Aerobic Training Improves Vagal Reactivation Regardless of Resting Vagal Control ANTONIO DUARTE1,2, PEDRO PAULO SOARES3, LINDA PESCATELLO4, and PAULO FARINATTI5 1 Brazilian Army Research Institute of Physical Fitness, Rio de Janeiro, BRAZIL; 2Physical Education Graduate Program, Gama Filho University, Rio de Janeiro, BRAZIL; 3Department of Physiology and Pharmacology, Fluminense Federal University, Nitero´i, Rio de Janeiro, BRAZIL; 4Department of Kinesiology, University of Connecticut, Storrs, CT; and 5Physical Activity and Health Promotion Laboratory, Rio de Janeiro State University, Rio de Janeiro, BRAZIL
ABSTRACT DUARTE, A., P. P. SOARES, L. PESCATELLO, and P. FARINATTI. Aerobic Training Improves Vagal Reactivation Regardless of Resting Vagal Control. Med. Sci. Sports Exerc., Vol. 47, No. 6, pp. 1159–1167, 2015. Purpose: Resting cardiac vagal modulation (RCVM) and postexercise vagal reactivation (PEVR) are markers of parasympathetic activity. We investigated whether adaptations in these markers to aerobic training are influenced by baseline autonomic control. Methods: Forty healthy men (19.2 T 0.8 yr) of similar ˙ O2peak = 50.4 T 5.7 mLIkgj1Iminj1) completed the study, being matched for autonomic activity and cardiorespiratory fitness (V randomized into four groups: training low-RCVM (TL; n = 11, high-frequency power component [HF] = 48.1 T 8.2 normalized units [n.u.]) and high-RCVM (TH; n = 11, HF = 63.1 T 5.9 n.u.) and nonexercise control low-RCVM (CL; n = 9, HF = 47.1 T 7.5 n.u) and high-RCVM (CH; n = 9, HF = 65.5 T 8.3 n.u.). Aerobic training groups exercised 3 dIwkj1 for 40 min at 75%–85% HR reserve for 12 wk. Before and after the training period, sequences of 5-min R–R intervals were recorded at rest and immediately after maximal treadmill test to estimate (a) RCVM (HF) and (b) PEVR (root mean square of successive R–R differences—rMSSD3–5min = mean ˙ O2peak in TL (11.7% T 5.4%, P G 0.01) and TH (7.0% T 2.9%, P G 0.01), value from 3 to 5 min of recovery). Results: Training improved V with no difference between groups (P = 0.70), but not in CL (0.8% T 3.9%) and CH (1.8% T 6.2%, P = 0.90). Only TL increased RCVM (56.6 T 13.3 n.u., P = 0.03), approaching TH level (58.9 T 12.3 n.u.; P = 0.60); rMSSD3-5min increased in both training groups (P G 0.01) but not in controls (P = 0.99). Relative changes in RCVM ($HF%) and PEVR ($rMSSD3–5min%) were significantly correlated in TL (r = 0.61, P = 0.04). Conclusions: PEVR after exercise increased in both exercise training groups, whereas RCVM increased only in the group with low vagal activity at baseline. Vagal reactivation may be improved by aerobic training, even when basal activity remains unaltered. Key Words: EXERCISE, AEROBIC POWER, AUTONOMIC NERVOUS SYSTEM, HR VARIABILITY, PARASYMPATHETIC MODULATION
Address for correspondence: Paulo Farinatti, Laborato´rio de Atividade Fı´sica e Promo0a˜o da Sau´de, Universidade do Estado do Rio de Janeiro, Rua Sa˜o Francisco Xavier 524, sala 8121-8133 F, Maracana˜, Rio de Janeiro, RJ, CEP: 20.550-013, Brazil; E-mail: [email protected]; [email protected]. Submitted for publication April 2014. Accepted for publication September 2014. 0195-9131/15/4706-1159/0 MEDICINE & SCIENCE IN SPORTS & EXERCISEÒ Copyright Ó 2014 by the American College of Sports Medicine DOI: 10.1249/MSS.0000000000000532
1159
Copyright © 2015 by the American College of Sports Medicine. Unauthorized reproduction of this article is prohibited.
BASIC SCIENCES
A
is mixed regarding the effectiveness of aerobic training and cardiorespiratory fitness on resting and postexercise markers of parasympathetic autonomic control of the heart. Some investigations have reported higher (1,3,8) and others no difference (5,18) in levels of resting parasympathetic modulation of HR in trained compared to untrained subjects. Furthermore, some authors proposed that resting cardiac vagal control improves with aerobic training (7,20,23,41), whereas others concluded that training would not influence autonomic modulation at rest (14,26,29). With regard to the postexercise autonomic control of the heart, some investigators have found a faster HR recovery after exercise, which reflects vagal reactivation (37), in aerobically trained compared to untrained individuals (8,33). The few available cross-sectional and longitudinal studies about the relationship between resting cardiac parasympathetic control and postexercise vagal reactivation have shown conflicting results (5,8,10,24,33,37). Reasons for these contradictory findings could reside in methodological issues such as sample characteristics (e.g., individuals_ baseline level of cardiac autonomic modulation and aerobic fitness) or the use of different exercise recovery protocols. Given that resting cardiac vagal modulation and postexercise HR recovery
utonomic nervous system dysfunction, indicated by relative or absolute decrease in vagal activity or an increase in sympathetic activity, has been linked to cardiovascular disease (34,39) and to an increased risk of potentially lethal cardiac arrhythmias in healthy individuals (21,27). Moreover, an inadequate HR recovery after physical exertion is indicative of reduced cardiac vagal reactivation, which is also a predictor of mortality (12). Aerobic exercise is considered to be a nonpharmacological intervention that may increase cardiac vagal control and promote cardioprotection (4,6). Nonetheless, the literature
BASIC SCIENCES
depend on parasympathetic activity, it would seem probable that aerobic training would induce favorable adaptations in these autonomic markers (40). Nevertheless, there is limited evidence about the effects of training on postexercise autonomic control of HR. Moreover, it is not well known if there is an association between possible adaptations of resting and postexercise vagal indices to aerobic training. In brief, adaptations in resting cardiac vagal modulation and postexercise vagal reactivation are expected with aerobic training. However, whether these effects are influenced by subjects_ initial level of autonomic control and the extent which those markers are mutually related have not been investigated. Because autonomic adaptations to exercise can be assessed at rest or within the immediate recovery from an exercise bout, it is important to determine whether these different outcomes are similarly affected by chronic exercise. If this is not the case, to evaluate the vagal activity in only one condition could lead to misinterpretation of the training effects on cardiac vagal modulation. Therefore, the present study investigated the effects of 12-wk aerobic training on resting cardiac vagal control and postexercise vagal reactivation as modulated by cardiorespi˙ O2peak) among healthy young men. We also ratory fitness (V examined whether the observed training-induced adaptations in resting cardiac vagal modulation influence postexercise vagal reactivation and whether this relationship is affected by baseline levels of resting vagal modulation. We hypothesized that (a) the increase in postexercise vagal reactivation would be positively related to an increase in resting parasympathetic activity achieved through exercise training and (b) the association between the gains in resting and postexercise vagal control with training would be dependent on the baseline pretraining level of resting vagal modulation.
METHODS Ethical approval. All subjects provided written informed consent. Procedures were approved by institutional ethics committee and registered at the Brazilian National Council of Ethics in Research (CAAE/CONEP: 0038.0.312.000-07). Subjects. Sixty-two healthy, young men (18–22 yr) volunteered to take part in the study. Candidates were screened for health problems and interviewed about their habitual physical
activity. Subject exclusion criteria were as follows: (a) chronic medical conditions and diseases, (b) current smoker, (c) use of any cardiovascular medication, (d) BMI 930 kgImj2, and (e) participation in regular exercise training (Q2 dIwkj1) within the last year. After the initial evaluation, 49 young healthy men qualified for the study. Despite the fact that they were not currently training, all of them had previous history of recreational physical activity. Subjects were randomized into training (n = 25; 19.4 T 1.4 yr, 172.6 T 8.3 cm, 66.2 T 9.2 kg, ˙ O2peak = 50.8 T 5.1 mLIkgj1Iminj1; mean T SD) and V control (n = 24; 19.0 T 0.6 yr, 175.3 T 6.7 cm, 67.4 T 10.7 kg, ˙ O2peak = 49.9 T 6.1 mLIkgj1Iminj1) groups with similar V characteristics (P 9 0.30). This selection was performed in pairs from numbers assigned to each individual from the database using the program Research Randomizer (Version 4.0; http://www.randomizer.org). Because one of the purposes of the present study was to verify the effects of baseline parasympathetic control differences in the individuals_ training response, after the training period, subjects were assigned into four different groups for comparative analyses. Selection criterion was the median value of subjects_ initial level of resting vagal modulation, computed as the HF band of HRV spectral analysis expressed in normalized units (n.u.). Therefore, participants in the training condition were assigned into two groups to further analysis, one with lower (TL; n = 13, HF = 48.1 T 8.2 n.u.) and other with higher (TH; n = 12, HF = 63.1 T 5.9 n.u.) baseline HF power. Likewise, to allow comparisons with the trained subjects, participants in the control group were also assigned to groups with lower (CL; n = 12, HF = 47.1 T 7.5 n.u.) and higher HF power (CH; n = 12, HF = 65.5 T 8.3 n.u.). Two subjects in TL and one in TH dropped out of the training program and were not considered for the analyses. In the control group, three subjects from both CL and CH were excluded because of engagement in training programs (n = 4) or medical problems (n = 2). Final analyses included 11 subjects in each training group and 9 subjects in each control group. Considering all variables analyzed and the final sample size, the lowest test power to detect statistical differences at P e 0.05 was 0.83. Experimental protocol. Figure 1 summarizes the study design. All participants underwent a pretraining assessment that included measurement of anthropometric characteristics
FIGURE 1—Experimental protocol design. Laboratory assessments were conducted on weeks 0 and 13. GXT, Graded exercise test; HRRes, HR reserve; HRV, HR variability.
1160
Official Journal of the American College of Sports Medicine
http://www.acsm-msse.org
Copyright © 2015 by the American College of Sports Medicine. Unauthorized reproduction of this article is prohibited.
