Impaired autonomic control of heart rate has been linked to increased risk of cardiovascular mortality (23,24,33). This increased risk may be a result of electrical instability in the heart that results from an autonomic environment dominated by sympathetic nervous system (SNS) influences. Electrical, surgical, and pharmacological reduction of parasympathetic nervous system (PSNS) tone or stimulation of sympathetic activity have been shown to reduce the threshold for ventricular fibrillation (3,18,30). Thus, there is increasing interest in the use of noninvasive measures to examine cardiac autonomic function. One such measure that is frequently studied is heart rate variability (HRV). Measures of HRV, particularly those using frequency-domain techniques, can provide insight into the balance between the branches of the autonomic nervous system, and thus have potential diagnostic and prognostic significance.
HRV is reduced in patients with cardiac abnormalities such as diabetic neuropathies, myocardial infarction, and heart transplantation (3,30). Reduced HRV has been shown to be predictive of a second myocardial infarction and an independent predictor of risk for sudden cardiac death in asymptomatic individuals (23). Low HRV is also associated with insulin resistance, low levels of cardiorespiratory fitness, and other factors associated with increased risk of cardiovascular disease (16,24). Therefore, it appears that HRV is a useful noninvasive marker of risk for cardiovascular disease.
Chronic endurance exercise training is associated with increased HRV and a shift toward more PSNS influence on cardiovascular function (13,21,27), indicating an improved cardiac autonomic environment and improved cardiac health (3,30,33). This evidence provides some mechanistic insight into one of the cardio-protective benefits of chronic endurance training. Recently, it has been suggested that some of the benefits traditionally associated with endurance training may be realized quite rapidly, perhaps after only one bout of exercise (10,32). Single bouts of endurance exercise result in lower blood pressure, decreased low-density blood lipids, and increased insulin sensitivity (7,22,28). Because acute exercise affects the autonomic nervous system, insulin sensitivity, blood pressure, and other factors that ultimately influence the control of heart function, it is reasonable to hypothesize that changes in HRV as a result of endurance training may be at least partly the result of the accumulation of the acute effects of exercise, rather than or in addition to, adaptation to the stress of chronic training.
Although the improvements in HRV with chronic exercise training have been well documented (13,21,27), the effects of single bouts of exercise on HRV have not been studied extensively. Several investigators have examined the changes in HRV after a bout of submaximal exercise, but measurements were limited to a relatively short (15–180 min) period after the exercise bout (4,25,31). Other investigators have followed HRV changes for longer periods after exercise, but the exercise intervention consisted of intense or maximal workloads that were not typical of the dose of exercise associated with improvements in cardiovascular disease risk factors including HRV (8). In all of these previous investigations, exercise resulted in changes in HRV that indicated a shift toward more SNS contribution to autonomic control immediately postexercise, followed by a gradual return toward preexercise values.
Of interest from a clinical or epidemiological perspective is the question of the effects of a single bout of submaximal endurance exercise on HRV. This is the type of exercise prescribed in both clinical and lifestyle interventions for the enhancement of cardiovascular health. The positive changes in the internal environment and cardiac autonomic control as a result of individual bouts of exercise are obscured immediately postexercise by the predominant influence of the SNS. To investigate the acute cardio-protective effects of exercise, changes in HRV must be investigated over the course of a substantial recovery period to determine whether HRV increases above resting levels after the perturbation during exercise. This recovery period should be similar to that encountered in the field, where individuals on a regular exercise program usually exercise at 24- to 48-h intervals.
The purpose of this study was to investigate the effects of a single bout of submaximal exercise on HRV over the course of a 22-h recovery period. This study tested the hypothesis that during later recovery (~6 –22 h) from a single bout of submaximal exercise, overall HRV would be increased, and that power spectral indices of HRV would indicate increased PSNS influence in the control of heart function as compared to baseline. This investigation provides the foundation for future research in the investigation of additional mechanisms through which exercise provides an enhanced electrical stability to the heart.
