling (RGS) modulator of the parasympathetic regulation of mouse heart rate. J Biol Chem 289:2440–2449. https://doi.org/10.1074/jbc.M113.520742

Bower M, Salomon R, Allanson J et al (2012) Update of PAX2 mutations in renal coloboma syndrome and establishment of a locus-specific database. Hum Mutat 33:457–466. https://doi.org/10.1002/humu.22020

Barua M, Stellacci E, Stella L et al (2014) Mutations in PAX2 associate with adult-onset FSGS. J Am Soc Nephrol 25:1942–1953. https://doi.org/10.1681/ASN.2013070686

Nolte IM, Munoz ML, Tragante V et al (2017) Genetic loci associated with heart rate variability and their effects on cardiac disease risk. Nat Commun 8:15805. https://doi.org/10.1038/ncomms15805

Camm AJMM et al (1996) Heart rate variability: standards of measurement, physiological interpretation and clinical use. Task Force of the European Society of Cardiology and the North American Society of Pacing and Electrophysiology. Circulation 93:1043–1065

Article   Google Scholar  

Ioannidis JPA, Ntzani EE, Trikalinos TA, Contopoulos-Ioannidis DG (2001) Replication validity of genetic association studies. Nat Genet 29:306–309. https://doi.org/10.1038/ng749

Morgan TM, Krumholz HM, Lifton RP, Spertus JA (2007) Nonvalidation of reported genetic risk factors for acute coronary syndrome in a large-scale replication study. JAMA 297:1551. https://doi.org/10.1001/jama.297.14.1551

Imai K, Sato H, Hori M et al (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. https://doi.org/10.1016/0735-1097(94)90150-3

Kohli U, Diedrich A, Kannankeril PJ et al (2015) Genetic variation in alpha2-adrenoreceptors and heart rate recovery after exercise. Physiol Genomics 47:400–406. https://doi.org/10.1152/physiolgenomics.00124.2014

Hautala AJ (2006) Heart rate recovery after maximal exercise is associated with acetylcholine receptor M2 (CHRM2) gene polymorphism. AJP Hear Circ Physiol 291:H459–H466. https://doi.org/10.1152/ajpheart.01193.2005

Hautala AJ, Tulppo MP, Kiviniemi AM et al (2009) Acetylcholine receptor M2 gene variants, heart rate recovery, and risk of cardiac death after an acute myocardial infarction. Ann Med 41:197–207. https://doi.org/10.1080/07853890802477866

Brodde OE, Michel MC (1999) Adrenergic and muscarinic receptors in the human heart. Pharmacol Rev 51:651–690

CAS   PubMed   Google Scholar  

Whittaker VP (1990) The contribution of drugs and toxins to understanding of cholinergic function. Trends Pharmacol Sci 11:8–13. https://doi.org/10.1016/0165-6147(90)90034-6

Nikolaienko RM, Hammel M, Dubreuil V et al (2016) Structural basis for interactions between contactin family members and protein-tyrosine phosphatase receptor type G in neural tissues. J Biol Chem 291:21335–21349. https://doi.org/10.1074/jbc.M116.742163

Walsh CA, Morrow EM, Rubenstein JLR (2008) Autism and brain development. Cell 135:396–400. https://doi.org/10.1016/J.CELL.2008.10.015

Connelly MA, Grady RC, Mushinski JF, Marcu KB (1994) PANG, a gene encoding a neuronal glycoprotein, is ectopically activated by intracisternal A-type particle long terminal repeats in murine plasmacytomas. Proc Natl Acad Sci USA 91:1337–1341

Ruffle JK, Coen SJ, Giampietro V et al (2018) Morphology of subcortical brain nuclei is associated with autonomic function in healthy humans. Hum Brain Mapp 39:381–392. https://doi.org/10.1002/hbm.23850

Pfeffer PL, Payer B, Reim G et al (2002) The activation and maintenance of Pax2 expression at the mid-hindbrain boundary is controlled by separate enhancers. Development 129:307–318

Kapa S, DeSimone CV, Asirvatham SJ (2016) Innervation of the heart: an invisible grid within a black box. Trends Cardiovasc Med 26:245–257. https://doi.org/10.1016/j.tcm.2015.07.001

Utami KH, Winata CL, Hillmer AM et al (2014) Impaired development of neural-crest cell-derived organs and intellectual disability caused by MED13L haploinsufficiency. Hum Mutat 35:1311–1320. https://doi.org/10.1002/humu.22636

Asadollahi R, Zweier M, Gogoll L et al (2017) Genotype-phenotype evaluation of MED13L defects in the light of a novel truncating and a recurrent missense mutation. Eur J Med Genet 60:451–464. https://doi.org/10.1016/j.ejmg.2017.06.004

Piven OO, Winata CL (2017) The canonical way to make a heart: β-catenin and plakoglobin in heart development and remodeling. Exp Biol Med 242:1735–1745. https://doi.org/10.1177/1535370217732737

Nakagawa A, Naito AT, Sumida T et al (2016) Activation of endothelial β-catenin signaling induces heart failure. Sci Rep 6:25009. https://doi.org/10.1038/srep25009

Pahnke A, Conant G, Huyer LD et al (2016) The role of Wnt regulation in heart development, cardiac repair and disease: a tissue engineering perspective. Biochem Biophys Res Commun 473:698–703. https://doi.org/10.1016/j.bbrc.2015.11.060

Foulquier S, Daskalopoulos EP, Lluri G et al (2018) WNT signaling in cardiac and vascular disease. Pharmacol Rev 70:68–141. https://doi.org/10.1124/pr.117.013896

Berger I, Hershkovitz E, Shaag A et al (2008) Mitochondrial complex I deficiency caused by a deleterious NDUFA11 mutation. Ann Neurol 63:405–408. https://doi.org/10.1002/ana.21332

Jang S, Javadov S (2018) Elucidating the contribution of ETC complexes I and II to the respirasome formation in cardiac mitochondria. Sci Rep 8:17732. https://doi.org/10.1038/s41598-018-36040-9

Wiegreffe C, Simon R, Peschkes K et al (2015) Bcl11a (Ctip1) controls migration of cortical projection neurons through regulation of Sema3c. Neuron 87:311–325. https://doi.org/10.1016/j.neuron.2015.06.023

Dias C, Estruch SB, Graham SA et al (2016) BCL11A haploinsufficiency causes an intellectual disability syndrome and dysregulates transcription. Am J Hum Genet 99:253–274. https://doi.org/10.1016/j.ajhg.2016.05.030

Saiki Y, Yamazaki Y, Yoshida M et al (2000) Human EVI9, a homologue of the mouse myeloid leukemia gene, is expressed in the hematopoietic progenitors and down-regulated during myeloid differentiation of HL60 cells. Genomics 70:387–391. https://doi.org/10.1006/geno.2000.6385

Koskela J, Laiho J, Kähönen M et al (2008) Potassium channel KCNH2 K897T polymorphism and cardiac repolarization during exercise test: the Finnish Cardiovascular study. Scand J Clin Lab Invest 68:31–38. https://doi.org/10.1080/00365510701496488

Dai G, Zagotta WN (2017) Molecular mechanism of voltage-dependent potentiation of KCNH potassium channels. Elife 6:e26355. https://doi.org/10.7554/eLife.26355

Li X, Martinson AS, Layden MJ et al (2015) Ether-à-go-go family voltage-gated K+ channels evolved in an ancestral metazoan and functionally diversified in a cnidarian-bilaterian ancestor. J Exp Biol 218:526–536. https://doi.org/10.1242/jeb.110080

Kupershmidt S, Yang IC-H, Hayashi K et al (2003) The IKr drug response is modulated by KCR1 in transfected cardiac and noncardiac cell lines. FASEB J 17:2263–2265. https://doi.org/10.1096/fj.02-1057fje

Gilman AG (1987) G Proteins: transducers of receptor-generated signals. Annu Rev Biochem 56:615–649. https://doi.org/10.1146/annurev.bi.56.070187.003151

Petersen CI, McFarland TR, Stepanovic SZ et al (2004) In vivo identification of genes that modify ether-a-go-go-related gene activity in Caenorhabditis elegans may also affect human cardiac arrhythmia. Proc Natl Acad Sci 101:11773–11778. https://doi.org/10.1073/pnas.0306005101

Contractor A, Mulle C, Swanson GT (2011) Kainate receptors coming of age: milestones of two decades of research. Trends Neurosci 34:154–163. https://doi.org/10.1016/J.TINS.2010.12.002

Pischedda F, Szczurkowska J, Cirnaru MD et al (2014) A cell surface biotinylation assay to reveal membrane-associated neuronal cues: Negr1 regulates dendritic arborization. Mol Cell Proteomics 13:733–748. https://doi.org/10.1074/mcp.M113.031716

Wilmore JH, Stanforth PR, Gagnon J et al (2001) Heart rate and blood pressure changes with endurance training: the HERITAGE family study. Med Sci Sports Exerc 33:107–116

Rankinen T, Argyropoulos G, Rice T et al (2010) CREB1 is a strong genetic predictor of the variation in exercise heart rate response to regular exercise: the HERITAGE family study. Circ Cardiovasc Genet 3:294–299. https://doi.org/10.1161/CIRCGENETICS.109.925644

Rankinen T, Sung YJ, Sarzynski MA et al (2012) Heritability of submaximal exercise heart rate response to exercise training is accounted for by nine SNPs. J Appl Physiol 112:892–897. https://doi.org/10.1152/japplphysiol.01287.2011

Malaspina A, Kaushik N, de Belleroche J (2000) A 14-3-3 mRNA is up-regulated in amyotrophic lateral sclerosis spinal cord. J Neurochem 75:2511–2520

Lau JMC, Jin X, Ren J et al (2007) The 14-3-3tau phosphoserine-binding protein is required for cardiomyocyte survival. Mol Cell Biol 27:1455–1466. https://doi.org/10.1128/MCB.01369-06

Patberg KW, Shvilkin A, Plotnikov AN et al (2005) Cardiac memory: mechanisms and clinical implications. Hear Rhythm 2:1376–1382. https://doi.org/10.1016/j.hrthm.2005.08.021

Wu H, Zhou Y, Xiong Z-Q (2007) Transducer of regulated CREB and late phase long-term synaptic potentiation. FEBS J 274:3218–3223. https://doi.org/10.1111/j.1742-4658.2007.05891.x

Pickering C, Kiely J (2017) Exercise genetics: seeking clarity from noise. BMJ Open Sport Exerc Med 3:e000309. https://doi.org/10.1136/bmjsem-2017-000309

Karoly HC, Stevens CJ, Magnan RE et al (2012) Genetic influences on physiological and subjective responses to an aerobic exercise session among sedentary adults. J Cancer Epidemiol 2012:1–12. https://doi.org/10.1155/2012/540563

Wangler MF, Hu Y, Shulman JM (2017) Drosophila and genome-wide association studies: a review and resource for the functional dissection of human complex traits. Dis Model Mech 10:77–88. https://doi.org/10.1242/dmm.027680

Hamel V, Cheng K, Liao S et al (2017) De Novo human cardiac myocytes for medical research: promises and challenges. Stem Cells Int 2017:1–7. https://doi.org/10.1155/2017/4528941

Chen IY, Matsa E, Wu JC (2016) Induced pluripotent stem cells: at the heart of cardiovascular precision medicine. Nat Rev Cardiol 13:333–349. https://doi.org/10.1038/nrcardio.2016.36

Smith SC, Benjamin EJ, Bonow RO et al (2011) AHA/ACCF secondary prevention and risk reduction therapy for patients with coronary and other atherosclerotic vascular disease: 2011 update. J Am Coll Cardiol. https://doi.org/10.1161/CIR.0b013e318235eb4d

Ponikowski P, Voors AA, Anker SD et al (2016) 2016 ESC guidelines for the diagnosis and treatment of acute and chronic heart failure. Eur Heart J 37:2129–2200. https://doi.org/10.1093/eurheartj/ehw128

Swedberg K, Komajda M, Böhm M et al (2010) Ivabradine and outcomes in chronic heart failure (SHIFT): a randomised placebo-controlled study. Lancet 376:875–885. https://doi.org/10.1016/S0140-6736(10)61198-1

Kumamaru H, Kadoya K, Adler AF et al (2018) Generation and post-injury integration of human spinal cord neural stem cells. Nat Methods 15:723–731. https://doi.org/10.1038/s41592-018-0074-3

Mouse Genome Sequencing Consortium, Waterston RH, Lindblad-Toh K et al (2002) Initial sequencing and comparative analysis of the mouse genome. Nature 420:520–562. https://doi.org/10.1038/nature01262

van der Harst P, van Setten J, Verweij N et al (2016) 52 genetic loci influencing myocardial mass. J Am Coll Cardiol 68:1435–1448. https://doi.org/10.1016/j.jacc.2016.07.729

Bier E (2005) Drosophila, the golden bug, emerges as a tool for human genetics. Nat Rev Genet 6:9–23. https://doi.org/10.1038/nrg1503

Howe K, Clark MD, Torroja CF et al (2013) The zebrafish reference genome sequence and its relationship to the human genome. Nature 496:498–503. https://doi.org/10.1038/nature12111

Khera AV, Chaffin M, Aragam KG et al (2018) Genome-wide polygenic scores for common diseases identify individuals with risk equivalent to monogenic mutations. Nat Genet 50:1219–1224. https://doi.org/10.1038/s41588-018-0183-z

Said MA, Verweij N, van der Harst P (2018) Associations of combined genetic and lifestyle risks with incident cardiovascular disease and diabetes in the UK biobank study. JAMA Cardiol 3:693. https://doi.org/10.1001/jamacardio.2018.1717

Kraft P, Zeggini E, Ioannidis JPA (2009) Replication in genome-wide association studies. Stat Sci 1:1. https://doi.org/10.1214/09-STS290

Download references

N. Verweij is supported by “Nederlandse Organisatie voor Wetenschappelijk Onderzoek” VENI grant (016.186.125) in support of research into ECG changes in response to exercise.

Author information

Yordi J. van de Vegte and Balewgizie S. Tegegne have contributed equally to this work.

Authors and Affiliations

Department of Cardiology, University of Groningen, University Medical Center Groningen, Hanzeplein 1, 9700 RB, Groningen, The Netherlands

Yordi J. van de Vegte, Niek Verweij & Pim van der Harst

Department of Genetics, University of Groningen, University Medical Center Groningen, 9700 RB, Groningen, The Netherlands

Pim van der Harst

Department of Epidemiology, University of Groningen, University Medical Center Groningen, 9700 RB, Groningen, The Netherlands

Balewgizie S. Tegegne & Harold Snieder

Durrer Center for Cardiogenetic Research, Netherlands Heart Institute, 3511 GC, Utrecht, The Netherlands

You can also search for this author in PubMed   Google Scholar

Corresponding author

Correspondence to Pim van der Harst .

Ethics declarations

Conflict of interest.

The authors declare that they have no conflict of interest. Author N. Verweij is an employee of Genomics plc.

Additional information

Publisher's note.

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Electronic supplementary material

Below is the link to the electronic supplementary material.

Supplementary material 1 (XLSX 143 kb)

Rights and permissions.

Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0 International License ( http://creativecommons.org/licenses/by/4.0/ ), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made.

Reprints and permissions

About this article

van de Vegte, Y.J., Tegegne, B.S., Verweij, N. et al. Genetics and the heart rate response to exercise. Cell. Mol. Life Sci. 76 , 2391–2409 (2019). https://doi.org/10.1007/s00018-019-03079-4

Download citation

Received : 11 October 2018

Accepted : 18 March 2019

Published : 27 March 2019

Issue Date : 01 June 2019

DOI : https://doi.org/10.1007/s00018-019-03079-4

Share this article

Anyone you share the following link with will be able to read this content:

Sorry, a shareable link is not currently available for this article.

Provided by the Springer Nature SharedIt content-sharing initiative

• Find a journal
• Publish with us
• Track your research

An official website of the United States government

The .gov means it’s official. Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

The site is secure. The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

• Publications
• Account settings
• My Bibliography
• Collections
• Citation manager

Save citation to file

Email citation, add to collections.

• Create a new collection
• Add to an existing collection

Add to My Bibliography

Your saved search, create a file for external citation management software, your rss feed.

• Search in PubMed
• Search in NLM Catalog
• Add to Search

Exercise conditioning and heart rate variability: evidence of a threshold effect

Affiliation.

• 1 Department of Medicine, Cedars-Sinai Research Institute, Cedars-Sinai Medical Center, Los Angeles, California, USA.
• PMID: 10941549
• PMCID: PMC6654924
• DOI: 10.1002/clc.4960230813

Background: A protective effect of exercise in preventing sudden cardiac death is supported by studies in healthy populations as well as in patients with cardiac disease. The mechanisms involved in this protective effect are unknown.

Hypothesis: We hypothesized that exercise conditioning would beneficially alter autonomic nervous system tone, measured by heart rate variability.

Methods: We prospectively studied 20 cardiac patients enrolled in a Phase 2 12-week cardiac rehabilitation program following a recent cardiac event. The patients underwent 24 h Holter monitoring at program entry and 12 weeks later. Heart rate variability analysis was assessed for both time domain and spectral analysis.

Results: The group demonstrated a modest mean conditioning effect, indicated by an average reduction in resting heart rate from 81 +/- 16 to 75 +/- 12 beats/min (p = 0.03), and an increase in training METS from 2.1 +/- 0.4 to 3.3 +/- 1.1 (p < 0.0001). Overall, 15 of 20 (75%) patients demonstrated increased total and high-frequency power, and mean high-frequency power was significantly increased (3.9 +/- 1.4 vs. 4.4 +/- 1.0 ln, p = 0.05). When stratified according to the magnitude of exercise conditioning, patients achieving an increase of > 1.5 training METS demonstrated significant increases in SDNN, SDANN index, SDNN index, pNN50, total power, and high-frequency power (all p < 0.05) (see text for explanation of abbreviations).

Conclusions: Exercise conditioning improves heart rate variability in cardiac patients, particularly in patients who achieve a threshold of > 1.5 training METS increase over a 12-week period. These study results are supportive of the concept that exercise training lowers the risk of sudden cardiac death via increased vagal tone, which likely beneficially alters ventricular fibrillatory and ischemic thresholds.