AEROBIC TRAINING AND VAGAL CONTROL
Considering that HRV describes the HR dynamics through measures in time and frequency domains (38) and that vagal-related HRV indices mostly reflect the magnitude of modulation in parasympathetic outflow, as opposed to overall parasympathetic tone (19), in the present study, markers in the time domain included resting HR (an indicator of autonomic nervous system tonus) and an index describing the HRV (the square root of the mean squared difference of successive R–R intervals—rMSSD), reflecting vagal dynamics in short periods (38). In the frequency domain, a fast Fourier transform (Welch method) with a Hanning window and 50% overlap was used to estimate the power density spectrum of R–R interval variability using a customized routine (MATLAB R12; MathWorks, Inc., Natick, MA) (36). Spectral power was obtained by the integration of power spectrum density function in the very-low-frequency (0.0033–0.04 Hz), low-frequency (LF; 0.04–0.15 Hz), and high-frequency (HF; 0.15–0.40 Hz) components. The HF power was used as an index of vagal activity, and the LF power was adopted as a primarily marker of sympathetic nervous system activity (28,38). The spectral values were expressed as normalized units (n.u.), calculated by dividing the power of each component by the total variance from which the very-low-frequency component had been subtracted and multiplied by 100. This procedure is acknowledged to increase the precision of data about the relative contributions of LF and HF bands to total variability and autonomic balance (38). Assessment and analysis of postexercise vagal reactivation. Within 5 s of GXT termination, participants were placed in the supine position. An HR monitor (Polar RS800; PolarElectro, Kempele, Finland) and transmitter belt (Wearlink Coded; Polar Electro) were used to record the beat-by-beat HR during the 5-min recovery period (32). The R–R interval series were transferred to a personal computer using the Polar Precision Performance SW 4.3 software (Polar Electro). To assess vagal reactivation in the first 5 min after the end of GXT, the time domain HRV index rMSSD was calculated sequentially (MATLAB R12; MathWorks, Inc.) at each 30 s of the recovery period (rMSSD30s) (16,31). To ensure the stability of data, in addition to the 30-s segments analysis of postexercise vagal reactivation, rMSSD was also calculated as the mean value within minutes 3–5 of the recovery period (rMSSD3–5min) (7). In addition, HR recovery was calculated by taking the absolute difference between the final HR at exercise completion and the HR recorded after 60 s of recovery (HRR60s). To avoid interference in the natural return of HR to baseline, participants maintained spontaneous breathing during recovery. Nevertheless, their respiratory rate remained in the HF range (90.15 Hz) during the recovery period and did not differ significantly both before and after training (P 9 0.05). Aerobic training program. The aerobic training protocol consisted of three 40-min running sessions per week for 12 wk, at an intensity corresponding to 75%–85% of HR reserve (HRRes). To monitor exercise intensity, TL and TH
Medicine & Science in Sports & Exercised
Copyright © 2015 by the American College of Sports Medicine. Unauthorized reproduction of this article is prohibited.
1161
BASIC SCIENCES
and resting autonomic control by means of HR variability (HRV) (38). Subjects were instructed to avoid drinking coffee or other caffeinated beverages for 12 h and to refrain from exercising for 48 h before the pretraining assessments. The HRV was assessed with subjects seated in a silent and thermoneutral room (23-C) for 30 min, after which a continuous ECG was obtained for 5 min with subjects in the supine position. Familiarization with running on a treadmill was done on two occasions before the graded exercise test (GXT). The GXT was used to determine peak oxygen con˙ O2peak) and postexercise vagal reactivation. sumption (V After completion of pretraining assessments, subjects assigned in the training group underwent 12 wk of aerobic training. Controls were advised to perform their normal activities of daily living and not to engage in exercise training, considered as two or more exercise sessions per week. Control subjects kept a physical activity diary to record all exercises performed, which was verified by the same researcher on a weekly basis. At the end of the training period, all pretraining assessments were repeated at the same time of the day (08:30–11:30 a.m.). Determination of peak O2 consumption. Subjects performed a cardiopulmonary treadmill GXT using a ramp protocol, which started with a 5-min warm-up at 8.0 kmIhj1 followed by a gradual work rate increase of 0.8 kmIhj1 every 1 min and grade held constant at 1%. Minute ventilation and gas exchange were measured breath-by-breath and registered as the mean value of each 5 s (Ultima CardiO2; Medical Graphics Corporation, St. Paul, MN) at an ambient temperature of 23-C. The GXT was terminated if at least three out of the four following criteria were attained: (a) maximum voluntary exhaustion defined by attaining a 10 on the Borg CR-10 scale, (b) 95% of the predicted HRmax (220 j age) or presence of an HR plateau ($HR between two consecutive work rates e4 beats per minute), (c) presence of a ˙ O2 plateau ($V ˙ O2 between two consecutive work rates V j1 of G2.1 mLIkg Iminj1), and (d) maximal RER (RERmax) 91.10. The highest value of O2 consumption measured during ˙ O2peak. the test was used as V Assessment and analysis of resting autonomic control. Continuous ECG signals were acquired during 5 min of rest using Ag/AgCl electrodes placed on the participant_s chest in a standard three-lead configuration (C5) by means of a digital data acquisition system (Biopac MP-150; Biopac Systems, Goleta, CA) in the supine position. Because of the influence of respiratory rate on HRV indices, the subject_s breathing frequency (thoracic and abdominal movements) was recorded by two respiratory effort transducers (TSD201; Biopac Systems) during spontaneous breathing. The respiratory rate of all subjects was in the high-frequency range (90.15 Hz). All signals were acquired at a sampling frequency of 1000 Hz using the AcqKnowledge 4 software program (Biopac Systems). Before analysis, all R–R intervals were visually inspected for artifacts to make interpolation corrections on the 5-min sequences (38). This was necessary in G1% of the sequences in each subject.
TABLE 1. Pretraining subjects_ characteristics (mean T SD) among the aerobic training and control high and low HF power groups. Aerobic Training Age, yr Height, cm Weight, kg Body fat, %
Control
TL (n = 11)
TH (n = 11)
CL (n = 9)
CH (n = 9)
19.2 T 172.6 T 66.2 T 12.1 T
19.5 T 174.3 T 66.4 T 10.8 T
19.1 T 0.3 174.4 T 5.6 68.3 T 12.1 12.0 T 6.4
18.9 T 177.2 T 70.6 T 12.1 T
1.0 8.3 9.2 5.4
1.2 6.8 5.0 4.0
0.6 6.8 12.1 7.4
BASIC SCIENCES
There were no significant differences between groups (P 9 0.05). CH, control group with baseline high levels of HF power; CL, control group with baseline low levels of HF power; HF, HR variability high-frequency spectral power; TH, training group with baseline high levels of HF power; TL, training group with baseline low levels of HF power.
subjects wore a telemetric HR monitor (Polar RS800; Polar Electro) during all training sessions. The duration of exercise and average HRRes were recorded, and the training impulse (TRIMP) in each session was calculated to ensure that the prescribed training intensity and volume were achieved (2). TRIMP integrates in a single variable both intensity and volume of training, being determined by multiplying the duration of the exercise bout (TD) by the average HR Res during the exercise, as follows: TRIMP = TD %HRRes 0.64 e(1.92 %HRres). Weekly training load was computed individually as the average values of TD and %HRRes and as the sum of TRIMP values of the three sessions performed in a given week. Statistical analysis. Normal distribution of data was tested by the Shapiro–Wilk test. Considering that all variables showed normal distribution, differences between groups in subjects_ characteristics at baseline were analyzed using separate one-way ANOVA. The TD, %HRRes, and TRIMP comparisons between TL and TH during the 12-wk training protocol were made by means of two-way ANOVA with repeated measures, with one between-group factor (groups = TL and TH) and one within-group factor (time = 12 wk of training). Within- and between-groups differences before (Pre) and after (Post) training with regard to vagal control indices at rest, cardiorespiratory fitness, HR recovery after 60 s of the end of exercise (HRR60s), and absolute index of postexercise vagal reactivation (rMSSD3–5min) were analyzed using twoway repeated-measures ANOVA with one between-group factor (groups = TL, TH, CL, and CH) and one within-group
factor (time = Pre and Post). The time course of variation in rMSSD30s as a result of training was analyzed using a threeway repeated measures ANOVA with one between-group factor (initial resting vagal condition; TL and TH) and two within-group factors, the assessment period (Pre and Post) and recovery time (10 measurements over a 5-min recovery period). Between-groups differences concerning the relative variation in resting HF spectral power ($% HF, n.u.) and relative change in postexercise vagal reactivation with training ($% rMSSD3–5min) [$%variable = ([posttraining j pretraining] / pretraining) 100] were tested using one-way ANOVA. In all cases, the Tukey HSD post hoc test was used to identify differences when the ANOVA models detected significant F ratios. Pearson correlation tested the correlation between $% HF (n.u.) and $% rMSSD3–5min in TL and TH groups. The magnitude of change after versus before training was estimated by calculating the effect size (ES), considering values G0.2 as having a small effect size, 0.5 as moderate effect size, and Q0.8 as high effect size (11). P e 0.05 was considered significant. All calculations were performed using Statistica 7.1 software (StatSoft, Tulsa, OK), and results are presented as mean T SD.
RESULTS Pretraining subjects_ characteristics. All groups were matched with regard to age, height, weight, and body fat (P 9 0.05; Table 1). As presented in Table 2, the pretraining HF
TABLE 2. Resting HR variability, cardiorespiratory fitness at peak exercise, and HR recovery after 60 s of exercise before (Pre) and after (Post) the aerobic training program. TL (n = 11) Pre HR, bpm rMSSD, ms HF, n.u. LF, n.u. LF/HF V˙O2peak, mLIkgj1Iminj1 RERmax HRmax, bpm HRR60s, bpm
61.8 T 79.1 T 48.1 T 48.2 T 1.08 T 50.6 T 1.26 T 202.4 T 54.2 T
TH (n = 11) Post
9.2 30.6 8.2 8.8 0.47 5.4 0.07 6.0 8.4
53.9 131.8 56.6 40.3 0.79 56.3 1.23 201.1 62.1
T 5.9* T 83.6*** T 13.3* T 12.8 T 0.40 T 5.3*** T 0.11 T 7.2 T 7.9*
Pre 55.1 108.2 63.1 33.2 0.54 51.1 1.22 199.5 55.8
T 6.3** T 42.2 T 5.9** T 5.7** T 0.13** T 5.1 T 0.09 T 8.8 T 6.4
CL (n = 9) Post
53.4 111.1 58.9 37.9 0.71 54.6 1.21 201.0 62.7
T 6.2 T 23.8 T 12.3 T 11.6 T 0.36 T 4.6*** T 0.05 T 7.7 T 6.8*
Pre 58.3 86.5 47.1 49.1 1.11 50.7 1.17 198.4 54.5
T 7.5 T 61.8 T 7.5 T 7.5 T 0.47 T 7.7 T 0.07 T 6.1 T 7.4
CH (n = 9) Post
56.7 101.3 46.4 49.1 1.18 50.5 1.17 196.4 54.8
T 7.6 T 57.7 T 10.5** T 10.3 T 0.40** T 6.2** T 0.05 T 10.2 T 8.3
Pre 56.1 118.0 65.5 31.6 0.50 49.1 1.21 198.7 56.1
T 6.1 T 67.9 T 8.3**** T 7.5**** T 0.17**** T 4.4 T 0.08 T 6.8 T 10.7
Post 55.4 109.4 60.6 36.7 0.69 49.9 1.22 196.6 56.5
T 8.8 T 52.8 T 14.7**** T 13.8**** T 0.40**** T 5.0 T 0.08 T 6.2 T 6.9
Values are means T SD. CH, control group with baseline high levels of HF power; CL, control group with baseline low levels of HF power; HF, high-frequency power; HRmax, maximal HR during graded exercise test; HRR60s, HR recovery after 60 s of the end of exercise; LF, low-frequency power; RERmax, maximal RER; rMSSD, root mean square of successive normal R–R interval differences; TH, training group with baseline high levels of HF power; TL, training group with baseline low levels of HF power; V˙O2peak, peak oxygen consumption. *P G 0.05 versus Pre. **P G 0.05 compared to the corresponding TL value. ***P G 0.01 versus Pre. ****P G 0.05 compared to the corresponding CL value.
1162
Official Journal of the American College of Sports Medicine
http://www.acsm-msse.org
Copyright © 2015 by the American College of Sports Medicine. Unauthorized reproduction of this article is prohibited.
FIGURE 3—Mean values T SD of (A) the relative change in highfrequency power of HRV ($% HFn.u.) and (B) the relative change in the square root of the mean squared difference of successive R–R intervals during the last 2 min of the 5-min postexercise recovery time ($% rMSSD3–5min) for the low (TL) and high (TH) baseline vagal modulation training groups and the low (CL) and high (CH) baseline vagal modulation control groups, as a result of training. Significant differences are disclosed in the figure.