Eleven healthy, young, moderately active males were recruited on a volunteer basis through word of mouth and flyers posted at the University and surrounding community, as the day-to-day reliability of resting HRV measures has been verified in this population (20). Subjects were 18 –35 yr old (mean (SD) age 26.8 (5.0) yr, body mass 75.4 (8.5) kg, body mass index (BMI) 23.6 (2.9), and peak oxygen uptake (V̇ O2peak) 49.5 (9.6) mL·kg−1 ·min−1). Subjects were excluded from the study if they had cardiovascular or musculoskeletal limitations to maximal exercise, if they were taking any medication or supplements that influenced the control of heart rate or autonomic function (e.g., β-blockers, angiotensin converting enzyme inhibitors, ephedrine), if they were smokers, or if they had any ECG dysrhythmias. Subjects provided written informed consent and completed a physical activity readiness questionnaire (PAR-Q) according to the guidelines established by the University of Massachusetts institutional review board.
Habituation and peak oxygen uptake testing.
On the first day of testing, subjects completed a habituation session to become familiar with the experimental protocol, and a cycle ergometer test to assess V̇O2peak. The habituation session included 15–30 of breathing in time with a metronome while lying supine. Practice continued until the subject was able to match his breathing frequency to the frequency of the metronome.
The cycle ergometer test consisted of a continuous protocol on a SensorMedics Ergometrics 800s electronically braked ergometer. Starting work rate (100 –150 W) and increment (20 –30 W) were determined based on the subject’s body size and familiarity with cycling exercise. Pedal cadence was allowed to vary throughout the test to accommodate the subject’s preference, as the ergometer adjusted resistance to maintain a constant work rate in response to changes in cadence. The work rate was increased every 2 min until the subject reached volitional exhaustion. During the test, subjects breathed through a rubber mouthpiece attached to a Hans-Rudolph high velocity two-way nonrebreathing valve (Kansas City, MO, model 2700, dead space 95 mL). Inspired volume was measured on a breath-by-breath basis using an electronic pneumotachograph (Fitness Instrument Technologies, Quogue, NY). Expired gasses were directed into a 3.0-L mixing chamber and continuously sampled (250 Hz) and analyzed for oxygen and carbon dioxide concentrations using Ametek oxygen and carbon dioxide analyzers (AEI, Pittsburgh, PA). The metabolic gas analysis system was calibrated before each test using gasses of known concentration and with a known volume of air. Ventilation rate and expired gas data were transmitted to a personal computer-based system (Fitness Instrument Technologies, Quogue, NY) that compiled respiratory gas exchange data at 15-s intervals. Heart rate was measured at 15-s intervals throughout each test using a wireless chest-band heart rate monitor (Polar Electro, Finland). Rating of perceived exertion (RPE) was assessed at the end of each stage, and at the end of the test, using the Borg scale, where a score of 6 represents “no effort at all” and a score of 20 represents “maximal effort.”
The test was considered a peak effort, and the highest oxygen consumption value was included in the analysis, if three of the following four criteria were met (a) a respiratory exchange ratio greater than 1.10, (b) a final RPE score of 17 or greater on the Borg scale, (c) a maximal heart rate within 15 beats·min−1 of age predicted maximum (220 − age), and (d) an increase in oxygen consumption between the final two stages of the test that was less than the average increase between the previous workloads. All subjects met criteria “a” through “c,” and eight subjects met criterion “d.”
On the second day of testing, separated from the first by a minimum of 72 h and a maximum of 1 wk, subjects reported to the lab immediately upon waking. Subjects were instructed to refrain from exercise outside of the study for 72 h before the second day of testing and not to consume any caffeine or alcohol for 24 h before testing as these agents have been shown to influence resting HRV (30). Subjects arrived at the laboratory building via automobile, with a minimum of walking. Subjects were given a meal of standardized energy content and macronutrient composition (Promax Bar, SportPharma, 4 kcal·kg−1 body mass, 53% carbohydrate, 17% fat, 30% protein) in order to minimize interindividual differences in the HRV response to feeding.
One hour after the meal, ECG data were obtained while the subject rested quietly in the supine position. Surface ECG was collected with pregelled disposable Ag-AgCl electrodes placed in a lead I configuration. Electrodes were placed on the anterior aspect of each forearm, just proximal to the wrist, and a ground electrode was placed on the interior shank. Resting ventilation was measured using a strain gauge (TSD101B Transducer, Biopac Systems Inc., Santa Barbara, CA) positioned around the abdomen, superior to the umbilicus. After the electrodes and ventilation transducer were placed, the subject rested in the supine position on a bed in a quiet, semidark room maintained at a temperature of 19–21°C. Subjects were provided with sheets and blankets to adjust their comfort level. Subjects were monitored to be certain that they remained awake and relaxed, and were instructed to remain motionless during the data collection.