PubMed Disclaimer

Similar articles

• Impact of Menstrual Cycle on Cardiac Autonomic Function Assessed by Heart Rate Variability and Heart Rate Recovery. Yazar Ş, Yazıcı M. Yazar Ş, et al. Med Princ Pract. 2016;25(4):374-7. doi: 10.1159/000444322. Epub 2016 Feb 1. Med Princ Pract. 2016. PMID: 26828607 Free PMC article.
• [The effect of cardiovascular rehabilitation on the variability of the RR cycle after a first uncomplicated acute myocardial infarct]. Carunchio A, Fera MS, Bordi L, Daniele R, Rulli F, Coletta C, Burattini M, Greco G, Martinelli MM, Porzio A, Lumia F, Ceci V. Carunchio A, et al. Ital Heart J Suppl. 2000 Feb;1(2):241-9. Ital Heart J Suppl. 2000. PMID: 10731382 Clinical Trial. Italian.
• High-intensity interval training and cardiac autonomic control in individuals with metabolic syndrome: A randomised trial. Ramos JS, Dalleck LC, Borrani F, Beetham KS, Mielke GI, Dias KA, Wallen MP, Keating SE, Fassett RG, Coombes JS. Ramos JS, et al. Int J Cardiol. 2017 Oct 15;245:245-252. doi: 10.1016/j.ijcard.2017.07.063. Epub 2017 Jul 21. Int J Cardiol. 2017. PMID: 28747269 Clinical Trial.
• Monitoring Athletic Training Status Through Autonomic Heart Rate Regulation: A Systematic Review and Meta-Analysis. Bellenger CR, Fuller JT, Thomson RL, Davison K, Robertson EY, Buckley JD. Bellenger CR, et al. Sports Med. 2016 Oct;46(10):1461-86. doi: 10.1007/s40279-016-0484-2. Sports Med. 2016. PMID: 26888648 Review.
• [Heart rate variability and physical exercise. Current status]. Hottenrott K, Hoos O, Esperer HD. Hottenrott K, et al. Herz. 2006 Sep;31(6):544-52. doi: 10.1007/s00059-006-2855-1. Herz. 2006. PMID: 17036185 Review. German.
• Exercise training and cardiac autonomic function following coronary artery bypass grafting: a systematic review and meta-analysis. Kushwaha P, Moiz JA, Mujaddadi A. Kushwaha P, et al. Egypt Heart J. 2022 Sep 23;74(1):67. doi: 10.1186/s43044-022-00306-5. Egypt Heart J. 2022. PMID: 36138168 Free PMC article. Review.
• Changes in the Complexity of Heart Rate Variability with Exercise Training Measured by Multiscale Entropy-Based Measurements. Fazan FS, Brognara F, Fazan Junior R, Murta Junior LO, Virgilio Silva LE. Fazan FS, et al. Entropy (Basel). 2018 Jan 17;20(1):47. doi: 10.3390/e20010047. Entropy (Basel). 2018. PMID: 33265153 Free PMC article.
• Effects of 6 Months of Exercise-Based Cardiac Rehabilitation on Autonomic Function and Neuro-Cardiovascular Stress Reactivity in Coronary Artery Disease Patients. Badrov MB, Wood KN, Lalande S, Sawicki CP, Borrell LJ, Barron CC, Vording JL, Fleischhauer A, Suskin N, McGowan CL, Shoemaker JK. Badrov MB, et al. J Am Heart Assoc. 2019 Sep 3;8(17):e012257. doi: 10.1161/JAHA.119.012257. Epub 2019 Aug 23. J Am Heart Assoc. 2019. PMID: 31438760 Free PMC article.
• Can strenuous exercise harm the heart? Insights from a study of cardiovascular neural regulation in amateur triathletes. Dalla Vecchia LA, Barbic F, De Maria B, Cozzolino D, Gatti R, Dipaola F, Brunetta E, Zamuner AR, Porta A, Furlan R. Dalla Vecchia LA, et al. PLoS One. 2019 May 7;14(5):e0216567. doi: 10.1371/journal.pone.0216567. eCollection 2019. PLoS One. 2019. PMID: 31063482 Free PMC article.
• Hidden Signals-The History and Methods of Heart Rate Variability. Ernst G. Ernst G. Front Public Health. 2017 Oct 16;5:265. doi: 10.3389/fpubh.2017.00265. eCollection 2017. Front Public Health. 2017. PMID: 29085816 Free PMC article. Review.
• Blair SN, Kohl HN III, Paffenbarger RS Jr, Clark DG, Cooper KH, Gibbons LW: Physical fitness and all‐cause mortality: A prospective study of healthy men and women. J Am Med Assoc 1989; 262: 2395–2401 - PubMed
• Paffenbarger RS Jr, Hyde RT, Wing AL, Lee IM, Jung DL, Kampert JB: The association of changes in physical‐activity level and other life‐style characteristics with mortality among men. N Engl J Med 1993; 328: 538–545 - PubMed
• O'Connor GT, Buring JE, Yusuf S, Goldhaber SZ, Olmstead EM, Paffenbarger RS Jr, Hennekens CH: An overview of randomized trials of rehabilitation with exercise after myocardial infarction. Circulation 1989; 80: 234–244 - PubMed
• Oldridge NB, Guyatt GH, Fisher MS, Rimm AA: Cardiac rehabilitation after myocardial infarction: Combined experience of randomized trials. J Am Med Assoc 1988; 260: 945–950 - PubMed
• Pagani M, Lambardi F, Guzzetti S, Rimoldi O, Furlan R, Pizzinelli P, Sandrone G, Malfatto G, Dell'Orto S, Piccaluga E, Turiel M, Baselli G, Cerutti S, Malliani A: Power spectral analysis of heart rate and arterial pressure as a marker of sympatho‐vagal interaction in man and conscious dog. Circ Res 1986; 59: 178–193 - PubMed

Publication types

• Search in MeSH

Log in using your username and password

• Search More Search for this keyword Advanced search
• Latest content
• Current issue
• Browse by collection
• BMJ Journals

You are here

• Volume 94, Issue 5
• The heart rate hypothesis: ready to be tested
• Article Text
• Article info
• Citation Tools
• Rapid Responses
• Article metrics

• A S Hall 1 ,
• 1 British Heart Foundation Heart Research Centre at Leeds, University of Leeds, Leeds, UK
• 2 Centre for Health Economics, University of York, York, UK
• Professor of Clinical Cardiology, Alistair S Hall, G Floor, Jubilee Wing, Leeds General Infirmary, Leeds, LS2 3EX, UK; A.S.Hall{at}leeds.ac.uk

There is increasing evidence that increased heart rate may be an independent risk factor for cardiovascular morbidity and mortality both in patients with ischaemic heart disease and in the general population. Elevated heart rate in coronary artery disease is a major determinant of oxygen consumption and appears to evoke most episodes of ischaemia. Increased resting heart rate may also contribute to the development of atherosclerosis, facilitate plaque destabilisation and initiate arrhythmias, leading to acute coronary events and sudden death. Reducing heart rate is a central aim in the treatment of stable angina pectoris; this therapeutic approach may have an essential role in lowering the incidence of cardiovascular morbidity and mortality in patients with pre-existing ischaemic heart disease. However, this heart rate hypothesis has not thus far been proven. Evidence suggests that the use of heart rate-lowering drugs may have a beneficial effect; however, most treatments for angina have additional negative inotropic effects on the heart. This hypothesis can now be tested following the recent development of selective heart rate drugs.

• If channel inhibitors
• angina pectoris
• heart rate hypothesis

https://doi.org/10.1136/hrt.2007.118760

Statistics from Altmetric.com

Request permissions.

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

Competing interests: Professor Hall has received research grants from Astra-Zeneca, Servier UK, and Sanofi-Aventis UK; has received honoraria from Astra-Zeneca and Servier UK; and has been paid consultant fees by Servier UK.

Read the full text or download the PDF:

• Skip to primary navigation
• Skip to main content
• Skip to primary sidebar

• FREE Experiments
• Kitchen Science
• Climate Change
• Egg Experiments
• Fairy Tale Science
• Edible Science
• Human Health
• Inspirational Women
• Forces and Motion
• Science Fair Projects
• STEM Challenges
• Science Sparks Books
• Contact Science Sparks
• Science Resources for Home and School

How does exercise affect heart rate? Science Investigation

June 30, 2016 By Emma Vanstone 5 Comments

This investigation to find out how exercise affects heart rate is a great way to introduce correct scientific procedures and think about variables that can change and those that need to remain constant.

Exercise increases the rate at which energy is needed from food, increasing the body’s need for both nutrients and oxygen. This is why both pulse/heart rate and breathing rate increase when we exercise.

Pulse rate is an indication of heart rate, as the arteries expand each time the ventricles pump blood out of the heart.

The heart pumps extra food and oxygen to the muscles while breathing speeds up to get more oxygen into the body and remove carbon dioxide.

How does exercise affect heart rate investigation

Use the stethoscopes and timers to record how many heartbeats you can hear in 30 seconds.

Exercise – this could be 30 seconds of star jumps or a mini obstacle course.  

Step 3  

Use the timers and stethoscopes again to record how many heartbeats you can hear in 30 seconds.

Use my handy heart rate and exercise investigation results table to record your results or design your own!

Let’s think scientifically

A scientific investigation should be a fair test, think about what conditions you need to keep the same and what condition you will change. You should also repeat the testing three times and find the average heart rate.

Things to keep the same:

Heartbeats must be counted before and after exercising for the same amount of time .

The person whose heart rate is compared must be the same.

Things to change:  

Heart rate should be measured before and after exercise.

Make a prediction

What effect do you think exercise will have on heart rate?

Why do you think this?

Clue: When you exercise, your muscles need more food and oxygen from your blood, so your heart has to beat faster to transport them.

What is recovery time?  

Recovery time is the time it takes for the heart rate to return to normal. If you have time, can you work out how long this will take you?

During exercise, the pulse rate and breathing rate of a fitter person rise much less than in an unfit person. Fitter people also have a shorter recovery time.

Links to Maths

Design a method of recording your results. Can you work out the average heart rate for 10 participants before and after exercise?

Calculate the difference between a person’s heart rate before and after exercise.

Links to English

Can you write a letter to a friend telling them about your findings?

More Science for Kids

Find out how to  make your own stethoscope with a funnel, tape and cardboard tube.

Make a pumping model of a heart , or try one of our sports science investigations.

Suitable for:

Key Stage 1 Science: Animals, including Humans

Describe the importance for humans of exercise, eating the right amounts of different types of food, and hygiene.

Key Stage 2 Science: Animals, including Humans

Recognise the impact of diet, exercise, drugs and lifestyle on the way their bodies function

Last Updated on June 25, 2024 by Emma Vanstone

Safety Notice

Science Sparks ( Wild Sparks Enterprises Ltd ) are not liable for the actions of activity of any person who uses the information in this resource or in any of the suggested further resources. Science Sparks assume no liability with regard to injuries or damage to property that may occur as a result of using the information and carrying out the practical activities contained in this resource or in any of the suggested further resources.

These activities are designed to be carried out by children working with a parent, guardian or other appropriate adult. The adult involved is fully responsible for ensuring that the activities are carried out safely.

Reader Interactions

March 05, 2018 at 8:06 am

hi my name is tyrone

January 22, 2019 at 2:33 pm

helpful thanks!

February 23, 2019 at 7:51 am

This was extremely helpful for a school task, thx

April 02, 2019 at 3:37 pm

It is for my science fair

September 29, 2019 at 10:29 am

Helps a lot for science project.

Leave a Reply Cancel reply

Your email address will not be published. Required fields are marked *

The Biology Corner

Biology Teaching Resources

Investigation: How Does Exercise Affect Heart Rate

 This activity was designed for anatomy and physiology students who are studying how feedback loops work and also serves as a refresher on the scientific method and designing experiments (science and engineering practices).   

Students work in groups to develop a hypothesis about how exercise affects the heart rate.  This is fairly obvious for most students in my age group, but the next question asks them to determine the recovery time and compare how recovery times differ for individuals.

The group is responsible for designing the experiment, so they must determine how many individuals they will test, what type of exercise will be done and how they will measure heart rate, though a diagram is provided to give them a clue to measure using pulse rate.   

During this phase of the activity, I circulate among groups and help them with the details of their plan.

The final analysis  uses a CER format where students make a CLAIM that answers the experimental questions, and then provide a summary of their evidence.  Finally students discuss reasons for their observations by linking the results to their understanding of the physiology of the heart.

I often do this activity at the beginning of the year, but it could also be included in the chapter on the circulatory system.   It’s also possible, if you are short on time, to do an abbreviated version of this activity where you give students more direct instruction on how to take their pulse and have them do some light activity and just compare their own rate increase and recovery time.

Grade Level:  9-12 Time Required:  30 -45 minutes

NGSS Science and Engineering Practices :   1. Asking questions (for science)  2. Developing and using models 3. Planning and carrying out investigations 4. Analyzing and interpreting data

HS-LS1-3 Plan and conduct an investigation to provide evidence that feedback mechanisms maintain homeostasis.

Shannan Muskopf

Several factors can make your heart beat faster than normal while exercising. These include being out of shape, working out in hot weather, or pushing yourself too hard. It's important to pay attention to how you feel during exercise. If you're dizzy, short of breath, or have chest pain along with a very high heart rate, stop exercising right away.

Keeping track of your heart rate during workouts helps you  exercise safely and effectively . You can check your pulse manually or use a heart rate monitor. By staying within the right heart rate zone, you'll get the most benefit from your workouts while avoiding potential risks.

Key Takeaways

• Exercise heart rate  should stay between 50-85% of maximum
• Very high heart rates during workouts may signal overexertion
• Monitoring pulse helps ensure safe and effective exercise

Understanding Heart Rate and Exercise

Heart rate plays a key role in exercise. It changes based on how hard you work out and affects your fitness gains.

Fundamentals of Heart Rate

Heart rate is the number of times your heart beats per minute (bpm). A normal resting heart rate is between 60-100 bpm. Athletes may have lower resting rates, sometimes as low as 40 bpm.

Your maximum heart rate is the highest your heart can safely beat during activity. To estimate it, subtract your age from 220.

Maximum heart rate = 220 - age

During exercise, your heart rate increases to supply more oxygen to your muscles. How high it goes depends on how hard you're working.

Role of Heart Rate in Exercise

Heart rate helps measure exercise intensity. The harder you work, the faster your heart beats.

Target heart rate zones guide effective workouts:

• Moderate intensity: 50-70% of max heart rate
• High intensity: 70-85% of max heart rate

Staying in these zones helps improve fitness safely.  Regular aerobic exercise  can lower your resting heart rate over time, showing improved heart health.

Monitoring heart rate during workouts helps you:

• Gauge effort level
• Avoid overexertion
• Track fitness progress

Factors Influencing Heart Rate Response

Many things affect how your  heart rate changes  during exercise. Your body and outside factors both play a role in how fast your heart beats when you work out.

Individual Health and Fitness Levels

Age and fitness level  greatly impact heart rate response to exercise. Younger people often have lower resting heart rates and can reach  higher peaks during workouts . As we age, maximum heart rates tend to decrease.

Athletes and those with high fitness levels typically have lower resting heart rates. Their hearts pump more blood per beat. This means they can do more work at lower heart rates compared to less fit individuals.

Genetics also play a role. Some people naturally have higher or lower heart rates. Health risks like high blood pressure, diabetes, and high cholesterol can affect how the  heart responds to exercise .

Impact of External Factors

Environmental conditions  can change heart rate during exercise. Heat and high humidity make the heart work harder to cool the body. This leads to faster heart rates at the same exercise intensity.

Stress increases heart rate  even before exercise starts. Caffeine can boost heart rate and may make it higher during workouts.

Smoking harms heart health and can lead to higher exercise heart rates. Smokers often reach their maximum heart rate at lower exercise levels than non-smokers.

Altitude also matters. At higher elevations, the heart beats faster to get more oxygen to the body.

Risks of Elevated Heart Rate During Exercise

Pushing your heart rate too high during exercise can lead to both immediate dangers and potential long-term health issues. These risks range from mild discomfort to serious cardiac events.

Short-Term Dangers

Exercising at very high heart rates can cause several immediate symptoms. Dizziness and lightheadedness may occur as the body struggles to maintain blood flow to the brain.

Chest pain or tightness is another warning sign. This could indicate reduced oxygen supply to the heart muscle.

Palpitations or a racing heartbeat might be felt. These sensations can be frightening and may signal an irregular heart rhythm.

Shortness of breath often accompanies an elevated heart rate. The body may not be able to get enough oxygen to meet its increased demands.

In severe cases, people may experience nausea or even fainting. These symptoms require immediate attention and rest.

Long-Term Health Implications

Regularly exceeding safe heart rate limits during exercise can have lasting effects. High-intensity workouts may increase the risk of heart problems in some individuals.

Chronic overexertion can lead to changes in heart structure and function. This may increase the likelihood of arrhythmias like atrial fibrillation.

Repeated stress on the heart can contribute to the development of  hypertension . Over time, this raises the risk of  heart disease  and stroke.

In rare cases, extreme exercise and sustained high heart rates might trigger a heart attack. This is more likely in people with underlying heart conditions.

Proper exercise intensity and monitoring are key to avoiding these risks. Gradual progression and expert guidance can help ensure safe, effective workouts.

Monitoring and Measuring Exercise Heart Rate

Keeping track of your heart rate during exercise is key for safe and effective workouts. There are several ways to measure and assess your heart rate while exercising.

Using Heart Rate Monitors

Heart rate monitors are handy tools for tracking your pulse during workouts. Many fitness trackers and smartwatches have built-in heart rate sensors. These devices use light to detect blood flow in your wrist.

Chest strap monitors are another option. They wrap around your chest and pick up electrical signals from your heart. Chest straps tend to be more accurate than wrist-based monitors.

Most heart rate monitors show your current pulse and can alert you when you hit certain heart rate zones. Some also track your heart rate over time and link to smartphone apps for more detailed analysis.

Recognizing Heart Rate Zones

Heart rate zones help gauge exercise intensity. They're usually shown as a percentage of your  maximum heart rate .

• 50-60%: Light activity
• 60-70%: Moderate activity
• 70-80%: Hard activity
• 80-90%: Very hard activity
• 90-100%: Maximum effort

Knowing your zones helps you exercise at the right intensity for your goals. For example, moderate activity is good for building endurance, while hard activity  improves cardiovascular fitness .

Assessing Heart Rate without Gadgets

You can check your pulse manually if you don't have a heart rate monitor. Here's how:

• Find your pulse on your wrist or neck
• Count your heartbeats for 15 seconds
• Multiply that number by 4 to get beats per minute

The "talk test" is another way to gauge intensity. If you can talk easily, you're in a light to moderate zone. If you can only say a few words at a time, you're working hard.

Rate of perceived exertion  is also useful. On a scale of 1-10, light activity feels like a 2-3, moderate is 4-5, and hard is 6-7.

Managing High Heart Rate During Workouts

A high heart rate during exercise can be managed through several strategies. These include adjusting workout intensity, adding rest periods, and seeking medical advice when needed.

Adjusting Exercise Intensity

When your heart rate climbs too high, lowering the intensity of your workout can help. Try slowing down your pace or reducing the weight you're lifting. This allows your body to catch up and your heart rate to drop.

For cardio exercises, you can switch to a lower-impact activity. For example, change from running to walking or cycling. In strength training, use lighter weights or do fewer repetitions.

It's important to stay within your target heart rate zone during workouts. This zone is usually 50-85% of your maximum heart rate, depending on your fitness level and goals.

Incorporating Recovery Periods

Adding rest periods to your exercise routine can prevent your heart rate from getting too high. Take short breaks between sets or intervals to let your heart rate come down.

Try interval training. This involves alternating between high and low-intensity exercises. For example, sprint for 30 seconds, then walk for 1 minute.

Active recovery is another useful technique. Instead of stopping completely, continue moving at a very low intensity. This keeps blood flowing and helps your heart rate decrease gradually.

Consultation with Healthcare Professionals

If you often experience a very high heart rate during exercise, it's wise to talk to a doctor. They can check for any underlying medical conditions that might be causing this issue.

A healthcare professional can also help you create a safe exercise plan. This is especially important if you have heart problems or take medications that affect your heart rate.

They might recommend tests like an exercise stress test. This can show how your heart responds to physical activity. Based on the results, your doctor can suggest safe heart rate limits for your workouts.

Remember, some medications can affect your heart rate during exercise. Always inform your doctor about any medicines you're taking.

Improving Cardiovascular Health and Exercise Safety

Regular exercise benefits  heart health and lowers risks. Safe workout practices are key to avoiding issues from high heart rates. Knowing when to get medical help is also important.

Effective Exercise Strategies

Brisk walking  is a great low-impact way to improve heart health. It's easy on joints and suitable for most people.  Running  and swimming are also excellent cardio workouts.

For beginners, start slow and build up. Aim for 30 minutes of moderate activity 5 days a week. This can include:

• Brisk walking
• Light jogging

Gradually increase intensity as fitness improves.  High-intensity interval training  can boost heart health in less time. It involves short bursts of vigorous activity followed by rest periods.

Always warm up before exercise and cool down after. This helps prevent sudden changes in heart rate.

When to Seek Medical Attention

It's crucial to know the signs of overexertion. Stop exercising and seek help if feeling:

• Chest pain or pressure
• Extreme shortness of breath
• Dizziness or lightheadedness
• Irregular heartbeat

People with heart conditions should talk to a doctor before starting a new exercise program. A  stress test  or EKG can check heart health.

The  American Heart Association  advises adults to know their target heart rate. Exercising within this range helps maximize benefits while staying safe.

If unsure about exercise safety, consult a healthcare provider. They can offer personalized advice based on individual health status.

Frequently Asked Questions

High heart rates during exercise can pose risks and need careful management. Knowing when to slow down or stop is crucial for safety. Proper monitoring and adjustment techniques help maintain an ideal range.

What are the risks of exercising with a heart rate significantly above my target zone?

Exercising with a heart rate too far above the target zone can be dangerous. It may lead to dizziness, chest pain, or difficulty breathing. In extreme cases, it could cause heart damage or arrhythmias.

Pushing too hard also increases fatigue and injury risk. The body needs time to adapt to higher intensities safely.

How can I tell if my heart rate is dangerously high during a workout?

Signs of an excessively high heart rate include shortness of breath, chest discomfort, and lightheadedness. Feeling unusually tired or unable to speak in full sentences are also warning signs.

Using a heart rate monitor can provide precise data. If the rate exceeds 85% of the calculated maximum, it's time to ease up.

At what heart rate should I stop exercising to avoid health complications?

The exact number varies by age and fitness level. Generally, exercising above 85% of maximum heart rate is considered very high intensity.

For most adults, staying below 160-170 beats per minute is advisable. Those with health conditions may need lower limits set by a doctor.

Can a high heart rate during exercise affect my long-term cardiovascular health?

Consistently exercising at very high heart rates can strain the cardiovascular system. It may lead to increased risk of heart problems over time.

However, brief periods of high-intensity exercise can be beneficial when done safely. The key is balance and gradual progression in intensity.

What should I do if I experience a sudden spike in heart rate while exercising?