FIGURE 2—Mean values of the square root of the mean squared difference of successive R–R intervals, measured in successive 30-s intervals (rMSSD30s), presented by training groups of low (TL) and high (TH) baseline vagal modulation, before (Pre) and after (Post) the training protocol. *Significant difference in comparison to rMSSD30s at 30 s (P G 0.05). †Significant difference between Pre and Post values within groups (P G 0.05). To improve figure clarity, the error bars were omitted.
Training effects on postexercise vagal reactivation. The training program induced a significant increase (P G 0.05) in HRR60s in TL and TH (Table 2). No changes were observed in CL and CH (P 9 0.95). Figure 2 illustrates the time course of rMSSD30s in TL and TH during the 5-min recovery period after GXT at Pre and Post training. The rMSSD30s increased in TL from Pre to Post at 60, 90, 120, 150, 180, and 270 s of recovery and in TH at 60 s (P G 0.001). There was also a significant effect for the recovery time within each group (5 min; P G 0.001). After training, the rMSSD30s increased significantly at 60 s of recovery in both TL and TH, whereas before training, a significant increase was observed only after 180 s in TL and 120 s in TH. The rMSSD30s was not different between TL and TH either before or after training (P = 0.99). The absolute vagal reactivation index (i.e., rMSSD3–5min) increased from Pre to Post training in TL (Pre 5.5 T 4.4 vs Post 8.4 T 4.3, P G 0.001, ES = 1.24) and TH (Pre 7.7 T 4.0 vs Post 10.1 T 3.4, P G 0.01, ES = 0.74), but no Pre versus Post change was detected in the control groups (CL: Pre 5.0 T 1.9 vs Post 5.2 T 3.0, P = 0.99; CH: Pre 6.6 T 2.9 vs Post 6.9 T 3.6, P = 0.96). No statistical difference was found
AEROBIC TRAINING AND VAGAL CONTROL
Medicine & Science in Sports & Exercised
Copyright © 2015 by the American College of Sports Medicine. Unauthorized reproduction of this article is prohibited.
1163
BASIC SCIENCES
(n.u.) was lower (P G 0.001), whereas LF (n.u.) (P G 0.001) and LF/HF (P G 0.01) were higher in TL and CL compared to groups with higher vagal modulation (TH and CH). The four ˙ O2peak at baseline (P = 0.88). groups had similar V Aerobic training program adherence. All subjects in the training groups accomplished the planned 36 exercise sessions. The computed TD (P = 0.89), %HRRes (P = 0.93), and TRIMP (P = 0.84) were similar across TL and TH along the 12-wk protocol. The mean TD was 40.1 T 0.3 min for TL and 40.1 T 0.2 min for TH, and the mean %HRRes of the training sessions was 77.7% T 0.7 % for TL and 76.6% T 1.1 % for TH. The mean weekly TRIMP (A.U.) was 257.6 T 9.5 for TL and 260.3 T 10.3 for TH, meaning that these groups underwent the same training load during the study. Training effects on resting cardiac vagal modulation and cardiorespiratory fitness. HR decreased (P G 0.05) and rMSSD increased (P G 0.01, ES = 0.85) from Pre to Post in TL (Table 2). After training, TL and TH exhibited similar resting HR (P = 0.82) and rMSSD (P = 0.38). With regard to the spectral components of HRV, HF (n.u.) increased from Pre to Post in TL (P = 0.03, ES = 1.05) and did not change in TH (P = 0.91). Considering LF (n.u.) and LF/HF measurements, differences observed in Pre between TL and TH (P G 0.05) were not detected in Post (P 9 0.56). After training, LF/HF was lower in TL compared to CL (P = 0.04, ES = 0.98). ˙ O2peak increased from Pre to Post (P G 0.001) in TL The V and TH but not in CL and CH (P = 0.95). As a result of ˙ O2peak was similar (P = 0.70) training, the relative increase in V in TL (11.7% T 5.4%, ES = 0.82) and TH (7.0% T 2.9%, ES = 0.75) and higher (P G 0.03) than in CL (0.8% T 3.9%, ES = 0.03) and CH (1.8% T 6.2%, ES = 0.17), respectively.
BASIC SCIENCES
FIGURE 4—Relationships between the relative changes observed with training in the high-frequency component of HR variability (HFn.u.) and in the square root of the mean squared difference of successive R–R intervals during the last 2 min of the 5-min postexercise recovery time (rMSSD3–5min). Correlation coefficients were calculated based on data of low (TL, n = 11; empty squares) and high (TH, n = 11; full circles) baseline vagal modulation training groups.
between groups with regard to rMSSD3–5min in either Pre or Post experiment assessments (P 9 0.83). Changes in resting and postexercise vagal modulation with training. Figure 3A shows the betweengroups comparison for the relative variation in resting HF spectral power ($% HF, n.u.) due to training. The $% HF was higher in TL than in TH (P = 0.01) and CL (P G 0.03). The relative increase in $% rMSSD3–5min (Fig. 3B) was higher in TL and TH than in CL (P G 0.001) and CH (P G 0.03), respectively. Also, the increase in $% rMSSD3–5min was greater in TL than in TH (P G 0.02). Correlations among changes in resting and postexercise vagal modulation. Relationships between the relative variation in resting HF spectral power ($% HF, n.u.) and relative change in postexercise vagal reactivation ($% rMSSD3–5min) with training are represented in Figure 4. There was a significant positive correlation between $% HF (n.u.) and $% rMSSD3–5min in TL (r = 0.63, P = 0.04), accounting for 36% of the variation. On the other hand, $% HF (n.u.) and $% rMSSD3–5min were not associated in TH (r = j0.30, P = 0.37).
DISCUSSION This investigation examined the influence of moderate-tovigorous aerobic training on resting and postexercise vagal modulation among healthy young men with different levels of pretraining cardiac autonomic modulation. The 12-wk training protocol induced similar aerobic gains in both trained groups (TL and TH). On the other hand, resting vagal modulation (i.e., HF n.u.) increased only in subjects with lower pretraining HF power (TL; mean variation of 23%), whereas in those with higher levels of vagal modulation (TH), no significant variation was detected. A different behavior, however, was observed for the postexercise vagal reactivation, which improved in TL and TH after training
1164
Official Journal of the American College of Sports Medicine
regardless of the differences in pretraining resting vagal control. These major findings suggest that postexercise vagal reactivation gains with aerobic training may occur in the absence of improvements in resting cardiac vagal modulation. Recent studies (7,23,41) investigating the effects of training on autonomic modulation demonstrated that, when the training load is adequate, a positive response is generally observed in both cardiorespiratory fitness and cardiac vagal modulation. However, others (14,25) have reported an in˙ O2max with training but not in resting HF power crease in V in individuals who underwent aerobic training. These controversial observations suggest that autonomic adaptations to aerobic training may rely not only on the training intensity and volume but also on subjects_ initial aerobic capacity and autonomic control. ˙ O2max It should be highlighted that even when a gain in V is observed without an improvement in HF power (n.u.), this fact does not necessarily imply a lack of autonomic adaptation with training. The HF component of HRV reflects the magnitude of the fluctuation of cardiac vagal control and not the vagal tone level (19). When cardiac vagal control reaches high levels as a result of training, for example, further increasing in vagal tone may eliminate respiratory modulation of the cardiac rhythm, leading to a saturation of HF power (22). When this is the case, it is very difficult to detect a subsequent rise in cardiac vagal tone by means of HRV indices at rest. Accordingly, our findings indicate that the assessment of parasympathetic control should be carried out at rest and during the immediate recovery from an exercise bout when analyzing the effects of training on vagal modulation. Performing the assessments only in resting conditions may provide a limited interpretation of the data. With respect to the effects of training on postexercise vagal reactivation, both training groups showed an increase in HRR60s, rMSSD30s, and rMSSD3–5min within the 5-min recovery period after GXT. The faster recovery in rMSSD30s and the increase in HRR60s and rMSSD3–5min due to aerobic training reflect an improvement in postexercise cardiac vagal reactivation, reinforcing the potential value of exercise in raising parasympathetic activity (10,35,37). In addition, our findings showed that resting vagal modulation increased only in TL. This suggests that parasympathetic reactivation after a maximal bout of exercise might be improved by aerobic training, even in the absence of resting vagal modulation increase as observed in TH. It is well known that both HRV and vagal reactivation after exercise provide useful information about the autonomic balance. However, the present findings suggest that there may be some differences in the physiological determinants underlying the responses of these markers to training. Dewland et al. (13) addressed this issue using acetylcholinesterase inhibition and suggested that HRV would mostly express fluctuations of vagal outflow to the heart, whereas the recovery of HR after an effort would reflect the mean cholinergic signaling at the sinoatrial nodal junction. Actually, several authors have reported an increase of vagal reactivation with exercise training or as a consequence of
http://www.acsm-msse.org
Copyright © 2015 by the American College of Sports Medicine. Unauthorized reproduction of this article is prohibited.
AEROBIC TRAINING AND VAGAL CONTROL
recovery was assessed by rMSSD30s, which shows the inherent variability in short tracings of 30 s and is considered as a marker of vagal reactivation. Despite differences in the methodological design (cross-sectional longitudinal), the results of both studies emphasize that indices of vagal reactivity after exercise would be more sensible to show autonomic adaptations to training (or associated to higher cardiorespiratory fitness) than resting HRV alone. Limitations of this report warrant consideration. The spectral analysis of HRV was used to assess changes in cardiac autonomic control. Although the contribution to HRV modulation by sympathetic, parasympathetic, and respiratory influences can be identified by changes in specific frequency bands, neither LF (0.04–0.15 Hz) nor HF (0.15–0.40 Hz) power components can be considered as exclusive markers of sympathetic and parasympathetic autonomic controls, respectively. In addition, it is not known whether the autonomic reactivation patterns observed in this study may be generalized to the recovery period from exercise protocols with different characteristics. Finally, a potential ceiling effect on the response of resting vagal modulation to training was observed in TH subjects. The acquisition of ECG signals for longer periods (i.e., 24 h) could avoid this effect, identifying the point of saturation of HF power in relation to R–R interval variations with training (22). Alternatively, the use of sympathetic pharmacological blockade, leaving the HR control predominantly under parasympathetic influence, with subsequent maneuvers of parasympathetic stimulation and withdraw with vasoactive drugs (15), could have helped to evaluate the relationship between HRV indices and parasympathetic control of the heart. This information might have provided additional insight on the effects of resting vagal tone adjustments upon postexercise vagal modulation. However, this analysis was beyond the scope of the present investigation.
CONCLUSIONS In conclusion, a 12-wk moderate-to-vigorous aerobic training program increased resting cardiac vagal control (rMSSD and HF power) and postexercise vagal reactivation (HRR60s, rMSSD30s, and rMSSD3–5min) in healthy young men. In addition, it was observed that resting cardiac vagal modulation and postexercise vagal reactivation may respond differently to aerobic training, depending on the subjects_ pretraining level of resting vagal control. An increase in postexercise vagal reactivation may be observed, even in the absence of resting vagal modulation improvement. These findings have direct implications on the interpretation of the effects of endurance training on the autonomic control. This research was supported in part by the Pro-Defesa Project (Brazil Ministry of Defense and CAPES) and by grants from the Carlos Chagas Filho Foundation for the Research Support in Rio de Janeiro State (FAPERJ, process E-26/150.751/2007) and from
Medicine & Science in Sports & Exercised
Copyright © 2015 by the American College of Sports Medicine. Unauthorized reproduction of this article is prohibited.