After 5 min of rest, 10 min of ECG data were collected. Subjects matched their respiratory frequency to an auditory metronome set to 12 breaths·min−1 (0.2 Hz). There was no attempt to influence tidal volume. Paced breathing was important in many subjects to assure that the respiratory frequency and resultant peak in the power spectrum are outside of the range of LF power (30).
After the collection of resting ECG, subjects completed 60 min of cycle ergometer exercise at approximately 65% of V̇O2peak, as this dose of exercise is typical of that prescribed in the health literature, and has been shown to reduce circulating insulin for at least 24 h postexercise (22) and produce a variety of health related benefits when incorporated into a regular exercise program (29). Heart rate was monitored throughout the exercise period using a wireless chest-band heart rate monitor (Polar Electro). Subjects were required to maintain their heart rate within 5 beats per minute of the rate that corresponded to 65% of V̇O2peak from the cycle ergometer test.
ECG data were collected according to the procedure described above at 1, 3, 6, and 22 h postexercise. One hour before each postexercise measurement, subjects were given a meal identical to that provided before the baseline HRV measurement. After the 6-h postexercise measurement, subjects returned to their homes via automobile, with instructions not to exercise and not to consume any alcohol or caffeine. On the third day of testing, subjects reported to the lab immediately upon waking. Subjects were again provided with a standardized meal. One hour after this meal, resting ECG data were obtained according to the procedures described above.
A condition with no exercise was included in the design, wherein subjects served as their own controls, completing all testing with the exception of the 1-h submaximal exercise bout. The order in which the subjects underwent the exercise or nonexercise condition was determined using a balanced, randomized design.
ECG and ventilation data were collected using a BIOPAC (BIOPAC Systems Inc., Santa Barbara, CA) data acquisition system consisting of two single-channel differential input amplifier and signal conditioning modules (ECG100A and RSP100B) and an analog-to-digital (A/D) converter module (MP100). Data were collected at 1000 Hz and converted from analog to digital format with 16-bit resolution. R-wave peaks were detected automatically with an R-wave detection algorithm supplied by the manufacturer. Instantaneous heart rate (beats·min−1) and inter-beat interval (RR; ms) were calculated from the difference in time of successive R-wave peaks using software provided by BIOPAC Systems.
Frequency domain analyses were performed according to the methods described by Melanson (20), on a series of five consecutive minutes manually selected from the data series for each condition. Care was taken to make sure that each 5-min segment was free of movement artifact, premature beats, and any other artifact of unknown origin. From each series, the successive interbeat intervals were plotted against time, and interpolated with a 1024-point cubic spline method. The resultant signal was resampled at 5 Hz, and the signal was detrended and submitted to a Blackman window. Amplitude density of the signal was estimated using the Fast Fourier Transform (FFT) algorithm. These spectra were squared to obtain power spectral density (PSD) plots for each signal. Total power and the power in the various frequency bands were calculated by integrating the area under the PSD curve in the frequency ranges set forth by the Task Force of the European Society of Cardiology and the North American Society of Pacing and Electrophysiology (total power: 0.0– 0.40 Hz; low frequency power: 0.04–0.15 Hz; high frequency power: 0.15– 0.40 Hz) (30).
Four time-domain measures of HRV were calculated from the time series of interbeat intervals: mean interbeat interval (RR), standard deviation of RR intervals (SDNN), root mean squared standard deviation of RR (rMSSD), and proportion of normal r-r intervals that differed from the next interval by more than 50 ms (pNN50). Four frequency-domain measures of HRV were calculated from the power estimates, including total power (0–0.4 Hz), normalized low frequency power (LFNU, power at 0.04–0.15 Hz/power at 0.04–0.40 Hz), normalized high frequency power (HFNU, power 0.15–0.4 Hz/power at 0.04–0.40 Hz), and the ratio of low frequency power to high-frequency power (LF:HF).
Many of the outcome measures were not normally distributed and showed increasing variance with increasing mean. Therefore, variables were trans-formed using the natural logarithm function (loge) as necessary before analysis. This transformation resulted in data that approximately met the assumptions required for parametric statistical methods. Data were analyzed using a two-way (condition × time point) ANOVA, with repeated measures on condition and time point. When there was a significant interaction between condition and time point, post hoc contrasts were used to assess the differences between the conditions. The specifics of the statistical analyses follow.