If heart rate spikes suddenly, slow down or stop the activity. Take deep breaths and find a cool, shaded area if outdoors.

Drink water and rest until the heart rate returns to a normal range. If symptoms persist or worsen, seek medical attention promptly.

How can I manage my heart rate during exercise to stay within a safe range?

Use a heart rate monitor to track intensity. Start workouts slowly and gradually increase effort.

Aim for 50-70% of maximum heart rate for moderate exercise. For vigorous workouts, stay between 70-85%. Take regular breaks and stay hydrated to help control heart rate.

Related posts:

• What is a Dangerous Heart Rate? Risks and Thresholds
• Etekcity FSA HSA Smart Scale Review: Worth Your Health Dollars?

Leave a Comment

Your email address will not be published. Required fields are marked

An official website of the United States government

The .gov means it’s official. Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

The site is secure. The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

• Publications
• Account settings

Preview improvements coming to the PMC website in October 2024. Learn More or Try it out now .

• Advanced Search
• Journal List
• Front Physiol

Cardiac Autonomic Responses during Exercise and Post-exercise Recovery Using Heart Rate Variability and Systolic Time Intervals—A Review

Scott michael.

1 Discipline of Exercise and Sports Science, Faculty of Health Sciences, University of Sydney, Sydney, NSW, Australia

Kenneth S. Graham

2 New South Wales Institute of Sport, Sydney, NSW, Australia

Glen M. Davis, OAM

Cardiac parasympathetic activity may be non-invasively investigated using heart rate variability (HRV), although HRV is not widely accepted to reflect sympathetic activity. Instead, cardiac sympathetic activity may be investigated using systolic time intervals (STI), such as the pre-ejection period. Although these autonomic indices are typically measured during rest, the “reactivity hypothesis” suggests that investigating responses to a stressor (e.g., exercise) may be a valuable monitoring approach in clinical and high-performance settings. However, when interpreting these indices it is important to consider how the exercise dose itself (i.e., intensity, duration, and modality) may influence the response. Therefore, the purpose of this investigation was to review the literature regarding how the exercise dosage influences these autonomic indices during exercise and acute post-exercise recovery. There are substantial methodological variations throughout the literature regarding HRV responses to exercise, in terms of exercise protocols and HRV analysis techniques. Exercise intensity is the primary factor influencing HRV, with a greater intensity eliciting a lower HRV during exercise up to moderate-high intensity, with minimal change observed as intensity is increased further. Post-exercise, a greater preceding intensity is associated with a slower HRV recovery, although the dose-response remains unclear. A longer exercise duration has been reported to elicit a lower HRV only during low-moderate intensity and when accompanied by cardiovascular drift, while a small number of studies have reported conflicting results regarding whether a longer duration delays HRV recovery. “Modality” has been defined multiple ways, with limited evidence suggesting exercise of a greater muscle mass and/or energy expenditure may delay HRV recovery. STI responses during exercise and recovery have seldom been reported, although limited data suggests that intensity is a key determining factor. Concurrent monitoring of HRV and STI may be a valuable non-invasive approach to investigate autonomic stress reactivity; however, this integrative approach has not yet been applied with regards to exercise stressors.

Introduction

Quantifying the fluctuations in R-wave to R-wave intervals (RRI), referred to as heart rate variability (HRV), has been considered a useful method by which to monitor autonomic activity, in particular cardiac parasympathetic modulation (Camm et al., 1996 ). Monitoring HRV responses to an “exercise challenge test” may provide useful insight into autonomic stress reactivity. This is consistent with the “reactivity hypothesis” (Krantz and Manuck, 1984 ; Heponiemi et al., 2007 ), which proposes that cardiovascular responses to a stressor may be predictive of certain diseases (Treiber et al., 2003 ; Lovallo, 2005 ; Phillips, 2011 ), as well as useful in monitoring the training status of high performance athletes (Borresen and Lambert, 2008 ; Lamberts et al., 2009 ; Daanen et al., 2012 ). For example, HRV kinetics during submaximal (D'Agosto et al., 2014 ) or maximal (Boullosa et al., 2012 ) exercise may be predictive of aerobic fitness and exercise performance. Similarly, HRV recovery following exercise occurs more rapidly in individuals with greater aerobic fitness (Stanley et al., 2013 ).

However, exercise can be performed in a multitude of different forms, including “aerobic” exercise (dynamic rhythmic exercise involving a large muscle mass, e.g., running and cycling), resistance exercise (e.g., weight/resistance training) as well as other forms (e.g., non-rhythmic/stochastic exercise, mixed-mode exercise, yoga, etc.), which may all elicit different effects on cardiac autonomic activity and HRV measures. Furthermore, these different types are each characterized by multiple sub-divisions that may be considered to constitute the exercise “dosage.” The focus of this review is on dynamic “aerobic” exercise as this form has received the most attention regarding HRV responses and is commonly used for exercise stress tests. The American College of Sports Medicine (ACSM) states that an acute bout of aerobic exercise can be modified by three primary factors constituting the exercise dose: intensity, duration, and modality (Pollock et al., 1998 ). If HRV responses to exercise and post-exercise recovery are to be interpreted with any diagnostic/prognostic value, it is important to establish how these factors of exercise prescription influence the response.

The controversies regarding the interpretation of HRV as reflecting cardiac sympathetic activity mean that HRV measures are not universally accepted to provide sympathetic insight (Eckberg, 1997 ; Billman, 2013b ). Alternatively, systolic time intervals (STI), in particular the pre-ejection period (PEP), are demonstrated to reflect cardiac sympathetic influences on myocardial contractility (Harris et al., 1967 ; Ahmed et al., 1972 ; Cacioppo et al., 1994 ). Thus, monitoring the response of STI measures during exercise and recovery may provide insights into cardiac sympathetic activity (inotropic influences) to complement HRV measures of cardiac parasympathetic modulation (chronotropic influences). Accordingly, as for HRV, it is important to establish how the exercise dose affects the STI response to exercise and recovery.

Therefore, the purpose of this review is to: (a) summarize relevant literature relating to cardiac autonomic control during exercise and recovery; (b) present relevant background information on the measurement and interpretation of HRV; (c) examine and summarize the existing literature regarding how key exercise dose factors (intensity, duration, and modality) influence HRV responses to dynamic “aerobic” exercise, in particular during post-exercise recovery; (d) examine and summarize the existing literature regarding STI responses to exercise and recovery.

Cardiac autonomic regulation during exercise and recovery

During exercise, substantial cardiovascular adjustments must occur to meet the competing demands of working muscles (metabolic demands) and skin blood flow (thermoregulatory demands), while maintaining blood pressure and adequate perfusion to other organs. Although some of the underlying mechanisms of in-vivo cardiac autonomic regulation during exercise remain contested, a prevailing model has emerged (Raven et al., 2006 ; Nobrega et al., 2014 ; White and Raven, 2014 ; Fadel, 2015 ; Fisher et al., 2015 ; Michelini et al., 2015 ) following on from the foundational work of Rowell et al. (Rowell and Oleary, 1990 ; Rowell, 1993 ; Rowell et al., 1996 ) and others (Robinson et al., 1966 ; O'leary, 1993 ; Potts and Mitchell, 1998 ). This model (Figure ​ (Figure1) 1 ) proposes that upon initiation of exercise, descending “feed-forward” inputs from higher brain centers (“central command”) into the medullary cardiovascular center reset the arterial baroreflex to a higher operating point, triggering a rapid HR increase which is primarily mediated by reduced cardiac parasympathetic neural activity (cPNA), i.e., “parasympathetic withdrawal.” Rapid feedback from muscle mechanoreceptors contributes to initial parasympathetic withdrawal, while loading of the cardiopulmonary baroreceptors (due to an increase in venous return secondary to muscle pump action) likely also elicits cPNA withdrawal as well as an initial reduction in cardiac sympathetic neural activity (cSNA). Both cSNA and cPNA regulate HR throughout the entire exercise intensity continuum—cSNA working as a “tone-setter” and cPNA operating as a “rapid responder/modulator”—with the relative “balance” shifting from predominantly “parasympathetic control” at rest and low intensities to mainly “sympathetic control” at high intensities (White and Raven, 2014 ). As exercise intensity increases further, progressive baroreflex resetting as well as afferent feedback from muscle metaboreceptors trigger further cardiac parasympathetic withdrawal and sympathetic activation, the latter of which is increasingly augmented from moderate to maximal intensity by systemic sympatho-adrenal activation. These processes are summarized in Figure ​ Figure2A 2A .

Key aspects of cardiovascular autonomic regulation, particularly during exercise and recovery . Blood pressure appears to be the primary regulated variable. Acetylcholine-esterase at the parasympathetic-cardiac junction facilitates rapid “On” and “Off” signaling (<1 s), whereas sympathetic “Off” effects are substantially slower (20+ s). Note the indirect nature of HRV and PEP as indices of cardiac parasympathetic and sympathetic activity, respectively, as well as the substantial “cross talk” (pre-and post-junctional) of cardiac sympathetic/parasympathetic effects. Also note the common pathways through which different dosages of exercise (intensity, duration, and modality) may influence cardiac autonomic regulation. AC-cAMP-PKA, adenylate-cyclase/cyclic-AMP/Protein-kinase-A cascade; ACh, acetylcholine; aS, sympathetic outflow to adrenal medulla; β1 (β2), Beta1 (Beta2) adrenergic receptors; Ca 2+ , calcium ions; cP, cardiac parasympathetic outflow; cS, cardiac sympathetic outflow; CVLM, caudal ventrolateral medulla; E, epinephrine; Gi, G-protein inhibitory subunit; Gs, G-protein stimulatory subunit; HR, heart rate; HRV, heart rate variability; K + , potassium ions; M2, M2 muscarinic receptor; MLC, myosin light chain; NA, nucleus ambiguus; Na + , sodium ions; NE, norepinephrine; NPY, neuropeptide Y; NTS, Nucleus Tractus Solitarii; P-, phosphorylation; PEP, pre-ejection period; PG, parasympathetic ganglia; Q, cardiac output; RVLM, rostro ventrolateral medulla; SG, sympathetic ganglia; SV, stroke volume; vS, vascular sympathetic outflow.

Schematic illustration of autonomic regulation of HR during exercise and recovery . Panel (A) displays HR regulation during exercise as a function of intensity. Panel (B) displays HR regulation during recovery as a function of time. As exercise intensity increases, cardiac control shifts from predominantly parasympathetic control (blue, acting as a “rapid modulator”) to predominantly sympathetic control (red, acting as a “tone-setter”). During recovery, the mechanisms eliciting cardio-acceleration during exercise are reversed, as HR regulation is gradually shifted back to predominantly parasympathetic control. cPNA, cardiac parasympathetic neural activity; cSNA, cardiac sympathetic neural activity; CC, central command; Mechano, mechanoreflex; CBR, central baroreflexes; ABR, arterial baroreflex; Metabo, metaboreflex; Symp-Adr, sympatho-adrenal; Thermo, thermoregulatory influences.

Upon exercise cessation, the aforementioned processes mediating cardio-acceleration during exercise essentially occur in reverse. However, the details of which mechanisms mediate specific aspects of the post-exercise cardio-deceleration time-profile are to some extent less well-established, in part because of greater procedural variation (e.g., active vs. passive recovery and post-exercise posture). Nevertheless, the prevailing model (Imai et al., 1994 ; Kannankeril and Goldberger, 2002 ; Kannankeril et al., 2004 ; Pierpont and Voth, 2004 ; Coote, 2010 ; Pecanha et al., 2014 ) posits that abrupt removal of “central command” together with abolished feedback from muscle mechanoreceptors (for passive recovery) resets the arterial baroreflex to a lower level and causes an initial HR decrease, which is predominantly mediated by an increase in cPNA. Hence, this “fast phase” (i.e., initial minute) of HR recovery has often been attributed to “parasympathetic reactivation” (Perini et al., 1989 ; Imai et al., 1994 ; Cole et al., 1999 ; Coote, 2010 ; Pecanha et al., 2014 ), although some evidence has suggested sympathetic involvement as well (Nandi and Spodick, 1977 ; Kannankeril et al., 2004 ; Pichon et al., 2004 ). As recovery continues, a more gradual “slow phase” of cardio-deceleration is observed, likely mediated by both progressive parasympathetic reactivation and sympathetic withdrawal. These slower autonomic adjustments are believed to be elicited primarily by an intensity-dependent combination of gradual metabolite clearance (i.e., reduced metaboreflex input) and a reduction in circulating catecholamines, while thermoregulatory factors (direct thermoreceptor afferents and/or blood flow redistribution) may also be involved. These processes are summarized in Figure ​ Figure2B 2B .

What is HRV and how is it quantified?

In 1996, the Task Force of The European Society of Cardiology and The North American Society of Pacing and Electrophysiology published a set of standards for the measurement and physiological interpretation of HRV (Camm et al., 1996 ). Briefly, as the name implies, HRV quantifies the variability of heart rate, although this is a misnomer as heart rate (beats per minute, b.min −1 ) is usually expressed as heart period (milliseconds per beat, ms) before variability is quantified. A key point is that HRV quantification involves several steps of calculations; each of these steps may be approached with different methodologies (with multiple sub-variations of these differing methodologies). As such, the field of HRV research is inherently heterogeneous from a methodological standpoint. Firstly, the length of time of data collection (epoch) varies greatly. Following data collection and removal/correction of signal artifacts and non-sinus beats, a detrending algorithm is often applied to minimize non-stationary aspects of the HR signal. Simple linear detrending is commonly employed (Camm et al., 1996 ), although higher-order or more complex algorithms have also been used (Tarvainen et al., 2002 ).

Time-domain and frequency-domain analysis

Following data collection, correction, and detrending, the vast majority of HRV research employs time-domain and/or frequency-domain HRV measures. Regarding the time-domain (computationally the simplest and most consistent analysis method across studies), the two most common measures are the standard deviation of R-R intervals (SDRR), a measure of overall variability; and the root mean square of successive differences of R-R intervals (RMSSD), a measure of beat-to-beat variability. The latter is sometimes calculated slightly differently as the standard deviation of successive differences (SDsd).

Frequency-domain measures express HRV as a function of frequency, rather than time, since different spectral power components of HRV might relate to different elements of cardiac autonomic activity (Akselrod et al., 1985 ). There are different methods (and multiple sub-variations of methods) utilized to calculate HRV spectra, with the two most common employing a Fourier transform or autoregressive modeling. More recently, frequency-domain analysis has been applied to non-stationary signals using time-varying methods such as short-time Fourier transform (STFT) analysis (Cottin et al., 2006 ; Martinmaki et al., 2006 ; Kaikkonen et al., 2010 , 2012 ). Regardless of the method used, the primary components are low frequency (LF, often 0.04–0.15 Hz) and high frequency (HF, often 0.15–0.40 Hz) spectra. Very low frequency spectra (VLF, <0.04 Hz) is seldom reported. Together these constitute total power (TP). These may be expressed as absolute power (ms 2 ) or as power spectral density (ms 2 .Hz −1 ). Additionally, the ratio of LF to HF power (LF:HF) is also often reported, and HF and LF may also be normalized to TP (which may or may not include the VLF component) giving HF-nu and LF-nu (nu = normalized units). Variations of the upper and lower limits of each frequency band have also been employed (Radaelli et al., 1996 ; Avery et al., 2001 ; Pichon et al., 2004 ; Povea et al., 2005 ; Casties et al., 2006 ; Spadacini et al., 2006 ; Martinmaki et al., 2008 ).

Finally, HRV measures are often observed to be non-normally distributed. For this reason, a data transformation (typically natural-logarithm, Ln) is sometimes applied to yield an approximately normal distribution and permit parametric statistical analysis.

Other methods of quantifying HRV

In addition to time and frequency domain HRV, other approaches of expressing the variability of HR are sometimes employed. Poincaré plots warrant mention because the measure “standard deviation 1” (SD1) is used to quantify short-term beat-to-beat HRV (similar to RMSSD). In fact, SD1 is a scaled derivative of SDsd (which is very similar to RMSSD), i.e., SD1 = SDsd ÷ √2 (Brennan et al., 2001 ). Indeed, in studies reporting SD1 and SDsd or RMSSD, manually dividing the reported SD1 results by √2 gives nearly identical results to those reported for SDsd (Tulppo et al., 1996 ) and RMSSD (Leicht et al., 2008 ). Other measures of HRV based on non-linear dynamics are less often utilized, such as power-law analysis, entropy, dimensionality and fractals (Oida et al., 1997 ; Huikuri et al., 2003 , 2009 ; Perkiomaki et al., 2005 ; Baumert et al., 2009 ). Some reports have suggested that non-linear HRV measures may provide independent risk stratification and prognostic insight to complement time and frequency domain HRV measures (Makikallio et al., 1999 ; Huikuri et al., 2003 ; Stein et al., 2005 ), as well as offer certain advantages such as not requiring a stationary signal. In particular, fractal scaling exponents derived from detrended fluctuation analysis (DFA) might be of clinical value (Peng et al., 1995 ). However, Francis et al. ( 2002 ) suggested that fractal scaling exponents are fundamentally related to a weighted form of frequency-domain analysis. Regardless, the short term scaling exponent (α1) derived from DFA during incremental exercise may be modified by training (Karavirta et al., 2009 ). Moreover, it is interesting that the “inverted-U” bi-phasic response of DFAα1 during incremental exercise (Hautala et al., 2003 ) appears to be generally consistent with the bi-phasic nature of parasympathetic reflex control of HR as a function of exercise intensity (White and Raven, 2014 ). Notwithstanding the potential value of non-linear HRV measures (which warrant further investigation), these techniques are beyond the scope of this review.

Physiological interpretation of HRV

Interpretation of HRV as reflecting certain aspects of cardiac autonomic activity is complicated by the fact that rather than being a direct measure of autonomic nerve activity, HRV quantifies the modulation of the end-organ response, i.e., HR. HR is under the dual control of direct sympathetic and parasympathetic innervation at the sino-atrial (SA) node, but is also influenced by sympatho-adrenal activation and non-autonomic effects such as mechanical/hemodynamic influences and local reflexes (Bernardi et al., 1990 ; Rowell et al., 1996 ). Furthermore, complex non-linear pre-junctional and post-junctional parasympathetic-sympathetic interactions have been reported (Levy, 1971 ; Kawada et al., 1996 ; Sunagawa et al., 1998 ; Uijtdehaage and Thayer, 2000 ; Miyamoto et al., 2003 , 2004 ; Paton et al., 2005 ), whereby activity of one branch can in some instances augment and in other instances attenuate the activity and/or effect of the other branch. It is therefore critical to appreciate that HRV is an indirect indicator of cardiac autonomic modulation.

Interpretation of HRV measures associated with respiratory sinus arrhythmia

Despite the inherent limitations of HRV as an indirect measure of modulation of cardiac effect, oscillations in cPNA (secondary to respiration) are generally considered to be the primary contributor to HRV measures expressing beat-to-beat/rapid HR variations (Pomeranz et al., 1985 ; Randall et al., 1991 ; Malik and Camm, 1993 ; Camm et al., 1996 ; Goldberger et al., 2006 ). Parasympathetic blockade reduces these “cPNA-HRV measures” (e.g., RMSSD, SD1 and HF) in a dose-dependent manner (Tulppo et al., 1996 ; Medigue et al., 2001 ; Hautala et al., 2003 ). Furthermore, during exercise the response of these measures is generally consistent with current understandings of cardiac autonomic regulation, i.e., an increase in cSNA and a decrease in cPNA (Robinson et al., 1966 ; Rowell and Oleary, 1990 ; White and Raven, 2014 ). Under normal conditions these measures are gradually reduced during progressive exercise (Yamamoto et al., 1991 ; Tulppo et al., 1996 , 1998 ; Hautala et al., 2003 ; Casties et al., 2006 ; Martinmaki et al., 2008 ; Karapetian et al., 2012 ), whereas this response is greatly reduced or abolished under parasympathetic blockade (Tulppo et al., 1996 ; Warren et al., 1997 ; Hautala et al., 2003 ).