1165
BASIC SCIENCES
high level of aerobic fitness (10,24,33,37), but there is limited evidence about the simultaneous variation that should occur with resting vagal modulation in those situations. In the present study, the magnitude of training effects on vagal HRV indices varied as a function of subjects_ pretraining level of resting vagal modulation (see Fig. 3). Whereas TL subjects showed an increase in the relative variation of resting HF (n.u.) and in the relative change of postexercise vagal reactivation (rMSSD3–5min%), TH subjects presented a different trend, with an increase only in the relative change of postexercise vagal reactivation and no variation in resting HFn.u.%. In addition, $% HF (n.u.) and $% rMSSD3–5min were positively correlated in TL subjects (r = 0.61, P = 0.04), whereas no significant correlation was detected in TH subjects. Our hypothesis that the improvement in postexercise vagal reactivation with training may be associated with the increase in resting autonomic modulation can be therefore accepted, but this effect seems to depend on subjects_ initial level of resting autonomic control. Our findings indicate that subjects with high levels of pretraining resting parasympathetic control may improve postexercise vagal reactivation without any increase in resting vagal modulation. In brief, the improvement in postexercise vagal reactivation with training may not necessarily be related to an increase in resting parasympathetic control. As aforementioned, most studies examining the relationship between resting vagal modulation and autonomic recovery of HR after exercise are crosssectional and analyzed this association measuring resting vagal control after the end of exercise (9,17). In such conditions, cardiac vagal modulation may still be influenced by the previous effort (30). In the few available longitudinal studies (7,10), different training strategies and assessment methods have been applied, which compromises comparison between their results. Buchheit et al. (10) trained participants in supramaximal intensity for 9 wk. Postexercise vagal reactivation, as reflected by rMSSD30s, and normalized HF power measured 5–10 min after exercise were not different from pretraining levels. In another study by these investigators (7), the resting parasympathetic control was positively associated with postexercise vagal reactivation only in subjects who ˙ O2max in response to training versus showed an increase in V those whose aerobic capacity did not change. In the present study, vagal reactivation after exercise was evaluated immediately after the end of the effort until 5 min of recovery, and all trained subjects were responsive to the training protocol ˙ O2peak). In this scenario, the (i.e., showed improvement in V relationship between resting vagal modulation and postexercise vagal reactivation was observed only in subjects with baseline low levels of resting vagal control (see Fig. 4). More recently, Lee and Mendoza (24) showed that resting HRV and HR recovery (HRrec) would be dissociated in highly fit men and women. In that cross-sectional study, HRrec was measured as the reduction of HR in the first minute after exercise cessation and spectral analysis of HRV was carried out at rest. In our study, vagal modulation at rest was also measured by the HF power of HRV, and parasympathetic
the National Council of Technological and Scientific Development (CNPq, process 305729/2006-3, process 420122/2005-2, and process 481434/2008-9).
The authors have no conflict of interest to declare. The results of the present study do not constitute endorsement by the American College of Sports Medicine.
BASIC SCIENCES
REFERENCES 1. Aubert AE, Seps B, Beckers F. Heart rate variability in athletes. Sports Med. 2003;33(12):889–919. 2. Banister EW. Modeling elite athletic performance. In: MacDougall JD, Wenger HA, Green, HJ, eds. Physiological Testing of the High-Performance Athlete. Champaign (IL): Human Kinetics; 1991. pp. 403–24. 3. Berkoff DJ, Cairns CB, Sanchez LD, Moorman CT 3rd. Heart rate variability in elite American track-and-field athletes. J Strength Cond Res. 2007;21(1):227–31. 4. Billman GE. Aerobic exercise conditioning: a nonpharmacological antiarrhythmic intervention. J Appl Physiol. 2002;92(2):446–54. 5. Bosquet L, Gamelin FX, Berthoin S. Is aerobic endurance a determinant of cardiac autonomic regulation? Eur J Appl Physiol. 2007;100(3):363–9. 6. Buch AN, Coote JH, Townend JN. Mortality, cardiac vagal control and physical training—what’s the link? Exp Physiol. 2002;87(4):423–35. 7. Buchheit M, Chivot A, Parouty J, et al. Monitoring endurance running performance using cardiac parasympathetic function. Eur J Appl Physiol. 2010;108(6):1153–67. 8. Buchheit M, Gindre C. Cardiac parasympathetic regulation: respective associations with cardiorespiratory fitness and training load. Am J Physiol Heart Circ Physiol. 2006;291(1):H451–8. 9. Buchheit M, Laursen PB, Ahmaidi S. Parasympathetic reactivation after repeated sprint exercise. Am J Physiol Heart Circ Physiol. 2007;293(1):H133–41. 10. Buchheit M, Millet GP, Parisy A, Pourchez S, Laursen PB, Ahmaidi S. Supramaximal training and postexercise parasympathetic reactivation in adolescents. Med Sci Sports Exerc. 2008; 40(2):362–71. 11. Cohen J. Statistical Power Analysis for the Behavioral Sciences. 2nd ed. Hillsdale (NJ): Lawrence Erlbaum Associates; 1988. 12. Cole CR, Blackstone EH, Pashkow FJ, Snader CE, Lauer MS. Heart-rate recovery immediately after exercise as a predictor of mortality. N Engl J Med. 1999;341(18):1351–7. 13. Dewland TA, Androne AS, Lee FA, Lampert RJ, Katz SD. Effect of acetylcholinesterase inhibition with pyridostigmine on cardiac parasympathetic function in sedentary adults and trained athletes. Am J Physiol Heart Circ Physiol. 2007;293(1): H86–92. 14. Gamelin F, Berthoin S, Sayah H, Libersa C, Bosquet L. Effect of training and detraining on heart rate variability in healthy young men. Int J Sports Med. 2007;28(7):564–70. 15. Goldberger JJ, Challapalli S, Tung R, Parker MA, Kadish AH. Relationship of heart rate variability to parasympathetic effect. Circulation. 2001;103:1977–83. 16. Goldberger JJ, Le FK, Lahiri M, Kannankeril PJ, Ng J, Kadish AH. Assessment of parasympathetic reactivation after exercise. Am J Physiol Heart Circ Physiol. 2006;290(6):H2446–52. 17. Goulopoulou S, Heffernan KS, Fernhall BO, Yates G, BaxterJones ADG, Unnithan VB. Heart rate variability during recovery from a Wingate test in adolescent males. Med Sci Sports Exerc. 2006;38(5):875–81. 18. Grant CC, Clark JR, van Rensburg DC, Viljoen M. Relationship between exercise capacity and heart rate variability: supine and in response to an orthostatic stressor. Auton Neurosci. 2009; 151(2):186–8.
1166
Official Journal of the American College of Sports Medicine
19. Hedman AE, Hartikainen JE, Tahvanainen KU, Hakumaki MO. The high frequency component of heart rate variability reflects cardiac parasympathetic modulation rather than parasympathetic Ftone_. Acta Physiol Scand. 1995;155(3):267–73. 20. Heydari M, Boutcher YN, Boutcher SH. High-intensity intermittent exercise and cardiovascular and autonomic function. Clin Auton Res. 2013;23(1):57–65. 21. Jouven X, Empana J-P, Schwartz PJ, Desnos M, Courbon D, Ducimetie`re P. Heart-rate profile during exercise as a predictor of sudden death. N Engl J Med. 2005;352(19):1951–8. 22. Kiviniemi AM, Hautala AJ, Makikallio TH, Seppanen T, Huikuri HV, Tulppo MP. Cardiac vagal outflow after aerobic training by analysis of high-frequency oscillation of the R–R interval. Eur J Appl Physiol. 2006;96(6):686–92. 23. Kiviniemi AM, Tulppo MP, Eskelinen JJ, et al. Cardiac autonomic function and high-intensity interval training in middle-age men. Med Sci Sports Exerc. 2014;46(10):1960–7. 24. Lee CM, Mendoza A. Dissociation of heart rate variability and heart rate recovery in well-trained athletes. Eur J Appl Physiol. 2012;112(7):2757–66. 25. Leicht AS, Allen GD, Hoey AJ. Influence of age and moderateintensity exercise training on heart rate variability in young and mature adults. Can J Appl Physiol. 2003;28(3):446–61. 26. Loimaala A, Huikuri H, Oja P, Pasanen M, Vuori I. Controlled 5-mo aerobic training improves heart rate but not heart rate variability or baroreflex sensitivity. J Appl Physiol. 2000;89(5):1825–9. 27. Lombardi F, Makikallio TH, Myerburg RJ, Huikuri HV. Sudden cardiac death: role of heart rate variability to identify patients at risk. Cardiovasc Res. 2001;50(2):210–7. 28. Malliani A, Julien C, Billman GE, et al. Cardiovascular variability is/is not an index of autonomic control of circulation. J Appl Physiol. 2006;101(2):684–8. 29. Martinmaki K, Hakkinen K, Mikkola J, Rusko H. Effect of lowdose endurance training on heart rate variability at rest and during an incremental maximal exercise test. Eur J Appl Physiol. 2008; 104(3):541–8. 30. Mourot L, Bouhaddi M, Tordi N, Rouillon JD, Regnard J. Shortand long-term effects of a single bout of exercise on heart rate variability: comparison between constant and interval training exercises. Eur J Appl Physiol. 2004;92(4–5):508–17. 31. Ng J, Sundaram S, Kadish AH, Goldberger JJ. Autonomic effects on the spectral analysis of heart rate variability after exercise. Am J Physiol Heart Circ Physiol. 2009;297(4):H1421–8. 32. Nunan D, Donovan GAY, Jakovljevic DG, Hodges LD, Sandercock GRH, Brodie DA. Validity and reliability of short-term heart-rate variability from the Polar S810. Med Sci Sports Exerc. 2009;41(1): 243–50. 33. Otsuki T, Maeda S, Iemitsu M, et al. Postexercise heart rate recovery accelerates in strength-trained athletes. Med Sci Sports Exerc. 2007;39(2):365–70. 34. Schroeder EB, Liao D, Chambless LE, Prineas RJ, Evans GW, Heiss G. Hypertension, blood pressure, and heart rate variability: the Atherosclerosis Risk in Communities (ARIC) study. Hypertension. 2003;42(6):1106–11. 35. Seiler S, Haugen O, Kuffel E. Autonomic recovery after exercise in trained athletes: intensity and duration effects. Med Sci Sports Exerc. 2007;39(8):1366–73.
http://www.acsm-msse.org
Copyright © 2015 by the American College of Sports Medicine. Unauthorized reproduction of this article is prohibited.
36. Soares PP, Moreno AM, Cravo SL, Nobrega AC. Coronary artery bypass surgery and longitudinal evaluation of the autonomic cardiovascular function. Crit Care. 2005;9(2):R124–31. 37. Sugawara J, Murakami H, Maeda S, Kuno S, Matsuda M. Change in post-exercise vagal reactivation with exercise training and detraining in young men. Eur J Appl Physiol. 2001;85(3–4):259–63. 38. 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. Eur Heart J. 1996;17(3):354–81.
39. Thayer JF, Lane RD. The role of vagal function in the risk for cardiovascular disease and mortality. Biol Psychol. 2007;74(2): 224–42. 40. Tulppo MP, Kiviniemi AM, Hautala AJ, et al. Sympatho-vagal interaction in the recovery phase of exercise. Clin Physiol Funct Imaging. 2011;31(4):272–81. 41. Vesterinen V, Hakkinen K, Hynynen E, Mikkola J, Hokka L, Nummela A. Heart rate variability in prediction of individual adaptation to endurance training in recreational endurance runners. Scand J Med Sci Sports. 2013;23(2):171–80.