Outcome measures were analyzed with the SAS statistical computing package version 8.02 (Cary, NC) for Windows, using the MIXED procedure with the REPEATED statement. This allowed for optimization of the covariance matrices to account for the correlation between observations within a given subject. Data are reported in figures as mean proportional changes from baseline values (measure at time t/baseline measure) or in tables as mean percentage difference in change between the conditions ((percent change in value from baseline to time t in exercise condition − percent change in value from baseline to time t in nonexercise condition)/percent change in value from baseline to time t in nonexercise condition), with approximate P values for comparisons between conditions. P values are an approximation to the probability of observing a difference as large or larger than the difference found in the sample data, given that the null hypothesis is correct. The accuracy of this approximation is dependent on the degree to which the data are normally distributed. Thus, the authors did not use a hard cut-off, or alpha level, to determine the statistical significance of the present results. Instead, the magnitude of differences is presented, along with approximate P values, to provide the reader with enough information to draw his or her own conclusions about the significance of any differences observed.
Mean RR period was similar between the exercise and nonexercise conditions (differences < 2%), except at 22 h postexercise. At that time point, RR was approximately 7% longer than in the nonexercise condition (P ≈ 0.04). Although SDNN was lower and rMSSD was higher at all time points in the exercise condition, these differences were modest compared with the variability in the measures (P > 0.28). In contrast, pNN50 was markedly increased at 1 h (~647% difference, P ≈ 0.06), 3 h (~667% difference, P ≈ 0.10), and 6 h postexercise (~359% difference, P ≈ 0.17). At 22 h postexercise, pNN50 appeared to be more similar to nonexercise levels (~124% difference, P ≈ 0.59). Figure 1 shows the mean ± SE proportional difference in pNN50 from baseline at each time point during the exercise and nonexercise conditions. Complete results of the analysis of time-domain measures are shown in Table 1.
HFNU was higher at all time points after exercise than it was in the nonexercise condition. Most notably, HFNU was approximately 21% higher 1 h postexercise (P ≈ 0.07) and nearly 18% higher at 22 h postexercise (P ≈ 0.10) than at the corresponding times during the nonexercise condition. Figure 2 shows the mean ± SE proportional difference in HFNU from baseline at each time point during the exercise and nonexercise conditions. LFNU was between 40% and 55% lower after exercise than it was at the corresponding time points during the nonexercise condition (P ≈ 0.03– 0.19). Figure 3 shows the mean ± SE proportional difference in LFNU from baseline at each time point during the exercise and nonexercise conditions. Subsequent to the changes in HFNU and LFNU, LF:HF was between 45% and 65% lower in the exercise condition compared to the nonexercise condition (P < 0.01– 0.19). Figure 4 shows the mean ± SE proportional difference in LF:HF from baseline at each time point during the exercise and nonexercise conditions. Table 2 provides complete results of the analysis of frequency domain measures.
There are several noteworthy findings of the present study. The first is that two markers of cardiac parasympathetic nervous activity (HFNU power and pNN50) were increased in the day after a single bout of submaximal endurance exercise. The magnitude of this change is slightly smaller than that observed in studies of the effects of chronic endurance training on indices of HRV (13,21,27). The changes observed in the present study were more similar to those observed as a result of short-term endurance training (17). The present results suggest that endurance exercise may acutely increase the contribution of the PSNS to control of resting cardiac function between 1 h and at least 22 h postexercise. Additionally, the present study found that other frequency domain measures of HRV (LFNU power and LF:HF power ratio) were reduced by a single bout of exercise, suggesting a shift in sympathovagal balance toward a larger relative role for the PSNS after exercise. The present results suggest that the effects of acute submaximal exercise on the ANS may be quite different than the effects of acute maximal exercise that have been reported by other authors, who have found HRV changes indicating an increased SNS contribution to autonomic cardiovascular function after acute exercise (1,8). Finally, the present results provide further insight into the time course of changes in HRV after a submaximal exercise bout.