However, cPNA-HRV measures are not exact quantitative measures of cardiac parasympathetic activity. Although greater parasympathetic activity is generally associated with greater modulation of parasympathetic effect, very high parasympathetic activity may lead to acetylcholine saturation at the SA node, thus decreasing HRV (Eckberg, 2003 ; Dewland et al., 2007 ). Also, while sympathetic activity tends to have lesser/minimal effects on these cPNA-HRV measures (Warren et al., 1997 ; Polanczyk et al., 1998 ; Tulppo et al., 1999 ; Ng et al., 2009 ), some influence must be acknowledged since sympathetic blockade may augment these measures (Taylor et al., 2001 ; Ng et al., 2009 ), possibly via a permissive effect due to decreased sympathetic inhibition of parasympathetic activity/effect. This highlights the importance of considering parasympathetic-sympathetic interactions (Figure ​ (Figure1) 1 ) whenever interpreting measures of cardiac autonomic activity. During exercise, cPNA-HRV measures usually reach an intensity-dependent minimum at a moderate intensity, e.g., 50–60% maximal oxygen uptake (VO 2 max) (Tulppo et al., 1996 ; Yamamoto et al., 2001 ; Cottin et al., 2006 ), whereas cardiac parasympathetic activity may demonstrate a progressive decrease up to maximal exercise intensity (White and Raven, 2014 ). Again, it might be that despite small parasympathetic effects on HR even at maximal intensity (Kannankeril et al., 2004 ), beat-to-beat modulation of these effects above moderate intensity may be inhibited by strong sympathetic (cardiac nerve and sympatho-adrenal) activation.

Interpretation of other HRV measures

Measures of overall variability (e.g., SDRR and TP) are considered to be influenced by both parasympathetic as well as sympathetic cardiac activity, although these measures tend to be associated with cPNA-HRV measures (Tulppo et al., 1996 ; Warren et al., 1997 ; Hautala et al., 2003 ; Martinmaki et al., 2008 ), suggesting that parasympathetic activity is the primary influence. Although HF-nu has sometimes been employed as a cPNA-HRV index (Saito and Nakamura, 1995 ; Avery et al., 2001 ; Casonatto et al., 2011 ; Teixeira et al., 2011 ), the response of this measure during exercise is not consistent with a decrease in cardiac parasympathetic activity (Hautala et al., 2003 ; Casties et al., 2006 ; Leicht et al., 2008 ; Martinmaki and Rusko, 2008 ).

Some studies have suggested that LF, LF-nu, and LF:HF may reflect sympathetic activity or “sympatho-vagal balance” (Pagani et al., 1986 ; Malliani et al., 1991 ; Yamamoto et al., 1991 ; Saito and Nakamura, 1995 ; Avery et al., 2001 ; Teixeira et al., 2011 ), although this is controversial (Cacioppo et al., 1994 ; Eckberg, 1997 ; Warren et al., 1997 ; Billman, 2006 , 2013b ). In particular, parasympathetic blockade greatly attenuates LF (Cacioppo et al., 1994 ; Warren et al., 1997 ; Hautala et al., 2003 ), suggesting strong parasympathetic influence. Furthermore, under conditions that would be expected to increase cardiac sympathetic activity, such as exercise (Robinson et al., 1966 ; White and Raven, 2014 ) and myocardial ischemia (Houle and Billman, 1999 ), a decrease in LF is often observed (Arai et al., 1989 ; Yamamoto et al., 1991 ; Casadei et al., 1995 ; Houle and Billman, 1999 ; Hautala et al., 2003 ; Casties et al., 2006 ; Martinmaki and Rusko, 2008 ; Martinmaki et al., 2008 ). Thus, although sympathetic activity does contribute to LF, parasympathetic activity appears to be a stronger influence (Pomeranz et al., 1985 ; Randall et al., 1991 ; Cacioppo et al., 1994 ). There is also a strong parasympathetic influence on ratio measures (LF-nu and particularly LF:HF; Cacioppo et al., 1994 ), likely because HF (predominantly reflecting respiratory sinus arrhythmia) is directly used in the calculation of these measures. Furthermore, inconsistent responses to stressors which would increase sympathetic activity have been observed (Radaelli et al., 1996 ; Hautala et al., 2003 ; Casties et al., 2006 ; Martinmaki and Rusko, 2008 ), although generally a decrease has been reported during higher intensity exercise and thus they are not consistent with sympathetic activity/dominance. Additionally, the concept of “sympatho-vagal balance” has been challenged at a conceptual level (Eckberg, 1997 ; Billman, 2013b ; White and Raven, 2014 ), in particular the underlying assumptions that sympathetic activity is a key contributor to LF and that sympathetic and parasympathetic activity/effects operate in a reciprocal manner with linear interactions. Despite these controversies, LF:HF is often employed measure of “sympatho-vagal balance,” where supposedly an increase indicates “sympathetic dominance” and a decrease indicates “parasympathetic dominance.”

Summary—HRV background and physiological interpretation

For clinicians and exercise/sport scientists, the primary interest in HRV relates to (a) its prognostic potential value in cardiac disease and sudden cardiac death (Camm et al., 1996 ; Tsuji et al., 1996 ; Kikuya et al., 2000 ), and (b) the general acceptance of beat-to-beat measures (cPNA-HRV) as indicators of cardiac parasympathetic modulation. Indeed, notwithstanding some interpretative controversies, these measures are regularly employed in this manner. The underlying mechanisms and interpretation of other HRV measures (such as LF, LF-nu, and LF:HF) are less established, likely resulting from complex sympathetic-parasympathetic interactions. While this has not prevented the use of these measures as indices of sympathetic activity or “sympatho-vagal balance,” the majority of evidence does not support this approach. In light of interpretative controversies, cPNA-HRV measures perhaps provide qualitative or ordinal (rather than parametric quantitative) insight into cardiac parasympathetic activity (Billman, 2006 ).

HRV during exercise

Several studies have investigated HRV during exercise (Bernardi et al., 1990 ; Perini et al., 1990 , 2000 ; Yamamoto et al., 1991 , 1992 ; Dixon et al., 1992 ; Radaelli et al., 1996 ; Tulppo et al., 1996 , 1998 , 1999 ; Shibata et al., 2002 ; Casties et al., 2006 ; Cottin et al., 2006 , 2007 ; Karapetian et al., 2008 ; Kaikkonen et al., 2010 ). However, in addition to widely varying HRV analysis methodologies amongst the HRV literature, studies employing exercise with HRV measurements also vary markedly in terms of the participants and exercise protocol. Studies have used a range of exercise modalities, with cycling the most common mode employed (Yamamoto et al., 1991 ; Radaelli et al., 1996 ; Tulppo et al., 1996 , 1998 ; Hautala et al., 2003 ; Casties et al., 2006 ; Cottin et al., 2006 ; Karapetian et al., 2008 ; Martinmaki and Rusko, 2008 ; Martinmaki et al., 2008 ), although walking/running has also been utilized (Hautala et al., 2003 ; Cottin et al., 2007 ; Botek et al., 2010 ; Kaikkonen et al., 2010 ). Other modes less commonly used are arm-cranking (Tulppo et al., 1999 ; Leicht et al., 2008 ), rowing (Cheng et al., 2005 ) and swimming (Di Michele et al., 2012 ). Regarding the effect of exercise dose factors (intensity, duration, and modality), intensity has received the most investigative attention, while fewer studies have investigated the effects of duration and modality.

Effect of exercise intensity

Several studies have investigated the effect of exercise intensity on HRV during exercise. In addition to different modalities being utilized between these studies, the duration for which participants exercised at each intensity varies greatly, such as 2 min (Tulppo et al., 1996 ; Martinmaki et al., 2008 ), 3 min (Tulppo et al., 1998 , 1999 ; Hautala et al., 2003 ; Karapetian et al., 2012 ), 5 min (Radaelli et al., 1996 ), 8 min (Casties et al., 2006 ; Michael et al., 2016 ), 10 min (Martinmaki and Rusko, 2008 ) and 15 or more min (Yamamoto et al., 1991 ; Saito and Nakamura, 1995 ; Leicht et al., 2008 ; Boettger et al., 2010 ). Nevertheless, an analysis of the literature allows us to identify some general responses for some HRV measures as a function of exercise intensity.

Time-domain measures

Exercise elicits a reduction in HRV when expressed in the time domain, regardless of whether the HRV metric is based on overall HRV calculated from R-R intervals (e.g., SDRR), or beat-to-beat HRV metrics based on the difference in R-R intervals (e.g., RMSSD). Several studies have reported that higher exercise intensity is associated with a lower SDRR (Yamamoto et al., 1991 ; Saito and Nakamura, 1995 ; Radaelli et al., 1996 ; Tulppo et al., 1996 ; Hautala et al., 2003 ; Casties et al., 2006 ; Spadacini et al., 2006 ; Leicht et al., 2008 ; Fisher et al., 2009 ; Karapetian et al., 2012 ). Compared to resting values which vary greatly (typically 40–100 ms), there is a somewhat consistent intensity dose-response, namely a curvilinear decay to 3–10 ms for exercise >160 b.min −1 . The results of some studies indicate that SDRR reaches a minimum at moderate-high intensity, and does not change substantially thereafter, whereas other studies suggest that small decreases in SDRR continue as intensity is increased toward maximal. Regardless, it is clear that SDRR is drastically reduced as a function of exercise intensity. When natural-log transformed, SDRR (Ln-SDRR) demonstrates a somewhat linear decrease as a function of exercise intensity.

The basic response of cPNA-HRV (e.g., RMSSD) is similar to that of SDRR, i.e., a higher exercise intensity is associated with a lower RMSSD, and that this follows a relatively consistent curvilinear decay profile as a function of exercise intensity (Tulppo et al., 1996 ; Povea et al., 2005 ; Leicht et al., 2008 ; Fisher et al., 2009 ; Boettger et al., 2010 ; Lunt et al., 2011 ; Karapetian et al., 2012 ). The intensity-dependent reduction in RMSSD occurs more rapidly than compared with SDRR, such that a minimum value (<5 ms) is reached at a moderate intensity, i.e., ~120–140 b.min −1 , or 50–60% VO 2 max. Thereafter, RMSSD does not change substantially, although slight increases have sometimes been observed at higher exercise intensities (e.g., >180 b.min −1 ). As discussed in the previous section, SD1 (from Poincaré plots) is essentially a scaled version of RMSSD, thus it is not surprising that the intensity dose-response for SD1 (Figure ​ (Figure3) 3 ) is the same as that of RMSSD (Tulppo et al., 1996 ; Leicht et al., 2008 ; Garcia-Tabar et al., 2013 ). When these measures are expressed relative to the underlying RRI, which has been suggested to minimize the purely mathematical influence of HR on HRV (Sacha, 2013 , 2014 ; Pradhapan et al., 2014 ), the same intensity dose-response decay profile is maintained (Tulppo et al., 1998 , 1999 ). When natural-log transformed, RMSSD (Ln-RMSSD) demonstrates an approximately linear decrease until moderate-high intensities (140–160 b.min −1 ), followed by no change or a small increase at higher intensities.

cPNA-HRV (SD1 from Poincaré plot, ms) during rest (hollow) and incremental exercise (filled) . Data are mean ± SD. Redrawn from Tulppo et al. ( 1996 ).

Regarding practical application, the point at which beat-to-beat cPNA-HRV measures such as RMSSD and SD1 reach a minimum has been termed a “HRV threshold” (HRVT). The HRVT appears to be a proxy for determining the intensity corresponding to the first ventilation threshold (VT1) or first lactate threshold (Karapetian et al., 2008 , 2012 ; Sales et al., 2011 ; Garcia-Tabar et al., 2013 ), thus making it a potentially useful tool in exercise testing/monitoring and training prescription.

Frequency domain measures

Similar to when expressed in the time domain, the frequency domain measures of HRV also demonstrate a pronounced reduction in response to exercise. While the conventional HF band of 0.15–0.40 Hz has often been employed during exercise (Tulppo et al., 1996 , 1998 ; Hautala et al., 2003 ; Leicht et al., 2008 ; Fisher et al., 2009 ; Boettger et al., 2010 ), this band may not be suitable during exercise where higher respiratory frequencies are observed, and therefore different lower limits such as 0.18 Hz have previously been employed (Radaelli et al., 1996 ; Spadacini et al., 2006 ), as have different upper limits including 0.35 Hz (Spadacini et al., 2006 ), 0.50 Hz (Dixon et al., 1992 ; Avery et al., 2001 ; Povea et al., 2005 ), 0.80 Hz (Lunt et al., 2011 ), 1.00 Hz (Pichon et al., 2004 ; Casties et al., 2006 ; Martinmaki and Rusko, 2008 ), 1.20 Hz (Martinmaki et al., 2008 ), 1.50 Hz (Perini et al., 2006 ; Michael et al., 2016 ), and 2.00 Hz (Cottin et al., 2006 ). Furthermore, most studies report absolute power (ms 2 ) Yamamoto et al., 1991 ; Saito and Nakamura, 1995 ; Tulppo et al., 1996 ; Hautala et al., 2003 ; Pichon et al., 2004 ; Spadacini et al., 2006 ; Leicht et al., 2008 ; Martinmaki and Rusko, 2008 ; Martinmaki et al., 2008 ; Fisher et al., 2009 , although some report power spectral density (ms 2 .Hz −1 ) (Avery et al., 2001 ; Povea et al., 2005 ; Casties et al., 2006 ). Several studies report raw measures (Yamamoto et al., 1991 ; Saito and Nakamura, 1995 ; Tulppo et al., 1996 ; Avery et al., 2001 ; Hautala et al., 2003 ; Pichon et al., 2004 ; Casties et al., 2006 ; Leicht et al., 2008 ; Fisher et al., 2009 ), while others report natural log transformed measures (Radaelli et al., 1996 ; Spadacini et al., 2006 ; Martinmaki and Rusko, 2008 ; Martinmaki et al., 2008 ; Kaikkonen et al., 2010 ; Lunt et al., 2011 ). Another discrepancy occurs when considering studies investigating normalized power, as the total power to which measures are normalized may (Saito and Nakamura, 1995 ; Warren et al., 1997 ; Povea et al., 2005 ; Boettger et al., 2010 ) or may not (Hautala et al., 2003 ; Pichon et al., 2004 ; Casties et al., 2006 ; Leicht et al., 2008 ; Martinmaki and Rusko, 2008 ) include the VLF component.

Despite the heterogeneity regarding the signal analysis methodology of frequency domain HRV, some common trends are observed. As for time-domain measures, LF, HF, and TP all demonstrate a substantial decay-type decrease with increasing exercise intensity, up to a particular intensity after which minimal spectral power remains and no further decrease is observed. The intensity above which no further decline is observed varies greatly, but is usually in the range of 120–180 b.min −1 (Radaelli et al., 1996 ; Tulppo et al., 1996 ; Avery et al., 2001 ; Hautala et al., 2003 ; Povea et al., 2005 ; Casties et al., 2006 ; Spadacini et al., 2006 ; Martinmaki et al., 2008 ; Fisher et al., 2009 ). As for beat-to-beat time domain measures, some studies have demonstrated that analysis of the HF decay with incremental exercise may serve as a surrogate method for determining the intensity associated with first ventilation threshold (Cottin et al., 2006 , 2007 ). Furthermore, HF power multiplied by the HF frequency may provide an approximation for the intensity associated with the second ventilation threshold (Cottin et al., 2006 , 2007 ; Buchheit et al., 2007b ).

Considering the relationship between normalized measures of LF and HF (usually relative to LF+HF or VLF+LF+HF), it is not surprising that these measures behave in an opposite manner to each other during exercise. Typically, LF-nu increases during low-moderate intensity exercise and decreases during higher intensity exercise, while HF-nu demonstrates the opposite response (Bernardi et al., 1990 ; Perini et al., 1990 , 1993 , 1998 ; Hautala et al., 2003 ; Pichon et al., 2004 ; Povea et al., 2005 ; Martinmaki and Rusko, 2008 ). However, conflicting responses have also been reported (Perini et al., 1993 ; Avery et al., 2001 ; Casties et al., 2006 ; Boettger et al., 2010 ). LF:HF demonstrates inconsistent responses to exercise. Some studies reported an increase until low-moderate intensity (110–130 b.min −1 ), followed by a decrease during higher intensities (Radaelli et al., 1996 ; Tulppo et al., 1996 ; Hautala et al., 2003 ). However, other studies reported a progressive decrease from rest with increasing exercise intensity (Casties et al., 2006 ), a progressive increase from rest (Saito and Nakamura, 1995 ; Avery et al., 2001 ), or minimal change at low-moderate intensity followed by a sharp increase at moderate intensity (Yamamoto et al., 1991 ). Some of these divergent findings may be attributed to varying methodology (especially HRV analysis techniques). However, it is interesting that these normalized and ratio measures rarely (if ever) behave in a manner that is consistent with the established exercise response of systems they purportedly reflect—namely cardiac parasympathetic activity (HF-nu) and sympathetic activity or sympatho-vagal balance (LF-nu and LF:HF).

Effect of exercise duration

Compared to the effect of exercise intensity, few studies have investigated the effect of exercise duration HRV responses during exercise. This may be in part due to the pronounced effect of intensity on HRV, which has two important implications for investigating other influences on HRV during exercise: (1) Due to the HRV-intensity relationship (whereby most HRV measures reach a near-zero minimum at/above moderate intensity), any potential influence of duration is minimized; and (2) During prolonged exercise HR may change despite no change in external load, i.e., cardiovascular drift (Montain and Coyle, 1992 ), meaning that “internal intensity” is important to consider. Thermoregulatory factors (in particular fluid losses from sweating) play a key role in cardiovascular drift as exercise is prolonged (Coyle and Gonzalez-Alonso, 2001 ), with significant cardiovascular autonomic implications due to changes in blood pressure and blood flow redistribution (Figure ​ (Figure1 1 ).

Kaikkonen et al. ( 2007 ) investigated the effect of exercise duration on frequency domain HRV measures during different exercise intensities in sedentary women, who performed low intensity (~45% VO 2 max) and high intensity (~77% VO 2 max) exercise for two different durations (approximately 38 vs. 76 min for low intensity and 30 vs. 60 min for high intensity). Despite a doubling of exercise duration, no significant difference in HRV during exercise was observed for either intensity. Similar findings were reported by the same authors in a later study (Kaikkonen et al., 2010 ) investigating the effect of a 300–400% increase in exercise duration, in which recreational-level athletes ran at ~66% VO 2 max for ~20 or ~90 min.

Pichon et al. ( 2004 ) investigated frequency domain HRV measures during different short-duration exercise bouts, i.e., 3, 6, and 9 min of exercise at 60 and 70% of power reached at VO 2 max (PVO 2 max), and 3 and 6 min at 80% PVO 2 max. HF was higher during 3 min compared to 6 and 9 min. LF also decreased with increased exercise duration. No significant effects were observed for normalized measures or LF:HF. However, as all three intensities were of at least moderate-high intensity, the heart rates differed substantially between exercise durations. The lowest intensity (60% VO 2 max) elicited a mean HR of 143, 161, and 167 b.min −1 at 3, 6, and 9 min, respectively. These differences were enhanced with higher intensities, likely in part because of exercise onset HR kinetics and the time required to reach steady state. Thus, despite these bouts being matched for relative intensity (in terms of %PVO 2 max), the elevated HR associated with higher intensity exercise of such short durations means it is difficult to separate the independent effect of exercise duration from that of exercise intensity in this study. Similarly, Moreno et al. ( 2013 ) investigated time and frequency domain HRV throughout 90 min of moderate intensity exercise (60% VO 2 peak). Despite HRV measures being greatly reduced by the time of the initial exercise recording (25 min), the authors found that HR increased throughout 90 min of exercise (from ~140 to ~150 b.min −1 ), while SDRR, RMSSD, LF and HF were further decreased by 90 min.

Effect of exercise modality

Relatively few studies have compared the HRV responses during different modalities of dynamic exercise. In addition to the methodological challenge involving the strong HRV-intensity relationship, it is difficult to standardize exercise “intensity” when comparing different modalities. Some of these issues relate to what basic metric to employ (e.g., power output, HR, or VO 2 ). For example, arm crank exercise will elicit a higher HR when compared with leg cycling for the same rate of energy expenditure (same absolute VO 2 ; Tulppo et al., 1999 ) or work rate (Leicht et al., 2008 ). There is also the issue of whether to match relative or absolute intensity (and whether this is mode-specific). For instance, peak HR will likely be 10–20 b.min −1 lower during arm cranking compared with leg cycling (Tulppo et al., 1999 ; Ranadive et al., 2011 ). The physiological response (e.g., HR or VO 2 ) elicited for any intensity of arm crank exercise is likely not equal to response elicited during leg cycling exercise, despite the workload representing the same percentage of mode-specific maximum HR or maximum VO 2 . Because of these inherent methodological challenges, the effect of modality on HRV during exercise has been investigated with varying approaches.