BASIC SCIENCES
AEROBIC TRAINING AND VAGAL CONTROL
Medicine & Science in Sports & Exercised
Copyright © 2015 by the American College of Sports Medicine. Unauthorized reproduction of this article is prohibited.
1167
ABSTRACT DUARTE, A., P. P. SOARES, L. PESCATELLO, and P. FARINATTI. Aerobic Training Improves Vagal Reactivation Regardless of Resting Vagal Control. Med. Sci. Sports Exerc., Vol. 47, No. 6, pp. 1159–1167, 2015. Purpose: Resting cardiac vagal modulation (RCVM) and postexercise vagal reactivation (PEVR) are markers of parasympathetic activity. We investigated whether adaptations in these markers to aerobic training are influenced by baseline autonomic control. Methods: Forty healthy men (19.2 T 0.8 yr) of similar ˙ O2peak = 50.4 T 5.7 mLIkgj1Iminj1) completed the study, being matched for autonomic activity and cardiorespiratory fitness (V randomized into four groups: training low-RCVM (TL; n = 11, high-frequency power component [HF] = 48.1 T 8.2 normalized units [n.u.]) and high-RCVM (TH; n = 11, HF = 63.1 T 5.9 n.u.) and nonexercise control low-RCVM (CL; n = 9, HF = 47.1 T 7.5 n.u) and high-RCVM (CH; n = 9, HF = 65.5 T 8.3 n.u.). Aerobic training groups exercised 3 dIwkj1 for 40 min at 75%–85% HR reserve for 12 wk. Before and after the training period, sequences of 5-min R–R intervals were recorded at rest and immediately after maximal treadmill test to estimate (a) RCVM (HF) and (b) PEVR (root mean square of successive R–R differences—rMSSD3–5min = mean ˙ O2peak in TL (11.7% T 5.4%, P G 0.01) and TH (7.0% T 2.9%, P G 0.01), value from 3 to 5 min of recovery). Results: Training improved V with no difference between groups (P = 0.70), but not in CL (0.8% T 3.9%) and CH (1.8% T 6.2%, P = 0.90). Only TL increased RCVM (56.6 T 13.3 n.u., P = 0.03), approaching TH level (58.9 T 12.3 n.u.; P = 0.60); rMSSD3-5min increased in both training groups (P G 0.01) but not in controls (P = 0.99). Relative changes in RCVM ($HF%) and PEVR ($rMSSD3–5min%) were significantly correlated in TL (r = 0.61, P = 0.04). Conclusions: PEVR after exercise increased in both exercise training groups, whereas RCVM increased only in the group with low vagal activity at baseline. Vagal reactivation may be improved by aerobic training, even when basal activity remains unaltered. Key Words: EXERCISE, AEROBIC POWER, AUTONOMIC NERVOUS SYSTEM, HR VARIABILITY, PARASYMPATHETIC MODULATION
Address for correspondence: Paulo Farinatti, Laborato´rio de Atividade Fı´sica e Promo0a˜o da Sau´de, Universidade do Estado do Rio de Janeiro, Rua Sa˜o Francisco Xavier 524, sala 8121-8133 F, Maracana˜, Rio de Janeiro, RJ, CEP: 20.550-013, Brazil; E-mail: [email protected]; [email protected]. Submitted for publication April 2014. Accepted for publication September 2014. 0195-9131/15/4706-1159/0 MEDICINE & SCIENCE IN SPORTS & EXERCISEÒ Copyright Ó 2014 by the American College of Sports Medicine DOI: 10.1249/MSS.0000000000000532
1159
Copyright © 2015 by the American College of Sports Medicine. Unauthorized reproduction of this article is prohibited.
BASIC SCIENCES
A
is mixed regarding the effectiveness of aerobic training and cardiorespiratory fitness on resting and postexercise markers of parasympathetic autonomic control of the heart. Some investigations have reported higher (1,3,8) and others no difference (5,18) in levels of resting parasympathetic modulation of HR in trained compared to untrained subjects. Furthermore, some authors proposed that resting cardiac vagal control improves with aerobic training (7,20,23,41), whereas others concluded that training would not influence autonomic modulation at rest (14,26,29). With regard to the postexercise autonomic control of the heart, some investigators have found a faster HR recovery after exercise, which reflects vagal reactivation (37), in aerobically trained compared to untrained individuals (8,33). The few available cross-sectional and longitudinal studies about the relationship between resting cardiac parasympathetic control and postexercise vagal reactivation have shown conflicting results (5,8,10,24,33,37). Reasons for these contradictory findings could reside in methodological issues such as sample characteristics (e.g., individuals_ baseline level of cardiac autonomic modulation and aerobic fitness) or the use of different exercise recovery protocols. Given that resting cardiac vagal modulation and postexercise HR recovery
utonomic nervous system dysfunction, indicated by relative or absolute decrease in vagal activity or an increase in sympathetic activity, has been linked to cardiovascular disease (34,39) and to an increased risk of potentially lethal cardiac arrhythmias in healthy individuals (21,27). Moreover, an inadequate HR recovery after physical exertion is indicative of reduced cardiac vagal reactivation, which is also a predictor of mortality (12). Aerobic exercise is considered to be a nonpharmacological intervention that may increase cardiac vagal control and promote cardioprotection (4,6). Nonetheless, the literature
BASIC SCIENCES
depend on parasympathetic activity, it would seem probable that aerobic training would induce favorable adaptations in these autonomic markers (40). Nevertheless, there is limited evidence about the effects of training on postexercise autonomic control of HR. Moreover, it is not well known if there is an association between possible adaptations of resting and postexercise vagal indices to aerobic training. In brief, adaptations in resting cardiac vagal modulation and postexercise vagal reactivation are expected with aerobic training. However, whether these effects are influenced by subjects_ initial level of autonomic control and the extent which those markers are mutually related have not been investigated. Because autonomic adaptations to exercise can be assessed at rest or within the immediate recovery from an exercise bout, it is important to determine whether these different outcomes are similarly affected by chronic exercise. If this is not the case, to evaluate the vagal activity in only one condition could lead to misinterpretation of the training effects on cardiac vagal modulation. Therefore, the present study investigated the effects of 12-wk aerobic training on resting cardiac vagal control and postexercise vagal reactivation as modulated by cardiorespi˙ O2peak) among healthy young men. We also ratory fitness (V examined whether the observed training-induced adaptations in resting cardiac vagal modulation influence postexercise vagal reactivation and whether this relationship is affected by baseline levels of resting vagal modulation. We hypothesized that (a) the increase in postexercise vagal reactivation would be positively related to an increase in resting parasympathetic activity achieved through exercise training and (b) the association between the gains in resting and postexercise vagal control with training would be dependent on the baseline pretraining level of resting vagal modulation.
METHODS Ethical approval. All subjects provided written informed consent. Procedures were approved by institutional ethics committee and registered at the Brazilian National Council of Ethics in Research (CAAE/CONEP: 0038.0.312.000-07). Subjects. Sixty-two healthy, young men (18–22 yr) volunteered to take part in the study. Candidates were screened for health problems and interviewed about their habitual physical
activity. Subject exclusion criteria were as follows: (a) chronic medical conditions and diseases, (b) current smoker, (c) use of any cardiovascular medication, (d) BMI 930 kgImj2, and (e) participation in regular exercise training (Q2 dIwkj1) within the last year. After the initial evaluation, 49 young healthy men qualified for the study. Despite the fact that they were not currently training, all of them had previous history of recreational physical activity. Subjects were randomized into training (n = 25; 19.4 T 1.4 yr, 172.6 T 8.3 cm, 66.2 T 9.2 kg, ˙ O2peak = 50.8 T 5.1 mLIkgj1Iminj1; mean T SD) and V control (n = 24; 19.0 T 0.6 yr, 175.3 T 6.7 cm, 67.4 T 10.7 kg, ˙ O2peak = 49.9 T 6.1 mLIkgj1Iminj1) groups with similar V characteristics (P 9 0.30). This selection was performed in pairs from numbers assigned to each individual from the database using the program Research Randomizer (Version 4.0; http://www.randomizer.org). Because one of the purposes of the present study was to verify the effects of baseline parasympathetic control differences in the individuals_ training response, after the training period, subjects were assigned into four different groups for comparative analyses. Selection criterion was the median value of subjects_ initial level of resting vagal modulation, computed as the HF band of HRV spectral analysis expressed in normalized units (n.u.). Therefore, participants in the training condition were assigned into two groups to further analysis, one with lower (TL; n = 13, HF = 48.1 T 8.2 n.u.) and other with higher (TH; n = 12, HF = 63.1 T 5.9 n.u.) baseline HF power. Likewise, to allow comparisons with the trained subjects, participants in the control group were also assigned to groups with lower (CL; n = 12, HF = 47.1 T 7.5 n.u.) and higher HF power (CH; n = 12, HF = 65.5 T 8.3 n.u.). Two subjects in TL and one in TH dropped out of the training program and were not considered for the analyses. In the control group, three subjects from both CL and CH were excluded because of engagement in training programs (n = 4) or medical problems (n = 2). Final analyses included 11 subjects in each training group and 9 subjects in each control group. Considering all variables analyzed and the final sample size, the lowest test power to detect statistical differences at P e 0.05 was 0.83. Experimental protocol. Figure 1 summarizes the study design. All participants underwent a pretraining assessment that included measurement of anthropometric characteristics
FIGURE 1—Experimental protocol design. Laboratory assessments were conducted on weeks 0 and 13. GXT, Graded exercise test; HRRes, HR reserve; HRV, HR variability.
1160
Official Journal of the American College of Sports Medicine
http://www.acsm-msse.org
Copyright © 2015 by the American College of Sports Medicine. Unauthorized reproduction of this article is prohibited.
AEROBIC TRAINING AND VAGAL CONTROL
Considering that HRV describes the HR dynamics through measures in time and frequency domains (38) and that vagal-related HRV indices mostly reflect the magnitude of modulation in parasympathetic outflow, as opposed to overall parasympathetic tone (19), in the present study, markers in the time domain included resting HR (an indicator of autonomic nervous system tonus) and an index describing the HRV (the square root of the mean squared difference of successive R–R intervals—rMSSD), reflecting vagal dynamics in short periods (38). In the frequency domain, a fast Fourier transform (Welch method) with a Hanning window and 50% overlap was used to estimate the power density spectrum of R–R interval variability using a customized routine (MATLAB R12; MathWorks, Inc., Natick, MA) (36). Spectral power was obtained by the integration of power spectrum density function in the very-low-frequency (0.0033–0.04 Hz), low-frequency (LF; 0.04–0.15 Hz), and high-frequency (HF; 0.15–0.40 Hz) components. The HF power was used as an index of vagal activity, and the LF power was adopted as a primarily marker of sympathetic nervous system activity (28,38). The spectral values were expressed as normalized units (n.u.), calculated by dividing the power of each component by the total variance from which the very-low-frequency component had been subtracted and multiplied by 100. This procedure is acknowledged to increase the precision of data about the relative contributions of LF and HF bands to total variability and autonomic balance (38). Assessment and analysis of postexercise vagal reactivation. Within 5 s of GXT termination, participants were placed in the supine position. An HR monitor (Polar RS800; PolarElectro, Kempele, Finland) and transmitter belt (Wearlink Coded; Polar Electro) were used to record the beat-by-beat HR during the 5-min recovery period (32). The R–R interval series were transferred to a personal computer using the Polar Precision Performance SW 4.3 software (Polar Electro). To assess vagal reactivation in the first 5 min after the end of GXT, the time domain HRV index rMSSD was calculated sequentially (MATLAB R12; MathWorks, Inc.) at each 30 s of the recovery period (rMSSD30s) (16,31). To ensure the stability of data, in addition to the 30-s segments analysis of postexercise vagal reactivation, rMSSD was also calculated as the mean value within minutes 3–5 of the recovery period (rMSSD3–5min) (7). In addition, HR recovery was calculated by taking the absolute difference between the final HR at exercise completion and the HR recorded after 60 s of recovery (HRR60s). To avoid interference in the natural return of HR to baseline, participants maintained spontaneous breathing during recovery. Nevertheless, their respiratory rate remained in the HF range (90.15 Hz) during the recovery period and did not differ significantly both before and after training (P 9 0.05). Aerobic training program. The aerobic training protocol consisted of three 40-min running sessions per week for 12 wk, at an intensity corresponding to 75%–85% of HR reserve (HRRes). To monitor exercise intensity, TL and TH
Medicine & Science in Sports & Exercised
Copyright © 2015 by the American College of Sports Medicine. Unauthorized reproduction of this article is prohibited.