The results of the present investigation are supported by those of Bernardi et al. (2), who reported a shift toward greater parasympathetic influence in measures of baroreflex function 24 h after a prolonged trail run at altitude, and Hautala et al. (11), who observed a similar phenomenon with regard to baroreflex function, as well as an increase in HF HRV nearly 48 h after a 75-km cross-country ski race. Although the results mentioned above were based on much more strenuous and prolonged exercise than those of the present study, Halliwill and colleagues (9) described reductions in sympathetic nerve outflow and sympathetic baroreflex control after a 60-min bout of cycling exercise at 60% of V̇O2max; however, their investigation was limited to the period immediately after exercise. Overall, the present study adds to the body of knowledge regarding the effects of acute exercise on cardiac autonomic function, but there are many questions remaining to be answered.
Although the physiological meaning of changes in HF HRV is well established, considerable controversy exists in the literature regarding the interpretation of changes in the LF range of HRV. Several authors have suggested that LF power or LFNU power do not represent SNS activity, due to changes in these measures that occur with manipulations of PSNS function (6,12). These authors focus on the fact that the PSNS contributes a portion of LF power. Under conditions similar to those encountered in the present investigation, it has been shown that while LFNU and LF:HF are influenced by the PSNS, they are also strongly influenced by the SNS and as such can be interpreted as representing the dynamic balance between the two branches of the ANS (14,19,30).
Thus, the present results provide preliminary evidence that a single bout of submaximal dynamic exercise changes HRV in a manner that suggests a shift in ANS balance towards proportionally more PSNS influence. Such a shift in ANS balance is generally thought to indicate a healthier and more electrically stable environment for the myocardium. The changes in HRV observed in the present study suggest that in the day after a bout of submaximal endurance exercise, the heart may be more sensitive to neurological inputs, and more flexible in adapting to perturbations. It remains to be determined if these changes represent an acutely decreased risk of cardiac events, or if they are merely coincidental to recovery from exercise. Recent work by Powers et al. (26) suggests that there are direct cardioprotective effects of recent exercise, and the present results add to that body of evidence.
Additionally, many of the beneficial effects of recent exercise, such as changes in insulin sensitivity and blood pressure, are directly or indirectly linked to cardiovascular function. The results of this study suggest that HRV may be a useful, noninvasive measure that is sensitive to these types of changes. As such, HRV may be a useful clinical and research tool for assessing additional dimensions of, or summarizing, physiological changes that occur with exercise. Furthermore, measures of HRV may be useful for investigations of the physiological mechanisms underlying the beneficial effects of exercise. Further research is needed to determine the extent to which other physiological systems are influencing HRV postexercise.
Another important implication of the present findings is related to the use of HRV as a measure of the autonomic effects of a training program or other intervention. As the results of this study show that the effects of even submaximal exercise last at least 22 h, care must be taken to limit exercise in subjects before their undergoing HRV assessment. The present results also suggest the need for further research to clarify the differences between the effects of chronic training and those of recent exercise. It is clear that as the issue of the dose-response relationship between exercise and health benefits is explored more thoroughly, HRV is an outcome measure that bears further investigation.
There are several potential limitations to the interpretation of the results of this study. It has been shown that HF power and other markers of PSNS activity can be influenced by both the rate and depth of breathing. In the present study, subjects breathed at a paced frequency (0.2 Hz) in order to clearly separate HF and LF HRV. This frequency is similar to that used in previous studies in this laboratory (20,21) and approximates the mean un-paced respiration frequency in subjects from this population (young, lightly to moderately active).
All subjects were given identical instructions to breathe as naturally as possible while synchronizing their breathing to the metronome. Additionally, all measurements were made under the same conditions. Presumably then, the overall effect of controlled respiration would, if anything tend to reduce the magnitude of differences between the two conditions in respiration mediated (HF power, pNN50, rMSSD) components of HRV. As such, this potential problem should not confound the interpretation of any of the positive results of the present study.
Several measures in the present study showed a small chance of Type I error in concluding a statistical difference between the conditions. However, some of the other measures, most notably rMSSD, showed consistent differences between the conditions in the present sample, but statistical analysis suggested a large chance of Type I error in concluding that there was a difference between conditions. The sample size for this investigation was chosen a priori to provide a power of 0.90 to detect a 15% difference in HFNU at an α-level of 0.10 based on data from Melanson (20). HFNU was chosen because the investigators felt that HFNU was the most important indicator of beneficial changes in HRV.