Different modes of dynamic exercise

Tulppo et al. ( 1999 ) investigated HRV during incremental arm crank compared to incremental cycling exercise using SD1 and SD1n (SD1 divided by RRI) as parasympathetic measures. As expected for both modes of exercise, these measures decreased as intensity increased until a moderate power output (~50% VO 2 max) was reached. Below this power output, HR was higher and HRV was lower during arm exercise compared with leg exercise. The authors concluded that incremental arm exercise results in a more rapid vagal withdrawal compared with incremental leg exercise. Interestingly, manual re-plotting of the data reveals that when HRV is plotted against HR (rather than VO 2 ), there appears to be no difference between the modalities. These findings were in contrast to those of Leicht et al. ( 2008 ), who investigated HRV responses during lower body exercise (cycling), upper body exercise (arm crank), and whole body exercise (running) matched for HR. Absolute VO 2 was similar during cycling and running but lower during arm crank, whereas perceived exertion was higher during arm crank. Despite HR being matched, time and frequency domain measures were similar during cycling and running, but were higher during arm crank, thus demonstrating that exercise mode can affect HRV, independent of the underlying HR. Whilst methodological differences may explain part of these apparently discrepant findings, it is nonetheless clear that exercise intensity is a stronger determinant of HRV response to exercise than modality.

A small number of studies have investigated HRV during different conditions/modalities that are likely to elicit different circulatory/orthostatic effects. Di Rienzo et al. ( 2008 ) compared normal cycling to 0-g cycling (in space) at 75W, though no significant differences in HR or HRV (time and frequency domain) were observed. Somewhat comparable are the findings of Perini et al. ( 1993 ), who compared upright vs. supine cycling at different absolute workloads (50, 100, 150 W). RRI and TP both decreased with increasing intensity in a similar manner for both postures (although the normalized powers for VLF, LF, and HF varied as a function of intensity and posture). Takahashi et al. ( 2000 ) investigated HRV responses in older adults during incremental treadmill exercise on land and whilst immersed in water, with HF amplitude (rather than power) utilized as a measure of cardiac vagal outflow. HR differed during incremental exercise between the two conditions (lower in water at low intensities, then higher in water at high intensities, compared to land) although no significant difference was found in HF amplitude at any intensity. Similar findings were reported by Perini et al. ( 1998 ), who investigated the effect of water immersion on HRV responses to incremental cycling exercise and found that HRV responses were similar between the two conditions throughout incremental exercise. Thus, if these orthostatic conditions had any influence on HRV during exercise, the effects were minor when compared with the strong influence of exercise intensity.

Other “mode” comparisons

Although the focus of this review is on dynamic “aerobic” exercise, there have been some studies comparing the HRV responses amongst other “modes” of exercise that warrant mention, particularly considering the scarcity of research involving aerobic exercise modes and HRV. Cottin et al. ( 2004 ) investigated HRV measures during two different high intensity exercises, namely dynamic (cycling) vs. irregular (Judo Randori). HR was similar during both exercises (above 180 b.min −1 ), although frequency domain and Poincaré HRV measures were higher during Judo. Due to the high heart rates, the physiological interpretation of these findings is difficult, as non-neural mechanisms such as respiration are believed to substantially influence HRV at such high intensities (Casadei et al., 1995 , 1996 ; Cottin et al., 2006 , 2007 ). Furthermore, the metabolic drive to HR vs. the hormonal stress anticipatory response of combat activity (Salvador et al., 2003 ) represent different aspects of “modality” that are not easily accounted for or compared.

Gonzalez-Camarena et al. ( 2000 ) compared static exercise (isometric knee extension at 30% MCV) to dynamic exercise (cycling at 30 and 60% VO 2 max). HRV (time and frequency domain measures) was reduced during dynamic exercise whereas an increase was observed during static exercise, although HR was not matched between exercises. Similar finding were reported by Weippert et al. ( 2013 , 2015 ), who compared HRV during static exercise (supine isometric leg press) vs. dynamic exercise (supine cycling) at similar low-intensity heart rate levels (~88 b.min −1 ). HRV measures of overall variability (SDRR) as well as beat-to-beat variability purportedly reflecting cardiac parasympathetic activity (RMSSD, HF, SD1) were higher during static exercise compared to dynamic exercise (as was subject effort, blood pressure, and rate-pressure product).

Reliability of HRV during exercise

A small number of studies have investigated the test-retest reliability of HRV measures during exercise. Apart from aforementioned disparities amongst the literature regarding exercise protocols and HRV analysis methodology, there are also differences in the assessment of reliability. The intraclass correlation coefficient (ICC) reflects the ability of a test to differentiate between individuals and is therefore considered a measure of relative reliability, although there are multiple methods of calculating ICC (Weir, 2005 ). Alternatively, measures such as the coefficient of variation (CV), typical error of measurement (TEM, which may be expressed as a CV), and Bland-Altman limits of agreement (LoA) may be considered measures of absolute reliability as they reflect the trial-to-trial noise. Carrasco et al. ( 2003 ) demonstrated that the ICC for a range of time and frequency domain measures (Ln-transformed) during cycling exercise at 80 W ranged between 0.70 and 0.91, which was comparable to resting measures. Similar ICC-values (~0.9) were reported for Ln-transformed time domain measures during slow walking at 4 km/hr in another study (Boullosa et al., 2014 ). These authors also reported that that the TEM(CV) was 16–22%. Utilizing a Bland-Altman approach, Tulppo et al. ( 1998 ) demonstrated that the LoA were lower during cycling exercise for time and frequency domain measures compared to rest, although this may be partly due to the substantially reduced values during exercise compared with rest.

Summary—HRV during exercise

During exercise, HRV measures demonstrate a curvilinear decay as a function of exercise intensity, that is closely related to exercising HR. HRV measures associated with cardiac parasympathetic activity (e.g., RMSSD and HF) usually reach a near-zero minimum at moderate intensity (possibly being associated with the first ventilation/lactate threshold). These measures are sometimes observed to increase slightly as exercise intensity increases toward maximum, although this is likely mediated by non-neural mechanisms such as direct mechanical effects of respiration on the SA node. The data also leads to further questioning of the use of frequency domain ratio and normalized measures as indicators of sympathetic activity or “sympatho-vagal balance.” In addition to demonstrating inconsistent responses to exercise, the response of these measures is rarely consistent with our current understanding of autonomic control during exercise, namely progressive parasympathetic withdrawal and sympathetic activation (White and Raven, 2014 ).

Regarding exercise duration, the limited body of literature suggests that prolonged exercise duration can influence (attenuate) HRV during exercise, although this has only been observed in studies where there is a concomitant increase in HR (i.e., cardiovascular drift) and also when HRV has not already reached the intensity-dependent minimum. Indeed, this elevated HR (representing elevated internal intensity, regardless of the external load) may be the cause for any change in HRV, rather than any direct effect of exercise duration. Conversely, prolonged exercise duration may be associated with progressive parasympathetic withdrawal (indicated by lower HRV), and this in turn might contribute to cardiovascular drift (Kukielka et al., 2006 ). Regardless of the direction of any cause-effect relationship, it is difficult to infer what affects HRV under conditions of different HR-values, as HR likely has a purely mathematical effect on HRV (Billman, 2013a ; Sacha, 2013 , 2014 ), whereby a greater HR can reduced HRV despite no change in the actual variability of autonomic outflow.

While there is limited data available, the evidence suggests that exercise modality can modify HRV, although not all research supports this. Interpretation of results is difficult as any influence of exercise modality is confounded by the issue of matching for exercise “intensity.” Interestingly, studies investigating different dynamic modalities that would be expected to elicit different orthostatic/circulatory conditions (e.g., posture, gravity, or water immersion) have typically found no substantial effect on HRV during exercise. Instead, the muscle group employed and the mode of contraction (e.g., static vs. dynamic) seem to be stronger modality-related factors influencing HRV response during exercise. Nevertheless, exercise intensity is the strongest determinant of HRV during exercise.

HRV during post-exercise recovery

Most HRV measures are substantially reduced during exercise (see previous section). HRV has also been employed as a tool to investigate post-exercise autonomic (predominantly parasympathetic) activity (Goldberger et al., 2006 ; Buchheit et al., 2007a , 2009a ; Al Haddad et al., 2010 ; Stanley et al., 2013 ; Ahmadian et al., 2015 ). Upon exercise cessation, HR and HRV demonstrate a time-dependent recovery and eventual return to pre-exercise levels (Stanley et al., 2013 ). Recovery conditions such as posture have also been shown to affect HRV recovery (Barak et al., 2010 ), with a more upright posture slowing recovery (Buchheit et al., 2009a ). Rapid (though incomplete) recovery is commonly observed in the initial minutes following exercise (Buchheit et al., 2007a , 2009b ; Kaikkonen et al., 2007 , 2010 , 2012 ; Martinmaki and Rusko, 2008 ; Al Haddad et al., 2011 ; Stanley et al., 2013 ). While complete recovery may take up to 48 h following some bouts of exercise and may sometimes involve an “overshoot” above pre-exercise levels prior to 48 h (Furlan et al., 1993 ; Hautala et al., 2001 ; Stanley et al., 2013 ), the focus of this review is on the immediate post-exercise recovery period (e.g., 0–10 min). As is the case during exercise, the majority of studies investigating the effect of exercise on HRV recovery have focused on different exercise intensities, while fewer studies have investigated the effects of exercise duration or modality.

Several studies have investigated HRV during recovery from different exercise intensities. A recent review (Stanley et al., 2013 ) quantitatively summarized the findings of some of these studies demonstrating that a higher exercise intensity is associated with a slower recovery of cPNA-HRV measures, specifically Ln-transformed RMSSD or HF (Furlan et al., 1993 ; Terziotti et al., 2001 ; Mourot et al., 2004 ; Parekh and Lee, 2005 ; Niewiadomski et al., 2007 ; Seiler et al., 2007 ; Kaikkonen et al., 2008 ). These data are redrawn in Figure ​ Figure4. 4 . Other studies have reported overall similar findings for cPNA-HRV and other HRV measures, namely that a greater exercise intensity results in a slower HR and HRV recovery (Perini et al., 1990 ; Hayashi et al., 1992 ; Buchheit et al., 2007a ; Kaikkonen et al., 2007 , 2010 , 2012 ; Martinmaki and Rusko, 2008 ; Gladwell et al., 2010 ; Al Haddad et al., 2011 ; Casonatto et al., 2011 ; Dupuy et al., 2012 ). From a mechanistic standpoint, the effect of intensity on HRV recovery is likely associated with the amount of non-oxidative energy contribution and subsequent stimulation of the muscle metaboreflex (Buchheit et al., 2007a ).

Time course of cPNA-HRV recovery following different intensities of preceding exercise . Ln natural log transform. RMSSD root mean square of successive differences. HF High frequency power. Low intensity: <70% VO 2 max. Moderate intensity: 70–82% VO 2 max. High intensity: >82% VO 2 max. Data are mean ± SD. Redrawn from Stanley et al. ( 2013 ).

However, the details of the intensity dose-response has not been clearly elucidated, particularly during the initial 10 min following exercise when comparing moderate vs. high exercise intensities. Most studies investigating HRV following more than two intensities report differences between some, but not all exercise intensities (Kaikkonen et al., 2007 ; Seiler et al., 2007 ; Gladwell et al., 2010 ; Casonatto et al., 2011 ). Seiler et al. ( 2007 ) suggested that the first ventilation threshold might demarcate an autonomic “binary threshold” regarding HRV recovery, whereby exercise below this intensity is associated with rapid HRV recovery while exercise above this intensity results in delayed HRV recovery which is intensity-independent (at least in highly trained athletes). In contrast, inspection of Figure ​ Figure4 4 suggests that exercise intensity may elicit a more graded (rather than binary) effect on HRV recovery (at least during the first hour post-exercise). This latter interpretation is consistent with the findings of a recent study by this research group demonstrating a graded response following three different exercise intensities (Michael et al., 2016 ).

Similar to responses during acute exercise, few studies have investigated the effect of exercise duration on HRV during recovery after exercise, in particular when controlling for the intensity. Seiler et al. ( 2007 ) investigated the recovery of HRV following either 60 or 120 min of low intensity running exercise (below VT1 at ~60% VO 2 max) in a highly trained athletic population. By the earliest time-point investigated post-exercise (5–10 min), HRV (including RMSSD) had recovered to pre-exercise levels and there was no significant effect of duration on HRV measures throughout the 4 h recovery period. These findings are consistent with those reported by Casonatto et al. ( 2011 ), who investigated the effect of exercise duration on HRV during 60 min of recovery by having healthy participants cycle for 30 min or ~45 min at 60% VO 2 max. There was no significant duration effect for the HRV measures assessed during recovery (RMSSD, LF-nu, HF-nu, and LF:HF).

The findings of Kaikkonen et al. ( 2007 ) also suggest a lack of any duration effect on HRV during the immediate recovery period. In that study, the time-course of immediate HRV recovery in sedentary women was investigated using STFT frequency-domain analysis. The participants performed low intensity (~45% VO 2 max) and moderate-high intensity (~77% VO 2 max) running for ~38 vs. ~76 min and ~30 vs. ~60 min for low and moderate-high intensity, respectively. Despite a strong effect of exercise intensity on the time-course of HRV recovery from exercise, there was no significant effect of exercise duration for either intensity from 1 to 30 min recovery.

While these studies suggest that a 100% increase in exercise duration does not alter HRV recovery following exercise, the results of another study by Kaikkonen et al. ( 2010 ) indicate that exercise duration may influence post-exercise HRV when duration is extended by 300–400%. In that study, recreation-level athletes ran at ~66% VO 2 max for ~20 and ~90 min. During the initial 3 min of a 15 min recovery, the longer duration resulted in lower Ln-LF, Ln-HF, and Ln-TP.

Very few studies have examined post-exercise HRV recovery following different modalities of “aerobic” exercise. Cunha et al. ( 2015 ) investigated the immediate recovery period following incremental exercise of three modalities: walking, cycling, and running. During the 5 min recovery period, it was reported that HR recovery as well as the recovery of RMSSD was more rapid following exercise involving a smaller muscle mass or energy expenditure (i.e., cycling > walking > running), with the authors therefore concluding that muscle mass and/or energy expenditure are determinants of post-exercise parasympathetic reactivation. Similarly, active children demonstrated more rapid HR and HRV recovery following maximal incremental arm-cranking vs. cycling (Ahmadian et al., 2015 ), The findings of these two studies were also consistent with those of other studies investigating HR recovery (without HRV measurement) following exercise mode comparisons such as cycling vs. running (Rahimi et al., 2006 ; Maeder et al., 2009 ) and cycling vs. arm-cranking (Ranadive et al., 2011 ). These studies also utilized maximal incremental exercise. To our knowledge, HRV recovery has not been investigated following different exercise modalities utilizing submaximal intensity.

Reliability of HRV during post-exercise recovery

During acute post-exercise recovery, the few studies investigating test-retest reliability have reporting conflicting results. Some studies report moderate to good relative reliability, such as ICC-values of 0.58–0.91 during active recovery (Boullosa et al., 2014 ) and 0.69–0.92 during static recovery (Carrasco et al., 2003 ) for various HRV measures. Alternatively, Dupuy et al. ( 2012 ) reported ICC-values ranging from 0.14 to 0.97 for during static recovery. Absolute reliability has also been reported to vary markedly, with TEM(CV) reported as 19–27% (Boullosa et al., 2014 ) and 8–65% (Al Haddad et al., 2011 ), while Dupuy et al. ( 2012 ) reported CVs ranging from of 27 to 141%. Considering the methodological differences amongst these few studies, it is difficult to draw any conclusions regarding the test-retest reliability of HRV during recovery.

Summary—HRV during post-exercise recovery

A large body of research indicates that, as is the case during exercise, exercise intensity is a primary determinant of the immediate post-exercise recovery of HRV. Upon exercise cessation, HRV measures demonstrate a time-dependent recovery that is usually (though not always) slowed following a greater preceding exercise intensity. Thus, the intensity-dose response on HRV during recovery has not been clearly elucidated.

It is also not yet clear how exercise duration affects HRV during post-exercise recovery. While three studies report that a 100% increase in exercise duration does not alter HRV during immediate recovery, one study found that HRV recovery was slowed following a 300–400% increase in exercise duration (from ~20 to ~90 min). This may suggest that exercise duration must be prolonged beyond some critical length (either relative or absolute) before an effect on HRV recovery might be observed, however this remains speculative. Additionally, it is not clear how preceding exercise intensity and duration may interact to influence post-exercise HRV recovery. Finally, the two studies identified that investigated the effect of exercise modality on post-exercise HRV suggest that a greater active muscle mass and/or energy expenditure is associated with a slower HRV recovery, at least following maximal incremental exercise. Clearly, additional studies are needed to further elucidate the influences of exercise duration and modality on post-exercise HRV responses.

Systolic time intervals

Although some measures of HRV (namely cPNA-HRV measures, i.e., RMSSD, HF, and SD1) are generally accepted to provide insight into cardiac parasympathetic modulation, HRV is not widely considered to reflect cardiac sympathetic activity (notwithstanding the controversial use of LF, LF-nu, and particularly LF:HF as indicators of sympathetic activity or “sympatho-vagal balance”). This represents an important gap in the measurement of cardiac autonomic activity, in particular because; (a) sympathetic hyperactivity is associated with increased risk of morbidity/mortality (Leenen, 1999 ; Mancia et al., 1999 ; Licht et al., 2010 ; Schwartz and De Ferrari, 2011 ; Shanks and Herring, 2013 ; Vink et al., 2013 ), and (b) as highlighted previously, there exists complex cardiac sympathetic-parasympathetic interactions, such that it may be difficult to interpret measures of one arm of autonomic activity in the absence of any information of the activity of the other arm. Systolic time intervals (STI) are another class of non-invasive measures that may provide valuable insight into cardiac sympathetic activity to complement cPNA-HRV measures.

Physiological interpretation of STI

STI outcomes—such as the pre-ejection period (PEP), left ventricular ejection time (LVET), and PEP-to-LVET ratio (PEP:LVET)—are measures of cardiac performance that may be non-invasively measured using techniques such as bio-impedance cardiography. In particular, PEP may be used to assess cardiac sympathetic β-adrenergic activity (Ahmed et al., 1972 ; Sherwood et al., 1990 ; Cacioppo et al., 1994 ). The physiological rationale derives from sympathetic activity eliciting positive inotropic effects on the ventricular myocardium, thus increasing cardiac contractility and resulting in a more rapid development of force and intraventricular pressure. This decreases the time required to reach aortic pressure (the isovolumetric contraction time, IVCT) and therefore opening of the aortic valve, thereby attenuating PEP (hence PEP is inversely associated with cardiac sympathetic activity). In contrast to being richly innervated by cardiac sympathetic neurons, the ventricles are not extensively innervated by parasympathetic neurons, meaning that changes in contractility indices such as PEP are largely attributed to changes in cardiac sympathetic activity. Indeed, sympathetic stimulation by β-adrenergic agonists shortens PEP (Harris et al., 1967 ; Ahmed et al., 1972 ; Schachinger et al., 2001 ), whereas this response is reduced or abolished under conditions of β-adrenergic blockade (Harris et al., 1967 ; Benschop et al., 1994 ; Cacioppo et al., 1994 ). PEP:LVET has also been utilized as a measure of cardiac contractility, however LVET and PEP:LVET are influenced by the underlying HR (Cokkinos et al., 1976 ; Cacioppo et al., 1994 ) as well as parasympathetic activity (Cacioppo et al., 1994 ). In contrast, PEP is not substantially altered by changes in HR (Harris et al., 1967 ; Cokkinos et al., 1976 ). Thus, while it is widely acknowledged that LVET and PEP:LVET should be corrected for the underlying HR, the need for HR-correction of PEP (Weissler et al., 1968 ; Lewis et al., 1982 ) is contentious (Cokkinos et al., 1976 ; Spodick et al., 1984 ; Rousson et al., 1987 ; Cacioppo et al., 1994 ). As a result, PEP appears to be the STI measure of choice for reflecting cardiac sympathetic β-adrenergic activity (Rousson et al., 1987 ; Cacioppo et al., 1994 ).

The interpretation of PEP as reflecting cardiac sympathetic activity is not without limitations. In particular, non-sympathetic influences which may alter PEP need to be appreciated. Although the parasympathetic influence on the ventricles is considered to be weak, it is possible that parasympathetic outflow may inhibit positive inotropic sympathetic effects when sympathetic outflow is high (Levy, 1971 ; Azevedo and Parker, 1999 ; Figure ​ Figure1). 1 ). Another confounding factor relates to the fact that PEP encompasses the electromechanical delay as well as the IVCT. However, ~75% of the shortening of PEP in response to β-adrenergic agonists is due to shorting of IVCT (Harris et al., 1967 ), while changes during exercise and recovery are almost entirely accounted for by IVCT (Nandi and Spodick, 1977 ). In addition to these factors, cardiac loading may elicit non-sympathetic influences in PEP, in particular during exercise and post-exercise recovery. Namely, the Frank-Starling mechanism (length-dependent increase in cardiac contractile force, i.e., preload) and/or a decrease in aortic pressure (afterload) may shorten PEP (Lewis et al., 1977 , 1982 ; Buch et al., 1980 ; Joubert and Belz, 1987 ). However, the extent to which preload and afterload actually change during exercise and recovery is debated (Plotnick et al., 1986 ; Kimball et al., 1993 ; Warburton et al., 2002 ; Rowland, 2008 ; La Gerche and Gewillig, 2010 ), as is the extent to which PEP reflecting cardiac sympathetic influences may be confounded (Nandi and Spodick, 1977 ; Cousineau et al., 1978 ; Lewis et al., 1982 ; Obrist et al., 1987 ). Future research is needed to clarify the extent to which these potential confounding factors influence PEP during exercise and recovery.