1161
BASIC SCIENCES
and resting autonomic control by means of HR variability (HRV) (38). Subjects were instructed to avoid drinking coffee or other caffeinated beverages for 12 h and to refrain from exercising for 48 h before the pretraining assessments. The HRV was assessed with subjects seated in a silent and thermoneutral room (23-C) for 30 min, after which a continuous ECG was obtained for 5 min with subjects in the supine position. Familiarization with running on a treadmill was done on two occasions before the graded exercise test (GXT). The GXT was used to determine peak oxygen con˙ O2peak) and postexercise vagal reactivation. sumption (V After completion of pretraining assessments, subjects assigned in the training group underwent 12 wk of aerobic training. Controls were advised to perform their normal activities of daily living and not to engage in exercise training, considered as two or more exercise sessions per week. Control subjects kept a physical activity diary to record all exercises performed, which was verified by the same researcher on a weekly basis. At the end of the training period, all pretraining assessments were repeated at the same time of the day (08:30–11:30 a.m.). Determination of peak O2 consumption. Subjects performed a cardiopulmonary treadmill GXT using a ramp protocol, which started with a 5-min warm-up at 8.0 kmIhj1 followed by a gradual work rate increase of 0.8 kmIhj1 every 1 min and grade held constant at 1%. Minute ventilation and gas exchange were measured breath-by-breath and registered as the mean value of each 5 s (Ultima CardiO2; Medical Graphics Corporation, St. Paul, MN) at an ambient temperature of 23-C. The GXT was terminated if at least three out of the four following criteria were attained: (a) maximum voluntary exhaustion defined by attaining a 10 on the Borg CR-10 scale, (b) 95% of the predicted HRmax (220 j age) or presence of an HR plateau ($HR between two consecutive work rates e4 beats per minute), (c) presence of a ˙ O2 plateau ($V ˙ O2 between two consecutive work rates V j1 of G2.1 mLIkg Iminj1), and (d) maximal RER (RERmax) 91.10. The highest value of O2 consumption measured during ˙ O2peak. the test was used as V Assessment and analysis of resting autonomic control. Continuous ECG signals were acquired during 5 min of rest using Ag/AgCl electrodes placed on the participant_s chest in a standard three-lead configuration (C5) by means of a digital data acquisition system (Biopac MP-150; Biopac Systems, Goleta, CA) in the supine position. Because of the influence of respiratory rate on HRV indices, the subject_s breathing frequency (thoracic and abdominal movements) was recorded by two respiratory effort transducers (TSD201; Biopac Systems) during spontaneous breathing. The respiratory rate of all subjects was in the high-frequency range (90.15 Hz). All signals were acquired at a sampling frequency of 1000 Hz using the AcqKnowledge 4 software program (Biopac Systems). Before analysis, all R–R intervals were visually inspected for artifacts to make interpolation corrections on the 5-min sequences (38). This was necessary in G1% of the sequences in each subject.
TABLE 1. Pretraining subjects_ characteristics (mean T SD) among the aerobic training and control high and low HF power groups. Aerobic Training Age, yr Height, cm Weight, kg Body fat, %
Control
TL (n = 11)
TH (n = 11)
CL (n = 9)
CH (n = 9)
19.2 T 172.6 T 66.2 T 12.1 T
19.5 T 174.3 T 66.4 T 10.8 T
19.1 T 0.3 174.4 T 5.6 68.3 T 12.1 12.0 T 6.4
18.9 T 177.2 T 70.6 T 12.1 T
1.0 8.3 9.2 5.4
1.2 6.8 5.0 4.0
0.6 6.8 12.1 7.4
BASIC SCIENCES
There were no significant differences between groups (P 9 0.05). CH, control group with baseline high levels of HF power; CL, control group with baseline low levels of HF power; HF, HR variability high-frequency spectral power; TH, training group with baseline high levels of HF power; TL, training group with baseline low levels of HF power.
subjects wore a telemetric HR monitor (Polar RS800; Polar Electro) during all training sessions. The duration of exercise and average HRRes were recorded, and the training impulse (TRIMP) in each session was calculated to ensure that the prescribed training intensity and volume were achieved (2). TRIMP integrates in a single variable both intensity and volume of training, being determined by multiplying the duration of the exercise bout (TD) by the average HR Res during the exercise, as follows: TRIMP = TD %HRRes 0.64 e(1.92 %HRres). Weekly training load was computed individually as the average values of TD and %HRRes and as the sum of TRIMP values of the three sessions performed in a given week. Statistical analysis. Normal distribution of data was tested by the Shapiro–Wilk test. Considering that all variables showed normal distribution, differences between groups in subjects_ characteristics at baseline were analyzed using separate one-way ANOVA. The TD, %HRRes, and TRIMP comparisons between TL and TH during the 12-wk training protocol were made by means of two-way ANOVA with repeated measures, with one between-group factor (groups = TL and TH) and one within-group factor (time = 12 wk of training). Within- and between-groups differences before (Pre) and after (Post) training with regard to vagal control indices at rest, cardiorespiratory fitness, HR recovery after 60 s of the end of exercise (HRR60s), and absolute index of postexercise vagal reactivation (rMSSD3–5min) were analyzed using twoway repeated-measures ANOVA with one between-group factor (groups = TL, TH, CL, and CH) and one within-group
factor (time = Pre and Post). The time course of variation in rMSSD30s as a result of training was analyzed using a threeway repeated measures ANOVA with one between-group factor (initial resting vagal condition; TL and TH) and two within-group factors, the assessment period (Pre and Post) and recovery time (10 measurements over a 5-min recovery period). Between-groups differences concerning the relative variation in resting HF spectral power ($% HF, n.u.) and relative change in postexercise vagal reactivation with training ($% rMSSD3–5min) [$%variable = ([posttraining j pretraining] / pretraining) 100] were tested using one-way ANOVA. In all cases, the Tukey HSD post hoc test was used to identify differences when the ANOVA models detected significant F ratios. Pearson correlation tested the correlation between $% HF (n.u.) and $% rMSSD3–5min in TL and TH groups. The magnitude of change after versus before training was estimated by calculating the effect size (ES), considering values G0.2 as having a small effect size, 0.5 as moderate effect size, and Q0.8 as high effect size (11). P e 0.05 was considered significant. All calculations were performed using Statistica 7.1 software (StatSoft, Tulsa, OK), and results are presented as mean T SD.
RESULTS Pretraining subjects_ characteristics. All groups were matched with regard to age, height, weight, and body fat (P 9 0.05; Table 1). As presented in Table 2, the pretraining HF
TABLE 2. Resting HR variability, cardiorespiratory fitness at peak exercise, and HR recovery after 60 s of exercise before (Pre) and after (Post) the aerobic training program. TL (n = 11) Pre HR, bpm rMSSD, ms HF, n.u. LF, n.u. LF/HF V˙O2peak, mLIkgj1Iminj1 RERmax HRmax, bpm HRR60s, bpm
61.8 T 79.1 T 48.1 T 48.2 T 1.08 T 50.6 T 1.26 T 202.4 T 54.2 T
TH (n = 11) Post
9.2 30.6 8.2 8.8 0.47 5.4 0.07 6.0 8.4
53.9 131.8 56.6 40.3 0.79 56.3 1.23 201.1 62.1
T 5.9* T 83.6*** T 13.3* T 12.8 T 0.40 T 5.3*** T 0.11 T 7.2 T 7.9*
Pre 55.1 108.2 63.1 33.2 0.54 51.1 1.22 199.5 55.8
T 6.3** T 42.2 T 5.9** T 5.7** T 0.13** T 5.1 T 0.09 T 8.8 T 6.4
CL (n = 9) Post
53.4 111.1 58.9 37.9 0.71 54.6 1.21 201.0 62.7
T 6.2 T 23.8 T 12.3 T 11.6 T 0.36 T 4.6*** T 0.05 T 7.7 T 6.8*
Pre 58.3 86.5 47.1 49.1 1.11 50.7 1.17 198.4 54.5
T 7.5 T 61.8 T 7.5 T 7.5 T 0.47 T 7.7 T 0.07 T 6.1 T 7.4
CH (n = 9) Post
56.7 101.3 46.4 49.1 1.18 50.5 1.17 196.4 54.8
T 7.6 T 57.7 T 10.5** T 10.3 T 0.40** T 6.2** T 0.05 T 10.2 T 8.3
Pre 56.1 118.0 65.5 31.6 0.50 49.1 1.21 198.7 56.1
T 6.1 T 67.9 T 8.3**** T 7.5**** T 0.17**** T 4.4 T 0.08 T 6.8 T 10.7
Post 55.4 109.4 60.6 36.7 0.69 49.9 1.22 196.6 56.5
T 8.8 T 52.8 T 14.7**** T 13.8**** T 0.40**** T 5.0 T 0.08 T 6.2 T 6.9
Values are means T SD. CH, control group with baseline high levels of HF power; CL, control group with baseline low levels of HF power; HF, high-frequency power; HRmax, maximal HR during graded exercise test; HRR60s, HR recovery after 60 s of the end of exercise; LF, low-frequency power; RERmax, maximal RER; rMSSD, root mean square of successive normal R–R interval differences; TH, training group with baseline high levels of HF power; TL, training group with baseline low levels of HF power; V˙O2peak, peak oxygen consumption. *P G 0.05 versus Pre. **P G 0.05 compared to the corresponding TL value. ***P G 0.01 versus Pre. ****P G 0.05 compared to the corresponding CL value.
1162
Official Journal of the American College of Sports Medicine
http://www.acsm-msse.org
Copyright © 2015 by the American College of Sports Medicine. Unauthorized reproduction of this article is prohibited.
FIGURE 3—Mean values T SD of (A) the relative change in highfrequency power of HRV ($% HFn.u.) and (B) the relative change in the square root of the mean squared difference of successive R–R intervals during the last 2 min of the 5-min postexercise recovery time ($% rMSSD3–5min) for the low (TL) and high (TH) baseline vagal modulation training groups and the low (CL) and high (CH) baseline vagal modulation control groups, as a result of training. Significant differences are disclosed in the figure.
FIGURE 2—Mean values of the square root of the mean squared difference of successive R–R intervals, measured in successive 30-s intervals (rMSSD30s), presented by training groups of low (TL) and high (TH) baseline vagal modulation, before (Pre) and after (Post) the training protocol. *Significant difference in comparison to rMSSD30s at 30 s (P G 0.05). †Significant difference between Pre and Post values within groups (P G 0.05). To improve figure clarity, the error bars were omitted.