Although the sample size was nearly adequate for that purpose (observed power at α = 0.10 for 15% difference = 0.78), the power to detect a difference of the magnitude observed in rMSSD or total power (for example 27.5%) was frequently less than 0.20 given the variability in the present sample. Thus, potentially meaningful differences between the conditions may have been obscured by the variability of the data. The data used for estimating sample size a priori were taken from measurements that were obtained first thing in the morning after an overnight fast. Additional care should be taken in the future to include variability estimates that are more representative of the experimental conditions, as in this case the added effects of diurnal and postprandial variation in HRV caused the observed power to be below the desired level.
Another possible limitation to the interpretation of the results of this study is that the measurements were made postprandially. In early research on HRV, care was generally taken to ensure that subjects were fasted, to minimize the effects of digestion on measures of HRV. However, in the present study, it was determined that, for reasons of subject comfort and to more closely approximate a free-living situation, the subjects should be fed during the protocol.
Based on information from pilot testing, it was decided to use a meal of mixed macronutrient composition, and to adjust the energy content of the meal according to the subjects’ body mass. As measurements were all made under the same conditions, and subjects served as their own controls, there should be no confounding effects of the meals on the interpretation of the positive results of this study.
Another limitation to the application of the results of this study is the lack of other physiological or autonomic function measurements to accompany the assessment of HRV. In the absence of other information, any conclusions about the physiological meaning of the observed changes can only be based on earlier research, which may not adequately apply to the present experimental conditions. Furthermore, it is only educated speculation that the changes in HRV observed in this study suggest that cardiac autonomic function is improved by a single bout of exercise in a similar manner to the improvements seen with chronic training. However, there is no reason to suspect that acute changes in HRV after a single bout of exercise result from different mechanisms than those that cause changes with extended training.
An additional concern in interpreting the results of the present study is related to the nature of the experimental intervention. Most of the volunteers who participated in this investigation were moderately physically active, on the order of 5–7 h·wk−1 of moderate physical activity. As several of the measures in this study did not return to baseline levels on the morning after the nonexercise condition after undergoing diurnal changes during the previous day, it is possible that the exercise condition was more representative of the subjects’ normal state, whereas the lack of exercise in the nonexercise condition may have represented a perturbation to the normal state of the subjects. In this case, the nonexercise condition may represent an acute “detraining” stimulus, with measures steadily drifting away from their normal values as the time since the subjects’ last exercise stimulus increased. This effect has been observed with physiological measures that are related to HRV, particularly insulin levels (5). Future investigations are needed to determine whether the present results were due to acute effects of exercise, changes due to detraining, or some combination of the two.
Along similar lines, the cardiovascular fitness level of the subjects in this study suggests an additional caveat in the interpretation of the present results. As the subjects were relatively young and fit (mean age = 26.8 yr, mean V̇O2peak = 49.5 mL·kg−1·min−1) it is possible that the ANS response to a similar bout of exercise would be different among less fit individuals. In particular, it seems plausible that in an older, less active, or less fit sample, the dose of exercise used in this study may represent a greater challenge to the ANS. This increased challenge may result in a greater SNS response to the exercise bout, which could affect both the time course and magnitude of postexercise changes in HRV. Effects of fitness level on autonomic response to acute exercise have been observed by other investigators (11). As such, it is important to limit conclusions about the effects of single bouts of exercise on resting HRV to the population from which the present sample was drawn.
The results of the present study are quite provocative and suggest many basic, as well as applied questions to be addressed by future research. Foremost among all of the issues remaining to be addressed is the repeatability of the present results. Because this investigation took an original approach to the assessment of HRV after a bout of exercise, it is essential that future studies show these results to be repeatable in the same population, as well as in other groups. Perhaps future research could include more measurements between 6 and 22 h postexercise to complete our understanding of the time course of changes in HRV after exercise. Information about the duration of exercise-induced acute changes in HRV is essential in the construction of studies using HRV as an outcome, where it is important that effects of recent exercise are separated from the effects of other experimental interventions.