There are also methodological issues to consider. For example, bio-impedance cardiography is highly sensitive to movement artifact, which can make reliable signal acquisition difficult during exercise involving substantial upper body movement. Furthermore, postural changes as well as large and rapid thoracic movements (e.g., heavy breathing during hard exercise) can also influence thoracic impedance and make waveform identification and interpretation difficult.

STI during exercise and recovery

The response of PEP to exercise is generally consistent with what would be expected of a true (inverse) cardiac sympathetic indicator, i.e., an intensity-dependent decrease from rest to exercise (Ahmed et al., 1972 ; Nandi and Spodick, 1977 ; Miyamoto et al., 1983 ; Miles et al., 1984 ; Smith et al., 1989 ). PEP has rarely been investigated during post-exercise recovery. Nandi and Spodick ( 1977 ) demonstrated that the gradual recovery of PEP during the initial 5 min post-exercise period is intensity dependent across a range of absolute submaximal intensities (50, 100, and 150 W), i.e., greater intensity elicited a slower recovery. PEP:LVET has also been reported to be lower following maximal (compared with submaximal) exercise (Crisafulli et al., 2006b ), but not following different submaximal exercise intensities (Nandi and Spodick, 1977 ; Crisafulli et al., 2006b ). Regardless, the response of PEP during exercise is consistent with our understanding of intensity-dependent sympathetic activation. Very limited data suggests that post-exercise PEP recovery is also consistent with intensity-dependent sympathetic withdrawal.

No study has investigated the effect of exercise duration on STI outcomes either during exercise or post-exercise recovery. Regarding the effect of exercise modality, inspection of the data reported by Miles et al. ( 1984 ) suggests that PEP may be lower during arm-crank exercise compared with cycling for a HR range of ~80 to ~150 b.min −1 , however this is not clear due to different exercising heart rates. Alterations in posture (e.g., supine vs. upright cycling) would be expected to alter cardiac loading (in particular preload). Consistent with this, PEP during supine (compared with upright) cycling was reported to be lower at the same absolute intensities of 50 and 100 W, despite a lower HR at 100 W (Miyamoto et al., 1983 ), although a lack of any significant effect of posture on PEP during exercise has also been reported (Smith et al., 1989 ). Additionally, Crisafulli et al. ( 2006a , 2008 ) demonstrated that post-exercise muscle occlusion elicits an intensity-dependent delay in STI recovery. As for cPNA-HRV measures of post-exercise parasympathetic reactivation (see previous section), these findings highlight the important role that the muscle metaboreflex likely plays in regulating the post-exercise recovery of STI measures reflecting sympathetic withdrawal.

Summary—systolic time intervals

STI measures (in particular PEP) are a useful non-invasive indicator of cardiac sympathetic activity. As is the case for HRV, STI measures reflect the integrated end-organ response, in this case indirectly assessing inotropic cardiac effects that are understood to be under strong sympathetic influence. Accordingly, interpretative caveats/limitations relating to non-sympathetic influences (particularly potential cardiac loading effects, i.e., preload and afterload) need to be appreciated. Notwithstanding these limitations, the measurement of PEP likely provides valuable non-invasive insights into cardiac autonomic regulation, as PEP is strongly (inversely) associated with cardiac sympathetic activity.

The effect of different exercise dosages on the response of PEP (or other STI outcomes) during exercise has not been extensively investigated. Nevertheless, available data indicates that the (inverse) response of PEP is consistent with our understanding of sympathetic activity during exercise and recovery. Additional studies are needed to confirm this, particularly during the recovery period. It is not known how exercise duration influences PEP responses to exercise and recovery, nor is it well established how these responses are altered by different exercise modalities. Additionally, the test-retest reliability of these measures during and following exercise requires further research.

Overall summary and future perspectives

HRV is used as a non-invasive tool to monitor cardiac autonomic activity, with a wide range of quantitative approaches utilized. In particular, a large body of evidence indicates that cPNA-HRV measures (e.g., RMSSD, SD1, and HF) generally reflect cardiac parasympathetic modulation, although not without limitations. In contrast, the majority of evidence does not support the interpretation of HRV measures (in particular LF, LF-nu, and LF:HF) as reflecting sympathetic activity or “sympatho-vagal balance.”

Exercise elicits substantial changes in HRV measures, and several studies have investigated HRV during exercise and immediate post-exercise recovery. As is the case for studies investigating resting HRV measures, HRV analysis methodology varies widely. Regarding exercise and recovery, this issue is compounded by the wide range of exercise and recovery protocols employed. When combining these factors, very few (if any) studies investigating HRV during exercise and/or recovery are directly comparable, particularly when frequency-domain HRV measures are utilized. Additionally, some caution is advised when interpreting HRV measures calculated using very short epochs (e.g., 30 s) during non-stationary conditions (particularly during immediate post-exercise recovery), as non-oscillatory changes in HR may contribute to HRV. Nevertheless, a review of the literature concerning the effects of the primary exercise dosage factors for dynamic exercise (intensity, duration, and modality) reveals some noteworthy responses.

The literature indicates that the intensity of exercise is the primary exercise dose factor determining HRV responses during both exercise and post-exercise recovery. Most HRV measures demonstrate a reduction upon the initiation of exercise as well as an intensity-dependent decay toward near-zero levels. The intensity at which a near-zero minimum occurs depends upon the type of HRV outcome with cPNA-HRV measures typically reaching a minimum at 50–60% VO 2 max or around the first ventilation threshold. Normalized and ratio frequency-domain measures demonstrate inconsistent response during exercise that are not consistent with known aspects of cardiac autonomic activity. Upon exercise cessation, HRV recovers as a function of time, with a higher exercise intensity often (though not always) associated with a delayed recovery profile. Regarding the reliability of HRV during exercise and recovery, the few studies to date that have assessed test-retest reliability have reported mixed findings.

There is limited data on the effects of duration or modality on HRV during exercise and recovery. Methodologically, such investigations are difficult because of the close relationship between HR and HRV. Prolonged exercise duration has been associated with decreased HRV during exercise only when accompanied by a concomitant rise in HR. Prolonged exercise duration (100% increase) has been shown to elicit no significant influence on post-exercise HRV recovery, although a delayed recovery was reported when duration was prolonged by over 300% (e.g., from ~20 to ~90 min). Regarding modality, some studies report that the mode of exercise can alter the HRV response (even when HR is matched), although other studies have reported no significant modality effect. During recovery, very limited research suggests that modalities utilizing greater active muscle mass (and/or eliciting greater energy expenditure) are associated with slower HRV recovery, although this is preliminary. Potential intensity-duration and intensity-modality interactions on HRV during exercise have not been elucidated. Further research is required before conclusions on the effect of exercise duration and modality on HRV during exercise and recovery can be made.

Additional research is also needed regarding the effect of exercise dosage on non-linear HRV measures. Considering that HR is non-stationary during key segments of exercise and recovery, non-linear measures may be particularly useful for investigating HRV during these periods. Furthermore, the monitoring (prognostic/diagnostic) value of these measures requires further investigation.

Since HRV measures are not widely accepted to reflect cardiac sympathetic activity, it is noteworthy that STI indices likely provide useful non-invasive insights into cardiac performance. In particular, PEP is a valid (inverse) measure of cardiac sympathetic activity, notwithstanding some interpretative limitations. While a limited body of research indicates that the response of PEP to exercise and recovery is generally consistent with our understanding of cardiac sympathetic influences, the effects of different exercise dosages is not clear, particularly during the recovery period. Furthermore, the test-retest reliability of STI responses to exercise has not been established.

Finally, while HRV and STI reflect different aspects of cardiac function (chronotropic vs. inotropic), STI measures of cardiac sympathetic activity (PEP) may be used to complement HRV measures of cardiac parasympathetic activity (RMSSD, HF, and SD1) to provide a more comprehensive insight into cardiac autonomic regulation. Previous investigations have utilized this integrative approach under conditions of rest and psychological stressors (Berntson et al., 1994 , 2008 ; Brindle et al., 2014 ). However, concurrent monitoring of the exercise stress-reactivity of HRV indices for cPNA and STI indices for cSNA during exercise and post-exercise recovery has not been reported. This integrative approach warrants further investigation.

Author contributions

SM was the primary author who was responsible for the design and writing of the manuscript. KG and GD assisted with writing and editing the manuscript.

Conflict of interest statement

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

Acknowledgments

The authors wish to thank Dr. Ollie Jay for his assistance in the preparation of this manuscript.