Training effects on postexercise vagal reactivation. The training program induced a significant increase (P G 0.05) in HRR60s in TL and TH (Table 2). No changes were observed in CL and CH (P 9 0.95). Figure 2 illustrates the time course of rMSSD30s in TL and TH during the 5-min recovery period after GXT at Pre and Post training. The rMSSD30s increased in TL from Pre to Post at 60, 90, 120, 150, 180, and 270 s of recovery and in TH at 60 s (P G 0.001). There was also a significant effect for the recovery time within each group (5 min; P G 0.001). After training, the rMSSD30s increased significantly at 60 s of recovery in both TL and TH, whereas before training, a significant increase was observed only after 180 s in TL and 120 s in TH. The rMSSD30s was not different between TL and TH either before or after training (P = 0.99). The absolute vagal reactivation index (i.e., rMSSD3–5min) increased from Pre to Post training in TL (Pre 5.5 T 4.4 vs Post 8.4 T 4.3, P G 0.001, ES = 1.24) and TH (Pre 7.7 T 4.0 vs Post 10.1 T 3.4, P G 0.01, ES = 0.74), but no Pre versus Post change was detected in the control groups (CL: Pre 5.0 T 1.9 vs Post 5.2 T 3.0, P = 0.99; CH: Pre 6.6 T 2.9 vs Post 6.9 T 3.6, P = 0.96). No statistical difference was found
AEROBIC TRAINING AND VAGAL CONTROL
Medicine & Science in Sports & Exercised
Copyright © 2015 by the American College of Sports Medicine. Unauthorized reproduction of this article is prohibited.
1163
BASIC SCIENCES
(n.u.) was lower (P G 0.001), whereas LF (n.u.) (P G 0.001) and LF/HF (P G 0.01) were higher in TL and CL compared to groups with higher vagal modulation (TH and CH). The four ˙ O2peak at baseline (P = 0.88). groups had similar V Aerobic training program adherence. All subjects in the training groups accomplished the planned 36 exercise sessions. The computed TD (P = 0.89), %HRRes (P = 0.93), and TRIMP (P = 0.84) were similar across TL and TH along the 12-wk protocol. The mean TD was 40.1 T 0.3 min for TL and 40.1 T 0.2 min for TH, and the mean %HRRes of the training sessions was 77.7% T 0.7 % for TL and 76.6% T 1.1 % for TH. The mean weekly TRIMP (A.U.) was 257.6 T 9.5 for TL and 260.3 T 10.3 for TH, meaning that these groups underwent the same training load during the study. Training effects on resting cardiac vagal modulation and cardiorespiratory fitness. HR decreased (P G 0.05) and rMSSD increased (P G 0.01, ES = 0.85) from Pre to Post in TL (Table 2). After training, TL and TH exhibited similar resting HR (P = 0.82) and rMSSD (P = 0.38). With regard to the spectral components of HRV, HF (n.u.) increased from Pre to Post in TL (P = 0.03, ES = 1.05) and did not change in TH (P = 0.91). Considering LF (n.u.) and LF/HF measurements, differences observed in Pre between TL and TH (P G 0.05) were not detected in Post (P 9 0.56). After training, LF/HF was lower in TL compared to CL (P = 0.04, ES = 0.98). ˙ O2peak increased from Pre to Post (P G 0.001) in TL The V and TH but not in CL and CH (P = 0.95). As a result of ˙ O2peak was similar (P = 0.70) training, the relative increase in V in TL (11.7% T 5.4%, ES = 0.82) and TH (7.0% T 2.9%, ES = 0.75) and higher (P G 0.03) than in CL (0.8% T 3.9%, ES = 0.03) and CH (1.8% T 6.2%, ES = 0.17), respectively.
BASIC SCIENCES
FIGURE 4—Relationships between the relative changes observed with training in the high-frequency component of HR variability (HFn.u.) and in the square root of the mean squared difference of successive R–R intervals during the last 2 min of the 5-min postexercise recovery time (rMSSD3–5min). Correlation coefficients were calculated based on data of low (TL, n = 11; empty squares) and high (TH, n = 11; full circles) baseline vagal modulation training groups.
between groups with regard to rMSSD3–5min in either Pre or Post experiment assessments (P 9 0.83). Changes in resting and postexercise vagal modulation with training. Figure 3A shows the betweengroups comparison for the relative variation in resting HF spectral power ($% HF, n.u.) due to training. The $% HF was higher in TL than in TH (P = 0.01) and CL (P G 0.03). The relative increase in $% rMSSD3–5min (Fig. 3B) was higher in TL and TH than in CL (P G 0.001) and CH (P G 0.03), respectively. Also, the increase in $% rMSSD3–5min was greater in TL than in TH (P G 0.02). Correlations among changes in resting and postexercise vagal modulation. Relationships between the relative variation in resting HF spectral power ($% HF, n.u.) and relative change in postexercise vagal reactivation ($% rMSSD3–5min) with training are represented in Figure 4. There was a significant positive correlation between $% HF (n.u.) and $% rMSSD3–5min in TL (r = 0.63, P = 0.04), accounting for 36% of the variation. On the other hand, $% HF (n.u.) and $% rMSSD3–5min were not associated in TH (r = j0.30, P = 0.37).
DISCUSSION This investigation examined the influence of moderate-tovigorous aerobic training on resting and postexercise vagal modulation among healthy young men with different levels of pretraining cardiac autonomic modulation. The 12-wk training protocol induced similar aerobic gains in both trained groups (TL and TH). On the other hand, resting vagal modulation (i.e., HF n.u.) increased only in subjects with lower pretraining HF power (TL; mean variation of 23%), whereas in those with higher levels of vagal modulation (TH), no significant variation was detected. A different behavior, however, was observed for the postexercise vagal reactivation, which improved in TL and TH after training
1164
Official Journal of the American College of Sports Medicine
regardless of the differences in pretraining resting vagal control. These major findings suggest that postexercise vagal reactivation gains with aerobic training may occur in the absence of improvements in resting cardiac vagal modulation. Recent studies (7,23,41) investigating the effects of training on autonomic modulation demonstrated that, when the training load is adequate, a positive response is generally observed in both cardiorespiratory fitness and cardiac vagal modulation. However, others (14,25) have reported an in˙ O2max with training but not in resting HF power crease in V in individuals who underwent aerobic training. These controversial observations suggest that autonomic adaptations to aerobic training may rely not only on the training intensity and volume but also on subjects_ initial aerobic capacity and autonomic control. ˙ O2max It should be highlighted that even when a gain in V is observed without an improvement in HF power (n.u.), this fact does not necessarily imply a lack of autonomic adaptation with training. The HF component of HRV reflects the magnitude of the fluctuation of cardiac vagal control and not the vagal tone level (19). When cardiac vagal control reaches high levels as a result of training, for example, further increasing in vagal tone may eliminate respiratory modulation of the cardiac rhythm, leading to a saturation of HF power (22). When this is the case, it is very difficult to detect a subsequent rise in cardiac vagal tone by means of HRV indices at rest. Accordingly, our findings indicate that the assessment of parasympathetic control should be carried out at rest and during the immediate recovery from an exercise bout when analyzing the effects of training on vagal modulation. Performing the assessments only in resting conditions may provide a limited interpretation of the data. With respect to the effects of training on postexercise vagal reactivation, both training groups showed an increase in HRR60s, rMSSD30s, and rMSSD3–5min within the 5-min recovery period after GXT. The faster recovery in rMSSD30s and the increase in HRR60s and rMSSD3–5min due to aerobic training reflect an improvement in postexercise cardiac vagal reactivation, reinforcing the potential value of exercise in raising parasympathetic activity (10,35,37). In addition, our findings showed that resting vagal modulation increased only in TL. This suggests that parasympathetic reactivation after a maximal bout of exercise might be improved by aerobic training, even in the absence of resting vagal modulation increase as observed in TH. It is well known that both HRV and vagal reactivation after exercise provide useful information about the autonomic balance. However, the present findings suggest that there may be some differences in the physiological determinants underlying the responses of these markers to training. Dewland et al. (13) addressed this issue using acetylcholinesterase inhibition and suggested that HRV would mostly express fluctuations of vagal outflow to the heart, whereas the recovery of HR after an effort would reflect the mean cholinergic signaling at the sinoatrial nodal junction. Actually, several authors have reported an increase of vagal reactivation with exercise training or as a consequence of
http://www.acsm-msse.org
Copyright © 2015 by the American College of Sports Medicine. Unauthorized reproduction of this article is prohibited.
AEROBIC TRAINING AND VAGAL CONTROL
recovery was assessed by rMSSD30s, which shows the inherent variability in short tracings of 30 s and is considered as a marker of vagal reactivation. Despite differences in the methodological design (cross-sectional longitudinal), the results of both studies emphasize that indices of vagal reactivity after exercise would be more sensible to show autonomic adaptations to training (or associated to higher cardiorespiratory fitness) than resting HRV alone. Limitations of this report warrant consideration. The spectral analysis of HRV was used to assess changes in cardiac autonomic control. Although the contribution to HRV modulation by sympathetic, parasympathetic, and respiratory influences can be identified by changes in specific frequency bands, neither LF (0.04–0.15 Hz) nor HF (0.15–0.40 Hz) power components can be considered as exclusive markers of sympathetic and parasympathetic autonomic controls, respectively. In addition, it is not known whether the autonomic reactivation patterns observed in this study may be generalized to the recovery period from exercise protocols with different characteristics. Finally, a potential ceiling effect on the response of resting vagal modulation to training was observed in TH subjects. The acquisition of ECG signals for longer periods (i.e., 24 h) could avoid this effect, identifying the point of saturation of HF power in relation to R–R interval variations with training (22). Alternatively, the use of sympathetic pharmacological blockade, leaving the HR control predominantly under parasympathetic influence, with subsequent maneuvers of parasympathetic stimulation and withdraw with vasoactive drugs (15), could have helped to evaluate the relationship between HRV indices and parasympathetic control of the heart. This information might have provided additional insight on the effects of resting vagal tone adjustments upon postexercise vagal modulation. However, this analysis was beyond the scope of the present investigation.
CONCLUSIONS In conclusion, a 12-wk moderate-to-vigorous aerobic training program increased resting cardiac vagal control (rMSSD and HF power) and postexercise vagal reactivation (HRR60s, rMSSD30s, and rMSSD3–5min) in healthy young men. In addition, it was observed that resting cardiac vagal modulation and postexercise vagal reactivation may respond differently to aerobic training, depending on the subjects_ pretraining level of resting vagal control. An increase in postexercise vagal reactivation may be observed, even in the absence of resting vagal modulation improvement. These findings have direct implications on the interpretation of the effects of endurance training on the autonomic control. This research was supported in part by the Pro-Defesa Project (Brazil Ministry of Defense and CAPES) and by grants from the Carlos Chagas Filho Foundation for the Research Support in Rio de Janeiro State (FAPERJ, process E-26/150.751/2007) and from
Medicine & Science in Sports & Exercised
Copyright © 2015 by the American College of Sports Medicine. Unauthorized reproduction of this article is prohibited.