Foremost among the fundamental issues left to be addressed is the determination of the physiological bases of the observed changes. Insulin sensitivity and baroreflex action are obvious candidates, as both have been shown to be related both to exercise and to HRV (15,16,27). As in the present study, earlier studies that have reported changes similar to those observed in the present study have failed to collect sufficient data to provide a physiological explanation of their results. It has been suggested, however, that increased baroreflex sensitivity, changes in sympathetic nerve outflow, or changes in the levels of vasoactive substances in the circulation may be responsible for the observed effects (15,16,27). Plasma volume change as a result of exercise is another candidate for a mechanism to explain the observed changes in cardiovascular function. In the present study, it seems unlikely that any plasma volume changes induced by the exercise bout would have persisted throughout the entire measurement period, particularly given the fact that the subjects consumed water ad libitum throughout the exercise and during their visit to the lab. Another basic question remaining to be addressed is whether the acute changes observed in HRV are representative of other changes in ANS function. An essential component of future studies is the inclusion of clinical ANS measurements such as baroreflex sensitivity or pressor response along with HRV assessment, which would strengthen the interpretation of HRV measures as indices of autonomic function.
Among the questions of application suggested by the present results is the issue of the optimal dose of exercise. Future studies should compare the effects of different durations and intensities of exercise bouts on HRV and autonomic function. This information is extremely useful in crafting the public health message about the benefits of exercise. If smaller doses of exercise are shown to be beneficial, perhaps the public will see fewer barriers to initiating a program of habitual exercise. Future research should also attempt to distinguish the effects of recent exercise on HRV from those of chronic training, as this would potentially provide fertile research lines in the area of the physiological basis of the health benefits of exercise.
Although other studies have investigated the effects of single bouts of exercise on HRV, these investigations have generally been confined to more strenuous (usually maximal) exercise doses or have lacked the temporal resolution of the present investigation. The present results suggest that quantifiable changes in the ANS environment occur as a result of a single bout of exercise and that these changes persist for at least 22 h after the cessation of exercise. As the changes seen in the present investigation are similar to those observed in training studies, and in the opposite direction to known detrimental changes in HRV, it is suggested that beneficial changes in ANS function may be added to the list of positive effects of recent exercise. Future studies should focus on replicating the present results and elucidating further the time course of changes in HRV after submaximal exercise, as well as on determining the physiological mechanisms underlying the observed changes and attempting to ascertain the optimal dose of exercise required to achieve a beneficial effect on ANS function.
The authors would like to thank John Buonaccorsi of the Department of Mathematics and Statistics, University of Massachusetts, Amherst, for his assistance in the design and analysis phases of this study.
This study was funded, in part, by a grant from the LifeFitness Corporation.
1. Arai, Y., J. P. Saul, P. Albrecht, et al. Modulation of cardiac autonomic activity during and immediately after exercise. Am. J. Physiol.
2. Bernardi, L., C. Passino, R. Robergs, and O. Appenzeller. Acute and persistent effects of a 46-kilometer wilderness trail run at altitude: cardiovascular autonomic modulation and baroreflexes. Cardiovasc. Res.
3. Berntson, G. G., J. T. Bigger, D. L. Eckberg, et al. Heart rate variability: origins, methods, and interpretive caveats. Psychophysiology
34:623– 648, 1997.
4. Brenner, I. K., S. Thomas, and R. J. Shephard. Spectral analysis of heart rate variability during heat exposure and repeated exercise. Eur. J. Appl. Physiol. Occup. Physiol.
5. Burstein, R., C. Polychronakos, C. J. Toews, J. D. MacDougall, H. J. Guyda, and B. I. Posner. Acute reversal of the enhanced insulin action in trained athletes: association with insulin receptor changes. Diabetes
6. Eckberg, D. L. Sympathovagal balance: a critical appraisal. Circulation
7. Ferguson, M. A., N. L. Alderson, S. G. Trost, D. A. Essig, J. R. Burke, and J. L. Durstine. Effects of four different single exercise sessions on lipids, lipoproteins, and lipoprotein lipase. J. Appl. Physiol.
8. Furlan, R., S. Piazza, S. Dell’Orto, et al. Early and late effects of exercise and athletic training on neural mechanisms controlling heart rate. Cardiovasc. Res.
27:482– 488, 1993.
9. Halliwill, J. R., J. A. Taylor, and D. L. Eckberg. Impaired sympathetic vascular regulation in humans after acute dynamic exercise. J. Physiol.
10. Haskell, W. L. Dose-response issues from a biological perspective. In:Physical Activity, Fitness, and Health: International Proceedings and Consensus Statement
, C. Bouchard, R. J. Shephard, T. Stephens (Eds.). Champaign, IL: Human Kinetics, 1994, pp. 1030–1038.
11. Hautala, A., M. P. Tulppo, T. H. Makikallio, R. Laukkanen, S. Nissila, and H. V. Huikuri. Changes in cardiac autonomic regulation after prolonged maximal exercise. Clin. Physiol.