• Ahmadian M., Roshan V. D., Hosseinzadeh M. (2015). Parasympathetic reactivation in children: influence of two various modes of exercise . Clin. Auton. Res. 25 , 207–212. 10.1007/s10286-015-0289-7 [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Ahmed S. S., Schwartz C. J., Levinson G. E., Ettinger P. O. (1972). Systolic time intervals as measures of contractile state of left ventricular myocardium in man . Circulation 46 , 559–571. 10.1161/01.CIR.46.3.559 [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Akselrod S., Gordon D., Madwed J. B., Snidman N. C., Shannon D. C., Cohen R. J. (1985). Hemodynamic regulation - investigation by spectral analysis . Am. J. Physiol. 249 , H867–H875. [ PubMed ] [ Google Scholar ]
• Al Haddad H., Laursen P. B., Ahmaidi S., Buchheit M. (2010). Influence of cold water face immersion on post-exercise parasympathetic reactivation . Eur. J. Appl. Physiol. 108 , 599–606. 10.1007/s00421-009-1253-9 [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Al Haddad H., Laursen P. B., Chollet D., Ahmaidi S., Buchheit M. (2011). Reliability of resting and postexercise heart rate measures . Int. J. Sports Med. 32 , 598–605. 10.1055/s-0031-1275356 [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Arai Y., Saul J. P., Albrecht P., Hartley L. H., Lilly L. S., Cohen R. J., et al.. (1989). Modulation of cardiac autonomic activity during and immediately after exercise . Am. J. Physiol. 256 , H132–H141. [ PubMed ] [ Google Scholar ]
• Avery N. D., Wolfe L. A., Amara C. E., Davies G. A. L., Mcgrath M. J. (2001). Effects of human pregnancy on cardiac autonomic function above and below the ventilatory threshold . J. Appl. Physiol. 90 , 321–328. [ PubMed ] [ Google Scholar ]
• Azevedo E. R., Parker J. D. (1999). Parasympathetic control of cardiac sympathetic activity - normal ventricular function versus congestive heart failure . Circulation 100 , 274–279. 10.1161/01.CIR.100.3.274 [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Barak O. F., Jakovljevic D. G., Gacesa J. Z. P., Ovcin Z. B., Brodie D. A., Grujic N. G. (2010). Heart rate variability before and after cycle exercise in relation to different body positions . J. Sports Sci. Med. 9 , 176–182. [ PMC free article ] [ PubMed ] [ Google Scholar ]
• Baumert M., Lambert G. W., Dawood T., Lambert E. A., Esler M. D., Mcgrane M., et al.. (2009). Short-term heart rate variability and cardiac norepinephrine spillover in patients with depression and panic disorder . Am. J. Physiol. Heart Circ. Physiol. 297 , H674–H679. 10.1152/ajpheart.00236.2009 [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Benschop R. J., Nieuwenhuis E. E. S., Tromp E. A. M., Godaert G. L. R., Ballieux R. E., Vandoornen L. J. P. (1994). Effects of beta-adrenergic blockade on immunological and cardiovascular changes induced by mental stress . Circulation 89 , 762–769. 10.1161/01.CIR.89.2.762 [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Bernardi L., Salvucci F., Suardi R., Solda P. L., Calciati A., Perlini S., et al.. (1990). Evidence for an intrinsic mechanism regulating heart rate variability in the transplanted and the intact heart during submaximal dynamic exercise . Cardiovasc. Res. 24 , 969–981. 10.1093/cvr/24.12.969 [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Berntson G. G., Cacioppo J. T., Binkley P. F., Uchino B. N., Quigley K. S., Fieldstone A. (1994). Autonomic cardiac control .3. Psychological stress and cardiac response in autonomic space as revealed by pharmacological blockades . Psychophysiology 31 , 599–608. 10.1111/j.1469-8986.1994.tb02352.x [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Berntson G. G., Norman G. J., Hawkley L. C., Cacioppo J. T. (2008). Cardiac autonomic balance versus cardiac regulatory capacity . Psychophysiology 45 , 643–652. 10.1111/j.1469-8986.2008.00652.x [ PMC free article ] [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Billman G. E. (2006). Point: counterpoint: cardiovascular variability is/is not an index of autonomic control of circulation . J. Appl. Physiol. 101 , 684–685. 10.1152/japplphysiol.00562.2006 [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Billman G. E. (2013a). The effect of heart rate on the heart rate variability response to autonomic interventions . Front. Physiol. 4 :222. 10.3389/fphys.2013.00222 [ PMC free article ] [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Billman G. E. (2013b). The LF/HF ratio does not accurately measure cardiac sympatho-vagal balance . Front. Physiol. 4 :26 10.3389/fphys.2013.00026 [ PMC free article ] [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Boettger S., Puta C., Yeragani V. K., Donath L., Muller H. J., Gabriel H. H. W., et al.. (2010). Heart rate variability, QT variability, and electrodermal activity during exercise . Med. Sci. Sports Exerc. 42 , 443–448. 10.1249/MSS.0b013e3181b64db1 [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Borresen J., Lambert M. I. (2008). Autonomic control of heart rate during and after exercise - measurements and implications for monitoring training status . Sports Med. 38 , 633–646. 10.2165/00007256-200838080-00002 [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Botek M., Stejskal P., Krejci J., Jakubec A., Gaba A. (2010). Vagal threshold determination. Effect of age and gender . Int. J. Sports Med. 31 , 768–772. 10.1055/s-0030-1263141 [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Boullosa D. A., Barros E. S., del Rosso S., Nakamura F. Y., Leicht A. S. (2014). Reliability of heart rate measures during walking before and after running maximal efforts . Int. J. Sports Med. 35 , 999–1005. 10.1055/s-0034-1372637 [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Boullosa D. A., Nakamura F. Y., Perandini L. A., Leicht A. S. (2012). Autonomic correlates of Yo-Yo performance in soccer referees . Motriz Rev. Educacao Fisica 18 , 291–297. 10.1590/S1980-65742012000200009 [ CrossRef ] [ Google Scholar ]
• Brennan M., Palaniswami M., Kamen P. (2001). Do existing measures of Poincaré plot geometry reflect nonlinear features of heart rate variability? IEEE Trans. Biomed. Eng. 48 , 1342–1347. 10.1109/10.959330 [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Brindle R. C., Ginty A. T., Phillips A. C., Carroll D. (2014). A tale of two mechanisms: a meta-analytic approach toward understanding the autonomic basis of cardiovascular reactivity to acute psychological stress . Psychophysiology 51 , 964–976. 10.1111/psyp.12248 [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Buch J., Egeblad H., Hansen P. B., Kjaergard H., Waldorff S., Steiness E. (1980). Correlation between changes in systolic time intervals and left ventricular end-diastolic diameter after preload reduction - non-invasive monitoring of pharmacological intervention . Br. Heart J. 44 , 668–671. 10.1136/hrt.44.6.668 [ PMC free article ] [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Buchheit M., Al Haddad H., Laursen P. B., Ahmaidi S. (2009a). Effect of body posture on postexercise parasympathetic reactivation in men . Exp. Physiol. 94 , 795–804. 10.1113/expphysiol.2009.048041 [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Buchheit M., Al Haddad H., Millet G. P., Lepretre P. M., Newton M., Ahmaidi S. (2009b). Cardiorespiratory and cardiac autonomic responses to 30-15 intermittent fitness test in team sport players . J. Strength Cond. Res. 23 , 93–100. 10.1519/JSC.0b013e31818b9721 [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Buchheit M., Laursen P. B., Ahmaidi S. (2007a). Parasympathetic reactivation after repeated sprint exercise . Am. J. Physiol. Heart Circ. Physiol. 293 , H133–H141. 10.1152/ajpheart.00062.2007 [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Buchheit M., Solano R., Millet G. P. (2007b). Heart-rate deflection point and the second heart-rate variability threshold during running exercise in trained boys . Pediatr. Exerc. Sci. 19 , 192–204. 10.1123/pes.19.2.192 [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Cacioppo J. T., Berntson G. G., Binkley P. F., Quigley K. S., Uchino B. N., Fieldstone A. (1994). Autonomic cardiac control .2. Noninvasive indexes and basal response as revealed by autonomic blockades . Psychophysiology 31 , 586–598. 10.1111/j.1469-8986.1994.tb02351.x [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Camm A. J., Malik M., Bigger J. T., Breithardt G., Cerutti S., Cohen R. J., et al. (1996). Heart rate variability - standards of measurement, physiological interpretation, and clinical use . Circulation 93 , 1043–1065. 10.1161/01.CIR.93.5.1043 [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Carrasco S., González R., Gaitán M. J., Yáñez O. (2003). Reproducibility of heart rate variability from short-term recordings during five manoeuvres in normal subjects . J. Med. Eng. Technol. 27 , 241–248. 10.1080/0309190031000111380 [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Casadei B., Cochrane S., Johnston J., Conway J., Sleight P. (1995). Pitfalls in the interpretation of spectral analysis of the heart rate variability during exercise in humans . Acta Physiol. Scand. 153 , 125–131. 10.1111/j.1748-1716.1995.tb09843.x [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Casadei B., Moon J., Johnston J., Caiazza A., Sleight P. (1996). Is respiratory sinus arrhythmia a good index of cardiac vagal tone in exercise? J. Appl. Physiol. 81 , 556–564. [ PubMed ] [ Google Scholar ]
• Casonatto J., Tinucci T., Dourado A. C., Polito M. (2011). Cardiovascular and autonomic responses after exercise sessions with different intensities and durations . Clinics 66 , 453–458. 10.1590/S1807-59322011000300016 [ PMC free article ] [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Casties J. F., Mottet D., Le Gallais D. (2006). Non-linear analyses of heart rate variability during heavy exercise and recovery in cyclists . Int. J. Sports Med. 27 , 780–785. 10.1055/s-2005-872968 [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Cheng C.-F., Lin H.-M., Tsai H.-C., Chu C.-H., Lin J.-C. (2005). Analysis of heart rate variability during acute exposure to moderate altitude and rowing exercise . J. Exerc. Sci. Fit. 3 , 25–32. [ Google Scholar ]
• Cokkinos D. V., Heimonas E. T., Demopoulos J. N., Haralambakis A., Tsartsalis G., Gardikas C. D. (1976). Influence of heart rate increase on uncorrected pre-ejection period left ventricular ejection time (PEP-LVET) ratio in normal individuals . Br. Heart J. 38 , 683–688. 10.1136/hrt.38.7.683 [ PMC free article ] [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Cole C. R., Blackstone E. H., Pashkow F. J., Snader C. E., Lauer M. S. (1999). Heart-rate recovery immediately after exercise as a predictor of mortality . N. Engl. J. Med. 341 , 1351–1357. 10.1056/NEJM199910283411804 [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Coote J. H. (2010). Recovery of heart rate following intense dynamic exercise . Exp. Physiol. 95 , 431–440. 10.1113/expphysiol.2009.047548 [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Cottin F., Durbin F., Papelier Y. (2004). Heart rate variability during cycloergometric exercise or judo wrestling eliciting the same heart rate level . Eur. J. Appl. Physiol. 91 , 177–184. 10.1007/s00421-003-0969-1 [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Cottin F., Lepretre P. M., Lopes P., Papelier Y., Medigue C., Billat V. (2006). Assessment of ventilatory thresholds from heart rate variability in well-trained subjects during cycling . Int. J. Sports Med. 27 , 959–967. 10.1055/s-2006-923849 [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Cottin F., Medigue C., Lopes P., Lepretre P. M., Heubert R., Billat V. (2007). Ventilatory thresholds assessment from heart rate variability during an incremental exhaustive running test . Int. J. Sports Med. 28 , 287–294. 10.1055/s-2006-924355 [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Cousineau D., Lapointe L., Dechamplain J. (1978). Circulating catecholamines and systolic-time intervals in normotensive and hypertensive patients with and without left-ventricular hypertrophy . Am. Heart J. 96 , 227–234. 10.1016/0002-8703(78)90090-X [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Coyle E. F., Gonzalez-Alonso J. (2001). Cardiovascular drift during prolonged exercise: new perspectives . Exerc. Sport Sci. Rev. 29 , 88–92. 10.1097/00003677-200104000-00009 [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Crisafulli A., Milia R., Lobina A., Caddeo M., Tocco F., Concu A., et al.. (2008). Haemodynamic effect of metaboreflex activation in men after running above and below the velocity of the anaerobic threshold . Exp. Physiol. 93 , 447–457. 10.1113/expphysiol.2007.041863 [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Crisafulli A., Salis E., Pittau G., Lorrai L., Tocco F., Melis F., et al.. (2006a). Modulation of cardiac contractility by muscle metaboreflex following efforts of different intensities in humans . Am. J. Physiol. Heart Circ. Physiol. 291 , H3035–H3042. 10.1152/ajpheart.00221.2006 [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Crisafulli A., Tocco F., Pittau G., Lorrai L., Porru C., Salis E., et al.. (2006b). Effect of differences in post-exercise lactate accumulation in athletes' haemodynamics . Appl. Physiol. Nutr. Metab. 31 , 423–431. 10.1139/h06-017 [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Cunha F. A., Midgley A. W., Goncalves T., Soares P. P., Farinatti P. (2015). Parasympathetic reactivation after maximal CPET depends on exercise modality and resting vagal activity in healthy men . SpringerPlus 4 :100. 10.1186/s40064-015-0882-1 [ PMC free article ] [ PubMed ] [ CrossRef ] [ Google Scholar ]
• D'Agosto T., Peçanha T., Bartels R., Moreira D. N., Silva L. P., Nóbrega A. C. L., et al.. (2014). Cardiac autonomic responses at onset of exercise: effects of aerobic fitness . Int. J. Sports Med. 35 , 879–885. 10.1055/s-0034-1370911 [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Daanen H. A. M., Lamberts R. P., Kallen V. L., Jin A., Van Meeteren N. L. U. (2012). A systematic review on heart-rate recovery to monitor changes in training status in athletes . Int. J. Sports Physiol. Perform. 7 , 251–260. 10.1123/ijspp.7.3.251 [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Dewland T. A., Androne A. S., Lee F. A., Lampert R. J., Katz S. D. (2007). Effect of acetylcholinesterase inhibition with pyridostigmine on cardiac parasympathetic function in sedentary adults and trained athletes . Am. J. Physiol. Heart Circ. Physiol. 293 , H86–H92. 10.1152/ajpheart.01339.2006 [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Di Michele R., Gatta G., Di Leo A., Cortesi M., Andina F., Tam E., et al.. (2012). Estimation of the anaerobic threshold from heart rate variability in an incremental swimming test . J. Strength Cond. Res. 26 , 3059–3066. 10.1519/JSC.0b013e318245bde1 [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Di Rienzo M., Castiglioni P., Iellamo F., Volterrani M., Pagani M., Mancia G., et al.. (2008). Dynamic adaptation of cardiac baroreflex sensitivity to prolonged exposure to microgravity: data from a 16-day spaceflight . J. Appl. Physiol. 105 , 1569–1575. 10.1152/japplphysiol.90625.2008 [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Dixon E. M., Kamath M. V., Mccartney N., Fallen E. L. (1992). Neural regulation of heart rate variability in endurance athletes and sedentary controls . Cardiovasc. Res. 26 , 713–719. 10.1093/cvr/26.7.713 [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Dupuy O., Mekary S., Berryman N., Bherer L., Audiffren M., Bosquet L. (2012). Reliability of heart rate measures used to assess post-exercise parasympathetic reactivation . Clin. Physiol. Funct. Imaging 32 , 296–304. 10.1111/j.1475-097X.2012.01125.x [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Eckberg D. L. (1997). Sympathovagal balance - A critical appraisal . Circulation 96 , 3224–3232. 10.1161/01.CIR.96.9.3224 [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Eckberg D. L. (2003). The human respiratory gate . J. Physiol. 548 , 339–352. 10.1113/jphysiol.2002.037192 [ PMC free article ] [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Fadel P. J. (2015). Reflex control of the circulation during exercise . Scand. J. Med. Sci. Sports 25 , 74–82. 10.1111/sms.12600 [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Fisher J. P., Ogoh S., Junor C., Khaja A., Northrup M., Fadel P. J. (2009). Spontaneous baroreflex measures are unable to detect age-related impairments in cardiac baroreflex function during dynamic exercise in humans . Exp. Physiol. 94 , 447–458. 10.1113/expphysiol.2008.044867 [ PMC free article ] [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Fisher J. P., Young C. N., Fadel P. J. (2015). Autonomic adjustments to exercise in humans . Compr. Physiol. 5 , 475–512. 10.1002/cphy.c140022 [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Francis D. P., Willson K., Georgiadou P., Wensel R., Davies L. C., Coats A., et al.. (2002). Physiological basis of fractal complexity properties of heart rate variability in man . J. Physiol. 542 , 619–629. 10.1113/jphysiol.2001.013389 [ PMC free article ] [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Furlan R., Piazza S., Orto S. D., Gentile E., Cerutti S., Pagani M., et al.. (1993). Early and late effects of exercise and athletic training on neural mechanisms controlling heart-rate . Cardiovasc. Res. 27 , 482–488. 10.1093/cvr/27.3.482 [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Garcia-Tabar I., Sánchez-Medina L., Aramendi J. F., Ruesta M., Ibañez J., Gorostiaga E. M. (2013). Heart rate variability thresholds predict lactate thresholds in professional world-class road cyclists . J. Exerc. Physiol. Online 16 , 38–50. [ Google Scholar ]
• Gladwell V. F., Sandercock G. R. H., Birch S. L. (2010). Cardiac vagal activity following three intensities of exercise in humans . Clin. Physiol. Funct. Imaging 30 , 17–22. 10.1111/j.1475-097X.2009.00899.x [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Goldberger J. J., Le F. K., Lahiri M., Kannankeril P. J., Ng J., Kadish A. H. (2006). Assessment of parasympathetic reactivation after exercise . Am. J. Physiol. Heart Circ. Physiol. 290 , H2446–H2452. 10.1152/ajpheart.01118.2005 [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Gonzalez-Camarena R., Carrasco-Sosa S., Roman-Ramos R., Gaitan-Gonzalez M. J., Medina-Banuelos V., Azpiroz-Leehan J. (2000). Effect of static and dynamic exercise on heart rate and blood pressure variabilities . Med. Sci. Sports Exerc. 32 , 1719–1728. 10.1097/00005768-200010000-00010 [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Harris W. S., Schoenfeld C. D., Weissler A. M. (1967). Effects of adrenergic receptor activation and blockade on systolic preejection period heart rate and arterial pressure in man . J. Clin. Invest. 46 , 1704–1714. 10.1172/JCI105661 [ PMC free article ] [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Hautala A. J., Makikallio T. H., Seppanen T., Huikuri H. V., Tulppo M. P. (2003). Short-term correlation properties of R-R interval dynamics at different exercise intensity levels . Clin. Physiol. Funct. Imaging 23 , 215–223. 10.1046/j.1475-097X.2003.00499.x [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Hautala A., Tulppo M. P., Makikallio T. H., Laukkanen R., Nissila S., Huikuri H. V. (2001). Changes in cardiac autonomic regulation after prolonged maximal exercise . Clin. Physiol. 21 , 238–245. 10.1046/j.1365-2281.2001.00309.x [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Hayashi N., Nakamura Y., Muraoka I. (1992). Cardiac autonomic regulation after moderate and exhaustive exercises . Ann. Physiol. Anthropol. 11 , 333–338. 10.2114/ahs1983.11.333 [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Heponiemi T., Elovainio M., Pulkki L., Puttonen S., Raitakari O., Keltikangas-Järvinen L. (2007). Cardiac autonomic reactivity and recovery in predicting carotid atherosclerosis: the cardiovascular risk in young Finns study . Health Psychol. 26 , 13–21. 10.1037/0278-6133.26.1.13 [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Houle M. S., Billman G. E. (1999). Low-frequency component of the heart rate variability spectrum: a poor marker of sympathetic activity . Am. J. Physiol. Heart Circ. Physiol. 276 , H215–H223. [ PubMed ] [ Google Scholar ]
• Huikuri H. V., Makikallio T. H., Perkiomaki J. (2003). Measurement of heart rate variability by methods based on nonlinear dynamics . J. Electrocardiol. 36 , 95–99. 10.1016/j.jelectrocard.2003.09.021 [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Huikuri H. V., Perkiomaki J. S., Maestri R., Pinna G. D. (2009). Clinical impact of evaluation of cardiovascular control by novel methods of heart rate dynamics . Philos. Trans. Ser. A Math. Phys. Eng. Sci. 367 , 1223–1238. 10.1098/rsta.2008.0294 [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Imai K., Sato H., Hori M., Kusuoka H., Ozaki H., Yokoyama H., et al.. (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. 10.1016/0735-1097(94)90150-3 [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Joubert P. H., Belz G. G. (1987). Are pre-ejection period changes specific for inotropic effects? Eur. J. Clin. Pharmacol. 33 , 335–335. [ PubMed ] [ Google Scholar ]
• Kaikkonen P., Hynynen E., Mann T., Rusko H., Nummela A. (2010). Can HRV be used to evaluate training load in constant load exercises? Eur. J. Appl. Physiol. 108 , 435–442. 10.1007/s00421-009-1240-1 [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Kaikkonen P., Hynynen E., Mann T., Rusko H., Nummela A. (2012). Heart rate variability is related to training load variables in interval running exercises . Eur. J. Appl. Physiol. 112 , 829–838. 10.1007/s00421-011-2031-z [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Kaikkonen P., Nummela A., Rusko H. (2007). Heart rate variability dynamics during early recovery after different endurance exercises . Eur. J. Appl. Physiol. 102 , 79–86. 10.1007/s00421-007-0559-8 [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Kaikkonen P., Rusko H., Martinmaki K. (2008). Post-exercise heart rate variability of endurance athletes after different high-intensity exercise interventions . Scand. J. Med. Sci. Sports 18 , 511–519. 10.1111/j.1600-0838.2007.00728.x [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Kannankeril P. J., Goldberger J. J. (2002). Parasympathetic effects on cardiac electrophysiology during exercise and recovery . Am. J. Physiol. Heart Circ. Physiol. 282 , H2091–H2098. 10.1152/ajpheart.00825.2001 [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Kannankeril P. J., Le F. K., Kadish A. H., Goldberger J. J. (2004). Parasympathetic effects on heart rate recovery after exercise . J. Invest. Med. 52 , 394–401. 10.1136/jim-52-06-34 [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Karapetian G. K., Engels H. J., Gretebeck K. A., Gretebeck R. J. (2012). Effect of Caffeine on LT, VT and HRVT . Int. J. Sports Med. 33 , 507–513. 10.1055/s-0032-1301904 [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Karapetian G. K., Engels H. J., Gretebeck R. J. (2008). Use of heart rate variability to estimate LT and VT . Int. J. Sports Med. 29 , 652–657. 10.1055/s-2007-989423 [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Karavirta L., Tulppo M. P., Laaksonen D. E., Nyman K., Laukkanen R. T., Kinnunen H., et al.. (2009). Heart rate dynamics after combined endurance and strength training in older men . Med. Sci. Sports Exerc. 41 , 1436–1443. 10.1249/MSS.0b013e3181994a91 [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Kawada T., Ikeda Y., Sugimachi M., Shishido T., Kawaguchi O., Yamazaki T., et al.. (1996). Bidirectional augmentation of heart rate regulation by autonomic nervous system in rabbits . Am. J. Physiol. Heart Circ. Physiol. 271 , H288–H295. [ PubMed ] [ Google Scholar ]
• Kikuya M., Hozawa A., Ohokubo T., Tsuji I., Michimata M., Matsubara M., et al.. (2000). Prognostic significance of blood pressure and heart rate variabilities - the Ohasama study . Hypertension 36 , 901–906. 10.1161/01.HYP.36.5.901 [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Kimball T. R., Mays W. A., Khoury P. R., Mallie R., Claytor R. P. (1993). Echocardiographic determination of left-ventricular preload, afterload, and contractility during and after exercise . J. Pediatr. 122 , S89–S94. 10.1016/S0022-3476(09)90050-6 [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Krantz D. S., Manuck S. B. (1984). Acute psychophysiologic reactivity and risk of cardiovascular-disease - a review and methodologic critique . Psychol. Bull. 96 , 435–464. 10.1037/0033-2909.96.3.435 [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Kukielka M., Seals D. R., Billman G. E. (2006). Cardiac vagal modulation of heart rate during prolonged submaximal exercise in animals with healed myocardial infarctions: effects of training . Am. J. Physiol. Heart Circ. Physiol. 290 , H1680–H1685. 10.1152/ajpheart.01034.2005 [ PubMed ] [ CrossRef ] [ Google Scholar ]
• La Gerche A., Gewillig M. (2010). What limits cardiac performance during exercise in normal subjects and in healthy fontan patients? Int. J. Pediatr. 2010 :791291. 10.1155/2010/791291 [ PMC free article ] [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Lamberts R. P., Swart J., Noakes T. D., Lambert M. I. (2009). Changes in heart rate recovery after high-intensity training in well-trained cyclists . Eur. J. Appl. Physiol. 105 , 705–713. 10.1007/s00421-008-0952-y [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Leenen F. H. H. (1999). Cardiovascular consequences of sympathetic hyperactivity . Can. J. Cardiol. 15 , 2A–7A. [ PubMed ] [ Google Scholar ]
• Leicht A. S., Sinclair W. H., Spinks W. L. (2008). Effect of exercise mode on heart rate variability during steady state exercise . Eur. J. Appl. Physiol. 102 , 195–204. 10.1007/s00421-007-0574-9 [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Levy M. N. (1971). Sympathetic-parasympathetic interactions in heart . Circ. Res. 29 , 437–445. 10.1161/01.RES.29.5.437 [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Lewis R. P., Boudoulas H., Leier C. V., Unverferth D. V., Weissler A. M. (1982). Usefulness of the systolic time intervals in cardiovascular clinical cardiology . Trans. Am. Clin. Climatol. Assoc. 93 , 108–120. [ PMC free article ] [ PubMed ] [ Google Scholar ]