1165
BASIC SCIENCES
high level of aerobic fitness (10,24,33,37), but there is limited evidence about the simultaneous variation that should occur with resting vagal modulation in those situations. In the present study, the magnitude of training effects on vagal HRV indices varied as a function of subjects_ pretraining level of resting vagal modulation (see Fig. 3). Whereas TL subjects showed an increase in the relative variation of resting HF (n.u.) and in the relative change of postexercise vagal reactivation (rMSSD3–5min%), TH subjects presented a different trend, with an increase only in the relative change of postexercise vagal reactivation and no variation in resting HFn.u.%. In addition, $% HF (n.u.) and $% rMSSD3–5min were positively correlated in TL subjects (r = 0.61, P = 0.04), whereas no significant correlation was detected in TH subjects. Our hypothesis that the improvement in postexercise vagal reactivation with training may be associated with the increase in resting autonomic modulation can be therefore accepted, but this effect seems to depend on subjects_ initial level of resting autonomic control. Our findings indicate that subjects with high levels of pretraining resting parasympathetic control may improve postexercise vagal reactivation without any increase in resting vagal modulation. In brief, the improvement in postexercise vagal reactivation with training may not necessarily be related to an increase in resting parasympathetic control. As aforementioned, most studies examining the relationship between resting vagal modulation and autonomic recovery of HR after exercise are crosssectional and analyzed this association measuring resting vagal control after the end of exercise (9,17). In such conditions, cardiac vagal modulation may still be influenced by the previous effort (30). In the few available longitudinal studies (7,10), different training strategies and assessment methods have been applied, which compromises comparison between their results. Buchheit et al. (10) trained participants in supramaximal intensity for 9 wk. Postexercise vagal reactivation, as reflected by rMSSD30s, and normalized HF power measured 5–10 min after exercise were not different from pretraining levels. In another study by these investigators (7), the resting parasympathetic control was positively associated with postexercise vagal reactivation only in subjects who ˙ O2max in response to training versus showed an increase in V those whose aerobic capacity did not change. In the present study, vagal reactivation after exercise was evaluated immediately after the end of the effort until 5 min of recovery, and all trained subjects were responsive to the training protocol ˙ O2peak). In this scenario, the (i.e., showed improvement in V relationship between resting vagal modulation and postexercise vagal reactivation was observed only in subjects with baseline low levels of resting vagal control (see Fig. 4). More recently, Lee and Mendoza (24) showed that resting HRV and HR recovery (HRrec) would be dissociated in highly fit men and women. In that cross-sectional study, HRrec was measured as the reduction of HR in the first minute after exercise cessation and spectral analysis of HRV was carried out at rest. In our study, vagal modulation at rest was also measured by the HF power of HRV, and parasympathetic
the National Council of Technological and Scientific Development (CNPq, process 305729/2006-3, process 420122/2005-2, and process 481434/2008-9).
The authors have no conflict of interest to declare. The results of the present study do not constitute endorsement by the American College of Sports Medicine.
BASIC SCIENCES
REFERENCES 1. Aubert AE, Seps B, Beckers F. Heart rate variability in athletes. Sports Med. 2003;33(12):889–919. 2. Banister EW. Modeling elite athletic performance. In: MacDougall JD, Wenger HA, Green, HJ, eds. Physiological Testing of the High-Performance Athlete. Champaign (IL): Human Kinetics; 1991. pp. 403–24. 3. Berkoff DJ, Cairns CB, Sanchez LD, Moorman CT 3rd. Heart rate variability in elite American track-and-field athletes. J Strength Cond Res. 2007;21(1):227–31. 4. Billman GE. Aerobic exercise conditioning: a nonpharmacological antiarrhythmic intervention. J Appl Physiol. 2002;92(2):446–54. 5. Bosquet L, Gamelin FX, Berthoin S. Is aerobic endurance a determinant of cardiac autonomic regulation? Eur J Appl Physiol. 2007;100(3):363–9. 6. Buch AN, Coote JH, Townend JN. Mortality, cardiac vagal control and physical training—what’s the link? Exp Physiol. 2002;87(4):423–35. 7. Buchheit M, Chivot A, Parouty J, et al. Monitoring endurance running performance using cardiac parasympathetic function. Eur J Appl Physiol. 2010;108(6):1153–67. 8. Buchheit M, Gindre C. Cardiac parasympathetic regulation: respective associations with cardiorespiratory fitness and training load. Am J Physiol Heart Circ Physiol. 2006;291(1):H451–8. 9. Buchheit M, Laursen PB, Ahmaidi S. Parasympathetic reactivation after repeated sprint exercise. Am J Physiol Heart Circ Physiol. 2007;293(1):H133–41. 10. Buchheit M, Millet GP, Parisy A, Pourchez S, Laursen PB, Ahmaidi S. Supramaximal training and postexercise parasympathetic reactivation in adolescents. Med Sci Sports Exerc. 2008; 40(2):362–71. 11. Cohen J. Statistical Power Analysis for the Behavioral Sciences. 2nd ed. Hillsdale (NJ): Lawrence Erlbaum Associates; 1988. 12. Cole CR, Blackstone EH, Pashkow FJ, Snader CE, Lauer MS. Heart-rate recovery immediately after exercise as a predictor of mortality. N Engl J Med. 1999;341(18):1351–7. 13. Dewland TA, Androne AS, Lee FA, Lampert RJ, Katz SD. Effect of acetylcholinesterase inhibition with pyridostigmine on cardiac parasympathetic function in sedentary adults and trained athletes. Am J Physiol Heart Circ Physiol. 2007;293(1): H86–92. 14. Gamelin F, Berthoin S, Sayah H, Libersa C, Bosquet L. Effect of training and detraining on heart rate variability in healthy young men. Int J Sports Med. 2007;28(7):564–70. 15. Goldberger JJ, Challapalli S, Tung R, Parker MA, Kadish AH. Relationship of heart rate variability to parasympathetic effect. Circulation. 2001;103:1977–83. 16. Goldberger JJ, Le FK, Lahiri M, Kannankeril PJ, Ng J, Kadish AH. Assessment of parasympathetic reactivation after exercise. Am J Physiol Heart Circ Physiol. 2006;290(6):H2446–52. 17. Goulopoulou S, Heffernan KS, Fernhall BO, Yates G, BaxterJones ADG, Unnithan VB. Heart rate variability during recovery from a Wingate test in adolescent males. Med Sci Sports Exerc. 2006;38(5):875–81. 18. Grant CC, Clark JR, van Rensburg DC, Viljoen M. Relationship between exercise capacity and heart rate variability: supine and in response to an orthostatic stressor. Auton Neurosci. 2009; 151(2):186–8.
1166
Official Journal of the American College of Sports Medicine
19. Hedman AE, Hartikainen JE, Tahvanainen KU, Hakumaki MO. The high frequency component of heart rate variability reflects cardiac parasympathetic modulation rather than parasympathetic Ftone_. Acta Physiol Scand. 1995;155(3):267–73. 20. Heydari M, Boutcher YN, Boutcher SH. High-intensity intermittent exercise and cardiovascular and autonomic function. Clin Auton Res. 2013;23(1):57–65. 21. Jouven X, Empana J-P, Schwartz PJ, Desnos M, Courbon D, Ducimetie`re P. Heart-rate profile during exercise as a predictor of sudden death. N Engl J Med. 2005;352(19):1951–8. 22. Kiviniemi AM, Hautala AJ, Makikallio TH, Seppanen T, Huikuri HV, Tulppo MP. Cardiac vagal outflow after aerobic training by analysis of high-frequency oscillation of the R–R interval. Eur J Appl Physiol. 2006;96(6):686–92. 23. Kiviniemi AM, Tulppo MP, Eskelinen JJ, et al. Cardiac autonomic function and high-intensity interval training in middle-age men. Med Sci Sports Exerc. 2014;46(10):1960–7. 24. Lee CM, Mendoza A. Dissociation of heart rate variability and heart rate recovery in well-trained athletes. Eur J Appl Physiol. 2012;112(7):2757–66. 25. Leicht AS, Allen GD, Hoey AJ. Influence of age and moderateintensity exercise training on heart rate variability in young and mature adults. Can J Appl Physiol. 2003;28(3):446–61. 26. Loimaala A, Huikuri H, Oja P, Pasanen M, Vuori I. Controlled 5-mo aerobic training improves heart rate but not heart rate variability or baroreflex sensitivity. J Appl Physiol. 2000;89(5):1825–9. 27. Lombardi F, Makikallio TH, Myerburg RJ, Huikuri HV. Sudden cardiac death: role of heart rate variability to identify patients at risk. Cardiovasc Res. 2001;50(2):210–7. 28. Malliani A, Julien C, Billman GE, et al. Cardiovascular variability is/is not an index of autonomic control of circulation. J Appl Physiol. 2006;101(2):684–8. 29. Martinmaki K, Hakkinen K, Mikkola J, Rusko H. Effect of lowdose endurance training on heart rate variability at rest and during an incremental maximal exercise test. Eur J Appl Physiol. 2008; 104(3):541–8. 30. Mourot L, Bouhaddi M, Tordi N, Rouillon JD, Regnard J. Shortand long-term effects of a single bout of exercise on heart rate variability: comparison between constant and interval training exercises. Eur J Appl Physiol. 2004;92(4–5):508–17. 31. Ng J, Sundaram S, Kadish AH, Goldberger JJ. Autonomic effects on the spectral analysis of heart rate variability after exercise. Am J Physiol Heart Circ Physiol. 2009;297(4):H1421–8. 32. Nunan D, Donovan GAY, Jakovljevic DG, Hodges LD, Sandercock GRH, Brodie DA. Validity and reliability of short-term heart-rate variability from the Polar S810. Med Sci Sports Exerc. 2009;41(1): 243–50. 33. Otsuki T, Maeda S, Iemitsu M, et al. Postexercise heart rate recovery accelerates in strength-trained athletes. Med Sci Sports Exerc. 2007;39(2):365–70. 34. Schroeder EB, Liao D, Chambless LE, Prineas RJ, Evans GW, Heiss G. Hypertension, blood pressure, and heart rate variability: the Atherosclerosis Risk in Communities (ARIC) study. Hypertension. 2003;42(6):1106–11. 35. Seiler S, Haugen O, Kuffel E. Autonomic recovery after exercise in trained athletes: intensity and duration effects. Med Sci Sports Exerc. 2007;39(8):1366–73.
http://www.acsm-msse.org
Copyright © 2015 by the American College of Sports Medicine. Unauthorized reproduction of this article is prohibited.
36. Soares PP, Moreno AM, Cravo SL, Nobrega AC. Coronary artery bypass surgery and longitudinal evaluation of the autonomic cardiovascular function. Crit Care. 2005;9(2):R124–31. 37. Sugawara J, Murakami H, Maeda S, Kuno S, Matsuda M. Change in post-exercise vagal reactivation with exercise training and detraining in young men. Eur J Appl Physiol. 2001;85(3–4):259–63. 38. 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. Eur Heart J. 1996;17(3):354–81.
39. Thayer JF, Lane RD. The role of vagal function in the risk for cardiovascular disease and mortality. Biol Psychol. 2007;74(2): 224–42. 40. Tulppo MP, Kiviniemi AM, Hautala AJ, et al. Sympatho-vagal interaction in the recovery phase of exercise. Clin Physiol Funct Imaging. 2011;31(4):272–81. 41. Vesterinen V, Hakkinen K, Hynynen E, Mikkola J, Hokka L, Nummela A. Heart rate variability in prediction of individual adaptation to endurance training in recreational endurance runners. Scand J Med Sci Sports. 2013;23(2):171–80.
BASIC SCIENCES
AEROBIC TRAINING AND VAGAL CONTROL
Medicine & Science in Sports & Exercised
Copyright © 2015 by the American College of Sports Medicine. Unauthorized reproduction of this article is prohibited.
1167