12. Houle, M. S., and G. E. Billman. Low-frequency component of the heart rate variability spectrum: a poor marker of sympathetic activity. Am. J. Physiol.
13. Iellamo, F., J. M. Legramante, M. Massaro, G. Raimondi, and A. Galante. Effects of a residential exercise training on baroreflex sensitivity and heart rate variability in patients with coronary artery disease: a randomized, controlled study. Circulation
14. Jasson, S., C. Medigue, P. Maison-Blanche, et al. Instant power spectrum analysis of heart rate variability during orthostatic tilt using a time-/frequency-domain method. Circulation
15. Kirwan, J. P., R. E. Bourey, W. M. Kohrt, M. A. Staten, and J. O. Holloszy. Effects of treadmill exercise to exhaustion on the insulin response to hyperglycemia in untrained men. J. Appl. Physiol.
16. Laitinen, T., I. K. Vauhkonen, L. K. Niskanen, et al. Power spectral analysis of heart rate variability during hyperinsulinemia in nondiabetic offspring of type 2 diabetic patients: evidence for possible early autonomic dysfunction in insulin-resistant subjects. Diabetes
17. Lee, C. M., R. H. Wood, and M. A. Welsch. Influence of short-term endurance exercise training on heart rate variability. Med. Sci. Sports Exerc.
18. Lown, B., and R. L. Verrier. Neural activity and ventricular fibrillation. N. Engl. J. Med.
19. Malliani, A. The pattern of sympathovagal balance explored in the frequency domain. News Physiol. Sci.
20. Melanson, E. L. Resting heart rate variability in men varying in habitual physical activity. Med. Sci. Sports Exerc.
21. Melanson, E. L., and P. S. Freedson. The effect of endurance training on resting heart rate variability in sedentary adult males. Eur. J. Appl Physiol.
85:442– 449, 2001.
22. Mikines, K. J., B. Sonne, B. Tronier, and H. Galbo. Effects of acute exercise
and detraining on insulin action in trained men. J. Appl. Physiol.
23. Molgaard, H., K. E. Sorensen, and P. Bjerregaard. Attenuated 24-h heart rate variability in apparently healthy subjects, subsequently suffering sudden cardiac death. Clin. Auton. Res.
24. Odemuyiwa, O., M. Malik, T. Farrell, Y. Bashir, J. Poloniecki, and J. Camm. Comparison of the predictive characteristics of heart rate variability index and left ventricular ejection fraction for all-cause mortality, arrhythmic events and sudden death after acute myocardial infarction. Am. J. Cardiol.
68:434– 439, 1991.
25. Oida, E., T. Moritani, and Y. Yamori. Tone-entropy analysis on cardiac recovery after dynamic exercise. J. Appl. Physiol.
26. Powers, S. K., S. L. Lennon, J. Quindry, and J. L. Mehta. Exercise and cardioprotection. Curr. Opin. Cardiol.
27. Seals, D. R., and P. B. Chase. Influence of physical training on heart rate variability and baroreflex circulatory control. J. Appl. Physiol.
28. Seals, D. R., H. G. Silverman, M. J. Reiling, and K. P. Davy. Effect of regular aerobic exercise on elevated blood pressure in postmenopausal women. Am. J. Cardiol.
29. Surgeon General’s Report on Physical Activity and Health.
Pittsburgh: U.S. Department of Health and Human Services, Centers for Disease Control and Prevention, 1996.
30. Task Force of the European Society of Cardiology and the North American Society of Pacing and Electrophysiology. Heart rate variability: standards of measurement, physiological interpretation and clinical use. Circulation
31. Terziotti, P., F. Schena, G. Gulli, and A. Cevese. Post-exercise recovery of autonomic cardiovascular control
: a study by spectrum and cross-spectrum analysis in humans. Eur. J. Appl. Physiol.
32. Thompson, P. D., S. F. Crouse, B. Goodpaster, D. Kelley, N. Moyna, and L. Pescatello. The acute versus the chronic response to exercise. Med. Sci. Sports Exerc.
33:S438–S445; discussion S452–S433, 2001.
33. Tsuji, H., M. G. Larson, F. J. Venditti, et al. Impact of reduced heart rate variability on risk for cardiac events: the Framingham Heart Study. Circulation