• Lewis R. P., Rittgers S. E., Forester W. F., Boudoulas H. (1977). Critical-review of systolic-time intervals . Circulation 56 , 146–158. 10.1161/01.CIR.56.2.146 [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Licht C. M. M., Vreeburg S. A., Dortland A. K. B. V.R., Giltay E. J., Hoogendijk W. J. G., Derijk R. H., et al.. (2010). Increased sympathetic and decreased parasympathetic activity rather than changes in hypothalamic-pituitary-adrenal axis activity is associated with metabolic abnormalities . J. Clin. Endocrinol. Metab. 95 , 2458–2466. 10.1210/jc.2009-2801 [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Lovallo W. R. (2005). Cardiovascular reactivity: mechanisms and pathways to cardiovascular disease . Int. J. Psychophysiol. 58 , 119–132. 10.1016/j.ijpsycho.2004.11.007 [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Lunt H. C., Corbett J., Barwood M. J., Tipton M. J. (2011). Cycling cadence affects heart rate variability . Physiol. Meas. 32 , 1133–1145. 10.1088/0967-3334/32/8/009 [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Maeder M. T., Ammann P., Rickli H., Brunner-La Rocca H. P. (2009). Impact of the exercise mode on heart rate recovery after maximal exercise . Eur. J. Appl. Physiol. 105 , 247–255. 10.1007/s00421-008-0896-2 [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Makikallio T. H., Koistinen J., Jordaens L., Tulppo M. P., Wood N., Golosarsky B., et al.. (1999). Heart rate dynamics before spontaneous onset of ventricular fibrillation in patients with healed myocardial infarcts . Am. J. Cardiol. 83 , 880–884. 10.1016/S0002-9149(98)01068-6 [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Malik M., Camm A. J. (1993). Components of heart rate variability - What they really mean and what we really measure . Am. J. Cardiol. 72 , 821–822. 10.1016/0002-9149(93)91070-X [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Malliani A., Pagani M., Lombardi F., Cerutti S. (1991). Cardiovascular neural regulation explored in the frequency domain . Circulation 84 , 482–492. 10.1161/01.CIR.84.2.482 [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Mancia G., Grassi C., Giannattasio C., Seravalle G. (1999). Sympathetic activation in the pathogenesis of hypertension and progression of organ damage . Hypertension 34 , 724–728. 10.1161/01.HYP.34.4.724 [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Martinmaki K., Hakkinen K., Mikkola J., Rusko H. (2008). Effect of low-dose endurance training on heart rate variability at rest and during an incremental maximal exercise test . Eur. J. Appl. Physiol. 104 , 541–548. 10.1007/s00421-008-0804-9 [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Martinmaki K., Rusko H. (2008). Time-frequency analysis of heart rate variability during immediate recovery from low and high intensity exercise . Eur. J. Appl. Physiol. 102 , 353–360. 10.1007/s00421-008-0687-9 [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Martinmaki K., Rusko H., Saalasti S., Ketunen J. (2006). Ability of short-time Fourier transform method to detect transient changes in vagal effects on hearts: a pharmacological blocking study . Am. J. Physiol. Heart Circ. Physiol. 290 , H2582–H2589. 10.1152/ajpheart.00058.2005 [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Medigue C., Girard A., Laude D., Monti A., Wargon M., Elghozi J. L. (2001). Relationship between pulse interval and respiratory sinus arrhythmia: a time- and frequency-domain analysis of the effects of atropine . Pflugers Arch. Eur. J. Physiol. 441 , 650–655. 10.1007/s004240000486 [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Michael S., Jay O., Halaki M., Graham K., Davis G. M. (2016). Submaximal exercise intensity modulates acute post-exercise heart rate variability . Eur. J. Appl. Physiol. 116 , 697–706. 10.1007/s00421-016-3327-9 [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Michelini L. C., O'leary D. S., Raven P. B., Nobrega A. C. L. (2015). Neural control of circulation and exercise: a translational approach disclosing interactions between central command, arterial baroreflex, and muscle metaboreflex . Am. J. Physiol. Heart Circ. Physiol. 309 , H381–H392. 10.1152/ajpheart.00077.2015 [ PMC free article ] [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Miles D. S., Sawka M. N., Hanpeter D. E., Foster J. E., Doerr B. M., Frey M. A. B. (1984). Central hemodynamics during progressive upper-body and lower-body exercise and recovery . J. Appl. Physiol. 57 , 366–370. [ PubMed ] [ Google Scholar ]
• Miyamoto T., Kawada T., Takaki H., Inagaki M., Yanagiya Y., Jin Y. T., et al.. (2003). High plasma norepinephrine attenuates the dynamic heart rate response to vagal stimulation . Am. J. Physiol. Heart Circ. Physiol. 284 , H2412–H2418. 10.1152/ajpheart.00660.2002 [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Miyamoto T., Kawada T., Yanagiya Y., Inagaki M., Takaki H., Sugimachi M., et al. (2004). Cardiac sympathetic nerve stimulation does not attenuate dynamic vagal control of heart rate via alpha-adrenergic mechanism . Am. J. Physiol. Heart Circ. Physiol. 287 , H860–H865. 10.1152/ajpheart.00752.2003 [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Miyamoto Y., Higuchi J., Abe Y., Hiura T., Nakazono Y., Mikami T. (1983). Dynamics of cardiac output and systolic time intervals in supine and upright exercise . J. Appl. Physiol. 55 , 1674–1681. [ PubMed ] [ Google Scholar ]
• Montain S. J., Coyle E. F. (1992). Influence of graded dehydration on hyperthermia and cardiovascular drift during exercise . J. Appl. Physiol. 73 , 1340–1350. [ PubMed ] [ Google Scholar ]
• Moreno I. L., Pastre C. M., Ferreira C., De Abreu L. C., Valenti V. E., Marques Vanderlei L. C. (2013). Effects of an isotonic beverage on autonomic regulation during and after exercise . J. Int. Soc. Sports Nutr. 10 :2. 10.1186/1550-2783-10-2 [ PMC free article ] [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Mourot L., Bouhaddi M., Tordi N., Rouillon J. D., Regnard J. (2004). Short- and long-term effects of a single bout of exercise on heart rate variability: comparison between constant and interval training exercises . Eur. J. Appl. Physiol. 92 , 508–517. 10.1007/s00421-004-1119-0 [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Nandi P. S., Spodick D. H. (1977). Recovery from exercise at varying work loads - Time course of responses of heart-rate and systolic intervals . Br. Heart J. 39 , 958–966. 10.1136/hrt.39.9.958 [ PMC free article ] [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Ng J., Sundaram S., Kadish A. H., Goldberger J. J. (2009). Autonomic effects on the spectral analysis of heart rate variability after exercise . Am. J. Physiol. Heart Circ. Physiol. 297 , H1421–H1428. 10.1152/ajpheart.00217.2009 [ PMC free article ] [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Niewiadomski W., Gasiorowska A., Krauss B., Mroz A., Cybulski G. (2007). Suppression of heart rate variability after supramaximal exertion . Clin. Physiol. Funct. Imaging 27 , 309–319. 10.1111/j.1475-097X.2007.00753.x [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Nobrega A. C. L., O'leary D., Silva B. M., Marongiu E., Piepoli M. F., Crisafulli A. (2014). Neural regulation of cardiovascular response to exercise: role of central command and peripheral afferents . Biomed. Res. Int. 2014 :478965. 10.1155/2014/478965 [ PMC free article ] [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Obrist P. A., Light K. C., James S. A., Strogatz D. S. (1987). Cardiovascular-responses to stress .1. Measures of myocardial response and relationship to high resting systolic pressure and parental hypertension . Psychophysiology 24 , 65–78. 10.1111/j.1469-8986.1987.tb01864.x [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Oida E., Moritani T., Yamori Y. (1997). Tone-entropy analysis on cardiac recovery after dynamic exercise . J. Appl. Physiol. 82 , 1794–1801. [ PubMed ] [ Google Scholar ]
• O'leary D. S. (1993). Autonomic mechanisms of muscle metaboreflex control of heart rate . J. Appl. Physiol. 74 , 1748–1754. [ PubMed ] [ Google Scholar ]
• Pagani M., Lombardi F., Guzzetti S., Rimoldi O., Furlan R., Pizzinelli P., et al. (1986). Power spectral-analysis of heart-rate and arterial-pressure variabilities as a marker of sympathovagal interaction in man and conscious dog . Circ. Res. 59 , 178–193. 10.1161/01.RES.59.2.178 [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Parekh A., Lee C. M. (2005). Heart rate variability after isocaloric exercise bouts of different intensities . Med. Sci. Sports Exerc. 37 , 599–605. 10.1249/01.MSS.0000159139.29220.9A [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Paton J. F. R., Boscan P., Pickering A. E., Nalivaiko E. (2005). The yin and yang of cardiac autonomic control: vago-sympathetic interactions revisited . Brain Res. Rev. 49 , 555–565. 10.1016/j.brainresrev.2005.02.005 [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Pecanha T., Silva-Junior N. D., De Moraes Forjaz C. L. (2014). Heart rate recovery: autonomic determinants, methods of assessment and association with mortality and cardiovascular diseases . Clin. Physiol. Funct. Imaging 34 , 327–339. 10.1111/cpf.12102 [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Peng C. K., Havlin S., Stanley H. E., Goldberger A. L. (1995). Quantification of scaling exponents and crossover phenomena in nonstationary heartbeat time-series . Chaos 5 , 82–87. 10.1063/1.166141 [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Perini R., Milesi S., Biancardi L., Pendergast D. R., Veicsteinas A. (1998). Heart rate variability in exercising humans: effect of water immersion . Eur. J. Appl. Physiol. Occup. Physiol. 77 , 326–332. 10.1007/s004210050341 [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Perini R., Milesi S., Fisher N. M., Pendergast D. R., Veicsteinas A. (2000). Heart rate variability during dynamic exercise in elderly males and females . Eur. J. Appl. Physiol. 82 , 8–15. 10.1007/s004210050645 [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Perini R., Orizio C., Baselli G., Cerutti S., Veicsteinas A. (1990). The influence of exercise intensity on the power spectrum of heart-rate-variability . Eur. J. Appl. Physiol. Occup. Physiol. 61 , 143–148. 10.1007/BF00236709 [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Perini R., Orizio C., Comande A., Castellano M., Beschi M., Veicsteinas A. (1989). Plasma norepinephrine and heart-rate dynamics during recovery from submaximal exercise in man . Eur. J. Appl. Physiol. Occup. Physiol. 58 , 879–883. 10.1007/BF02332222 [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Perini R., Orizio C., Milesi S., Biancardi L., Baselli G., Veicsteinas A. (1993). Body position affects the power spectrum of heart-rate-variability during dynamic exercise . Eur. J. Appl. Physiol. Occup. Physiol. 66 , 207–213. 10.1007/BF00235095 [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Perini R., Tironi A., Cautero M., Di Nino A., Tam E., Capelli C. (2006). Seasonal training and heart rate and blood pressure variabilities in young swimmers . Eur. J. Appl. Physiol . 97 , 395–403. 10.1007/s00421-006-0174-0 [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Perkiomaki J. S., Makikallio T. H., Huikuri H. V. (2005). Fractal and complexity measures of heart rate variability . Clin. Exp. Hypertens. 27 , 149–158. 10.1081/CEH-48742 [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Phillips A. C. (2011). Blunted as well as exaggerated cardiovascular reactivity to stress is associated with negative health outcomes . Jpn. Psychol. Res. 53 , 177–192. 10.1111/j.1468-5884.2011.00464.x [ CrossRef ] [ Google Scholar ]
• Pichon A. P., De Bisschop C., Roulaud M., Denjean A., Papelier Y. (2004). Spectral analysis of heart rate variability during exercise in trained subjects . Med. Sci. Sports Exerc. 36 , 1702–1708. 10.1249/01.MSS.0000142403.93205.35 [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Pierpont G. L., Voth E. J. (2004). Assessing autonomic function by analysis of heart rate recovery from exercise in healthy subjects . Am. J. Cardiol. 94 , 64–68. 10.1016/j.amjcard.2004.03.032 [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Plotnick G. D., Becker L. C., Fisher M. L. (1986). Changes in left ventricular function during recovery from upright bicycle exercise in normal persons and patients with coronary artery disease . Am. J. Cardiol. 58 , 247–251. 10.1016/0002-9149(86)90056-1 [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Polanczyk C. A., Rohde L. E. P., Moraes R. S., Ferlin E. L., Leite C., Ribeiro J. P. (1998). Sympathetic nervous system representation in time and frequency domain indices of heart rate variability . Eur. J. Appl. Physiol. Occup. Physiol. 79 , 69–73. 10.1007/s004210050475 [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Pollock M. L., Gaesser G. A., Butcher J. D., Després J. P., Dishman R. K., Franklin B. A., et al. (1998). ACSM position stand: the recommended quantity and quality of exercise for developing and maintaining cardiorespiratory and muscular fitness, and flexibility in healthy adults . Med. Sci. Sports Exerc. 30 , 975–991. 10.1249/00005768-199806000-00032 [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Pomeranz B., Macaulay R. J. B., Caudill M. A., Kutz I., Adam D., Gordon D., et al.. (1985). Assessment of autonomic function in humans by heart rate spectral analysis . Am. J. Physiol. 248 , H151–H153. [ PubMed ] [ Google Scholar ]
• Potts J. T., Mitchell J. H. (1998). Rapid resetting of carotid baroreceptor reflex by afferent input from skeletal muscle receptors . Am. J. Physiol. Heart Circ. Physiol. 275 , 2000–2008. [ PubMed ] [ Google Scholar ]
• Povea C., Schmitt L., Brugniaux J., Nicolet U., Richalet J. P., Fouillot J. P. (2005). Effects of intermittent hypoxia on heart rate variability during rest and exercise . High Alt. Med. Biol. 6 , 215–225. 10.1089/ham.2005.6.215 [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Pradhapan P., Tarvainen M. P., Nieminen T., Lehtinen R., Nikus K., Lehtimaki T., et al.. (2014). Effect of heart rate correction on pre- and post-exercise heart rate variability to predict risk of mortality - an experimental study on the FINCAVAS cohort . Front. Physiol. 5 :208. 10.3389/fphys.2014.00208 [ PMC free article ] [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Radaelli A., Valle F., Falcone C., Calciati A., Leuzzi S., Martinelli L., et al.. (1996). Determinants of heart rate variability in heart transplanted subjects during physical exercise . Eur. Heart J. 17 , 462–471. 10.1093/oxfordjournals.eurheartj.a014881 [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Rahimi K., Thomas A., Adam M., Hayerizadeh B. F., Schuler G., Secknus M. A. (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. Prevent. Rehabil. 13 , 45–50. 10.1097/00149831-200602000-00007 [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Ranadive S. M., Fahs C. A., Yan H., Rossow L. M., Agliovlastis S., Fernhall B. (2011). Heart rate recovery following maximal arm and leg-ergometry . Clin. Auton. Res. 21 , 117–120. 10.1007/s10286-010-0094-2 [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Randall D. C., Brown D. R., Raisch R. M., Yingling J. D., Randall W. C. (1991). SA nodal parasympathectomy delineates autonomic control of heart rate power spectrum . Am. J. Physiol. 260 , H985–H988. [ PubMed ] [ Google Scholar ]
• Raven P. B., Fadel P. J., Ogoh S. (2006). Arterial baroreflex resetting during exercise: a current perspective . Exp. Physiol. 91 , 37–49. 10.1113/expphysiol.2005.032250 [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Robinson B. F., Epstein S. E., Beiser G. D., Braunwal E. (1966). Control of heart rate by autonomic nervous system - studies in man on interrelation between baroreceptor mechanisms and exercise . Circ. Res. 19 , 400–411. 10.1161/01.RES.19.2.400 [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Rousson D., Galleyrand J., Silie M., Boissel J. P. (1987). Uncorrected preejection period - a simple noninvasive measurement for pharmacodynamic screening of inotropic activity . Eur. J. Clin. Pharmacol. 31 , 559–562. 10.1007/BF00606630 [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Rowell L. B. (1993). Human Cardiovascular Control . New York, NY: Oxford University Press. [ Google Scholar ]
• Rowell L. B., Oleary D. S. (1990). Reflex control of the circulation during exercise: chemoreflexes and mechanoreflexes . J. Appl. Physiol. 69 , 407–418. [ PubMed ] [ Google Scholar ]
• Rowell L. B., Shepherd J. T., Society A. P. (1996). Handbook of Physiology. Section 12. Exercise: Regulation and Integration of Multiple Systems. New York, NY: Published for the American Physiological Society by Oxford University Press. [ Google Scholar ]
• Rowland T. (2008). Echocardiography and circulatory response to progressive endurance exercise . Sports Med. 38 , 541–551. 10.2165/00007256-200838070-00002 [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Sacha J. (2013). Why should one normalize heart rate variability with respect to average heart rate . Front. Physiol. 4 :306. 10.3389/fphys.2013.00306 [ PMC free article ] [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Sacha J. (2014). Interaction between heart rate and heart rate variability . Ann. Noninvasive Electrocardiol. 19 , 207–216. 10.1111/anec.12148 [ PMC free article ] [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Saito M., Nakamura Y. (1995). Cardiac autonomic control and muscle sympathetic nerve activity during dynamic exercise . Jpn. J. Physiol. 45 , 961–977. 10.2170/jjphysiol.45.961 [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Sales M. M., Campbell C. S. G., Morais P. K., Ernesto C., Soares-Caldeira L. F., Russo P., et al.. (2011). Noninvasive method to estimate anaerobic threshold in individuals with type 2 diabetes . Diabetol. Metab. Syndr. 3 :1. 10.1186/1758-5996-3-1 [ PMC free article ] [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Salvador A., Suay F., Gonzalez-Bono E., Serrano M. A. (2003). Anticipatory cortisol, testosterone and psychological responses to judo competition in young men . Psychoneuroendocrinology 28 , 364–375. 10.1016/S0306-4530(02)00028-8 [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Schachinger H., Weinbacher M., Kiss A., Ritz R., Langewitz W. (2001). Cardiovascular indices of peripheral and central sympathetic activation . Psychosom. Med. 63 , 788–796. 10.1097/00006842-200109000-00012 [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Schwartz P. J., De Ferrari G. M. (2011). Sympathetic-parasympathetic interaction in health and disease: abnormalities and relevance in heart failure . Heart Fail. Rev. 16 , 101–107. 10.1007/s10741-010-9179-1 [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Seiler S., Haugen O., Kuffel E. (2007). Autonomic recovery after exercise in trained athletes: intensity and duration effects . Med. Sci. Sports Exerc. 39 , 1366–1373. 10.1249/mss.0b013e318060f17d [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Shanks J., Herring N. (2013). Peripheral cardiac sympathetic hyperactivity in cardiovascular disease: role of neuropeptides . Am. J. Physiol. Regul. Integr. Comp. Physiol. 305 , R1411–R1420. 10.1152/ajpregu.00118.2013 [ PMC free article ] [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Sherwood A., Allen M. T., Fahrenberg J., Kelsey R. M., Lovallo W. R., Vandoornen L. J. P. (1990). Methodological guidelines for impedance cardiography . Psychophysiology 27 , 1–23. 10.1111/j.1469-8986.1990.tb02171.x [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Shibata M., Moritani T., Miyawaki T., Hayashi T., Nakao K. (2002). Exercise prescription based upon cardiac vagal activity for middle-aged obese women . Int. J. Obes. 26 , 1356–1362. 10.1038/sj.ijo.0802131 [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Smith J. J., Muzi M., Barney J. A., Ceschi J., Hayes J., Ebert T. J. (1989). Impedance-derived cardiac indexes in supine and upright exercise . Ann. Biomed. Eng. 17 , 507–515. 10.1007/BF02368070 [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Spadacini G., Passino C., Leuzzi S., Valle F., Piepoli M., Calciati A., et al.. (2006). Frequency-dependent baroreflex control of blood pressure and heart rate during physical exercise . Int. J. Cardiol. 107 , 171–179. 10.1016/j.ijcard.2005.03.011 [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Spodick D. H., Doi Y. L., Bishop R. L., Hashimoto T. (1984). Systolic-time intervals reconsidered - Reevaluation of the preejection period - absence of relation to heart-rate . Am. J. Cardiol. 53 , 1667–1670. 10.1016/0002-9149(84)90599-X [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Stanley J., Peake J. M., Buchheit M. (2013). Cardiac parasympathetic reactivation following exercise: implications for training prescription . Sports Med. 43 , 1259–1277. 10.1007/s40279-013-0083-4 [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Stein P. K., Domitrovich P. P., Huikuri H. V., Kleiger R. E., Cast I. (2005). Traditional and nonlinear heart rate variability are each independently associated with mortality after myocardial infarction . J. Cardiovasc. Electrophysiol. 16 , 13–20. 10.1046/j.1540-8167.2005.04358.x [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Sunagawa K., Kawada T., Nakahara T. (1998). Dynamic nonlinear vago-sympathetic interaction in regulating heart rate . Heart Vessels 13 , 157–174. 10.1007/BF01745040 [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Takahashi T., Okada A., Saitoh T., Hayano J., Miyamoto Y. (2000). Difference in human cardiovascular response between upright and supine recovery from upright cycle exercise . Eur. J. Appl. Physiol. 81 , 233–239. 10.1007/s004210050036 [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Tarvainen M. P., Ranta-Aho P. O., Karjalainen P. A. (2002). An advanced detrending method with application to HRV analysis . IEEE Trans. Biomed. Eng. 49 , 172–175. 10.1109/10.979357 [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Taylor J. A., Myers C. W., Halliwill J. R., Seidel H., Eckberg D. L. (2001). Sympathetic restraint of respiratory sinus arrhythmia: implications for vagal-cardiac tone assessment in humans . Am. J. Physiol. Heart Circ. Physiol. 280 , H2804–H2814. [ PubMed ] [ Google Scholar ]
• Teixeira L., Ritti-Dias R. M., Tinucci T., Mion Junior D., De Moraes Forjaz C. L. (2011). Post-concurrent exercise hemodynamics and cardiac autonomic modulation . Eur. J. Appl. Physiol. 111 , 2069–2078. 10.1007/s00421-010-1811-1 [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Terziotti P., Schena F., Gulli G., Cevese A. (2001). Post-exercise recovery of autonomic cardiovascular control: a study by spectrum and cross-spectrum analysis in humans . Eur. J. Appl. Physiol. 84 , 187–194. 10.1007/s004210170003 [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Treiber F. A., Kamarck T., Schneiderman N., Sheffield D., Kapuku G., Taylor T. (2003). Cardiovascular reactivity and development of preclinical and clinical disease states . Psychosom. Med. 65 , 46–62. 10.1097/00006842-200301000-00007 [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Tsuji H., Larson M. G., Venditti F. J., Manders E. S., Evans J. C., Feldman C. L., et al.. (1996). Impact of reduced heart rate variability on risk for cardiac events - The Framingham Heart Study . Circulation 94 , 2850–2855. 10.1161/01.CIR.94.11.2850 [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Tulppo M. P., Makikallio T. H., Laukkanen R. T., Huikuri H. V. (1999). Differences in autonomic modulation of heart rate during arm and leg exercise . Clin. Physiol. 19 , 294–299. 10.1046/j.1365-2281.1999.00180.x [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Tulppo M. P., Makikallio T. H., Seppanen T., Laukkanen R. T., Huikuri H. V. (1998). Vagal modulation of heart rate during exercise: effects of age and physical fitness . Am. J. Physiol. Heart Circ. Physiol. 274 , H424–H429. [ PubMed ] [ Google Scholar ]
• Tulppo M. P., Makikallio T. H., Takala T. E. S., Seppanen T., Huikuri H. V. (1996). Quantitative beat-to-beat analysis of heart rate dynamics during exercise . Am. J. Physiol. Heart Circ. Physiol. 271 , H244–H252. [ PubMed ] [ Google Scholar ]
• Uijtdehaage S. H. J., Thayer J. F. (2000). Accentuated antagonism in the control of human heart rate . Clin. Auton. Res. 10 , 107–110. 10.1007/BF02278013 [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Vink E. E., De Jager R. L., Blankestijn P. J. (2013). Sympathetic hyperactivity in chronic kidney disease: pathophysiology and (new) treatment options . Curr. Hypertens. Rep. 15 , 95–101. 10.1007/s11906-013-0328-5 [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Warburton D. E. R., Haykowsky M. J., Quinney H. A., Blackmore D., Teo K. K., Humen D. P. (2002). Myocardial response to incremental exercise in endurance-trained athletes: influence of heart rate, contractility and the Frank-Starling effect . Exp. Physiol. 87 , 613–622. 10.1113/eph8702372 [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Warren J. H., Jaffe R. S., Wraa C. E., Stebbins C. L. (1997). Effect of autonomic blockade on power spectrum of heart rate variability during exercise . Am. J. Physiol. Regul. Integr. Comp. Physiol. 273 , R495–R502. [ PubMed ] [ Google Scholar ]
• Weippert M., Behrens K., Rieger A., Stoll R., Kreuzfeld S. (2013). Heart rate variability and blood pressure during dynamic and static exercise at similar heart rate levels . PLoS ONE 8 :e83690. 10.1371/journal.pone.0083690 [ PMC free article ] [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Weippert M., Behrens M., Gonschorek R., Bruhn S., Behrens K. (2015). Muscular contraction mode differently affects autonomic control during heart rate matched exercise . Front. Physiol. 6 :156. 10.3389/fphys.2015.00156 [ PMC free article ] [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Weir J. P. (2005). Quantifying test-retest reliability using the intraclass correlation coefficient and the SEM . J. Strength Cond. Res. 19 , 231–240. 10.1519/00124278-200502000-00038 [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Weissler A. M., Harris W. S., Schoenfe C. D. (1968). Systolic time intervals in heart failure in man . Circulation 37 , 149–159. 10.1161/01.CIR.37.2.149 [ PubMed ] [ CrossRef ] [ Google Scholar ]
• White D. W., Raven P. B. (2014). Autonomic neural control of heart rate during dynamic exercise: revisited . J. Physiol. 592 , 2491–2500. 10.1113/jphysiol.2014.271858 [ PMC free article ] [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Yamamoto K., Miyachi M., Saitoh T., Yoshioka A., Onodera S. (2001). Effects of endurance training on resting and post-exercise cardiac autonomic control . Med. Sci. Sports Exerc. 33 , 1496–1502. 10.1097/00005768-200109000-00012 [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Yamamoto Y., Hughson R. L., Nakamura Y. (1992). Autonomic nervous-system responses to exercise in relation to ventilatory threshold . Chest 101 , S206–S210. [ PubMed ] [ Google Scholar ]
• Yamamoto Y., Hughson R. L., Peterson J. C. (1991). Autonomic control of heart-rate during exercise studied by heart-rate-variability spectral-analysis . J. Appl. Physiol. 71 , 1136–1142. [ PubMed ] [ Google Scholar ]