In many sports, the competitive season involves a series of events that can stretch over several weeks or months (e.g., cycling, triathlon, and biathlon). In this particular context, peaking for major competitions each month, even every other week, usually poses the problem of choosing between recovering from previous competition and rebuilding the athlete’s fitness, or maintaining intensive training and capitalizing on adaptations acquired during the previous training cycle and competitive stimuli. Both approaches can be valid, and the choice should depend on the level of fatigue present after a race (or a series of competitions) and the time frame between the last competition and the next one. Whatever the strategy used, optimized taper periods characterized by large training volume reduction (∼−50%) over a prolonged period (∼1 or 2 wk) (5) cannot practically be planned when competitions are so close together. The resulting combination of frequent competitions with short tapers largely increases the risk of persistent fatigue. When the balance between appropriate training stress and adequate recovery is disrupted, an abnormal training response may occur, and a state of short-term “overreaching” (functional OR [F-OR] ) may develop, resulting in a decline in performance. Although the F-OR state is generally reversed when an appropriate period of recovery is provided (∼1–2 wk ), it can compromise the competition outcome in the short term. The currently accepted method for diagnosing F-OR is to monitor performance after completion of a resting period of several days or weeks (19). Nevertheless, this method, which only ensures a diagnosis by retrospection, is frequently rejected by coaches and athletes because it may disrupt the training continuum and could lead to potential detraining before the next competition. It is therefore important to identify early markers of F-OR in endurance athletes who require large training loads to achieve peak performance. However, critical reviews of existing scientific literature continue to conclude that the underlying causes of F-OR in endurance athletes remain uncertain (19,33,36,45).
Several studies have suggested that a down-regulation of the sympathetic nervous system and/or changes in the balance between parasympathetic and sympathetic tone was related to OR symptoms (20,31,44). A decreased HR (4,10,20,26,30,31) and a shift to the right in the lactate versus intensity curve (4,19,30,45) were observed during maximal exercise in F-OR endurance athletes and suggested autonomic nervous system (ANS) and metabolic perturbations. Recently, Le Meur et al. (30) strengthened these results by showing that an OR index, which combines HR and blood lactate concentration changes after a strenuous training period, could discriminate OR subjects from control counterparts with a 90% classification success rate.
During the past decade, some authors have proposed that monitoring changes in HR at rest may represent a practical method to detect early signs of F-OR (16). The beat-to-beat variation in resting HR or HR variability (HRV) has been suggested to assess ANS activity (42). As such, HRV analysis may serve as a practical noninvasive method of assessing cardiac autonomic status and possibly detecting F-OR. Several authors have investigated the effects of heavy training load programs on HRV indices. Nevertheless, most of them did not assess performance decrement in response to the prescribed overload program (16,25,38,39,48), making difficult to address any clear conclusion for truly OR athletes. This discrepancy might be because the terminology on OR and overtraining was still not well defined (34). When focusing on the studies during which HRV response was assessed in endurance athletes who demonstrated clear signs of OR (i.e., high perceived fatigue and short-term decreased performance), no clear change was observed (12,20,46). Hedelin et al. (20) reported unchanged HRV in nine canoeists after increasing the training load by 50% during a 6-d training camp, despite decreases in running time to exhaustion and V˙O2max. Similarly, Dupuy et al. (12) did not report any significant variation in HRV indices in 11 endurance athletes led to F-OR, despite significant decreases in HR values at rest and during exercise. Nevertheless, HRV measurements were systematically performed punctually before and after the overload period during these studies, and the use of isolated HRV recording to assess training-induced ANS activity has been questioned due to the high day-to-day variability of HRV values (40).
The aim of the present study was therefore to analyze whether changes in HRV parameters could differentiate an experimental group of triathletes voluntarily led to F-OR from a control group. After 1 wk of very light training (baseline), a 3-wk overload period including weekly performance tests was designed to lead the experimental group to F-OR. This period was then followed by a 1-wk taper to determine whether athletes’ performance levels would be restored after a week of reduced training load. By obtaining daily HRV samples, we aimed to evaluate if different HRV sampling frequencies (daily or weekly) would differently delineate the potential changes observed with F-OR. At last, we investigated whether the potential changes in ANS activity observed at rest in F-OR triathletes would be accompanied by changes in their cardiovascular response to maximal and submaximal exercises.
MATERIALS AND METHODS
Twenty-four trained triathletes volunteered to participate in this study (11). All subjects had been competing for 2 yr in triathlon and were training a minimum of six times per week. Their performance level over the short (Olympic) distance triathlon (i.e., 1.5 km swimming–40 km cycling–10 km running) ranged between 2 h and 2 h 15 min (mean ± SD performance = 128 ± 5 min). The experimental design of the study was approved by the Ethical Committee of Saint-Germain-en-Laye (acceptance no. 10054) and was performed in accordance with guidelines set forth in the Declaration of Helsinki. Before participation, subjects underwent medical assessment to ensure that they presented normal ECG patterns and could participate in this study. None of them reported chronic diseases or any prescribed medication. After comprehensive verbal and written explanations of the study, all subjects gave their written informed consent to participate.
The subjects were randomly assigned to either the control group (n = 8) or the intensified training group (IT, n = 16) according to a matched group experimental design based on maximal oxygen uptake (V˙O2max) and maximal aerobic speed. Mean ± SD age, V˙O2max, and maximal aerobic speed were 32 ± 8 yr, 62 ± 3 mL O2·min−1·kg−1, and 18.2 ± 1.1 km·h−1 for the control group and 30 ± 5 yr, 62 ± 5 mL O2·min−1·kg−1, and 18.4 ± 0.8 km·h−1 for the intensified training group, respectively.
The protocol is illustrated in Figure 1. The investigation was conducted during the competitive triathlon season to ensure a high fitness level for all participants. The training of each subject was monitored for a period of 7 wk in total, which was divided into four distinct phases. The first two phases were similar for both the IT and the control groups. The first phase (I) consisted of 3 wk, during which the subjects completed their usual amount and type of training during the competitive season (i.e., classic training). The second phase (II) consisted of 1 wk of low-intensity moderate training, during which the subjects were asked to reduce their habitual training volume by ∼50% while maintaining the training intensity, according to the guidelines provided by Bosquet et al. (5) concerning optimal tapering strategies in endurance sports. During the third period (III), the IT group completed a 3-wk intensified training program designed to deliberately overreach the subjects: the classic training load was increased by 40% (e.g., a 1-h run including six repetitions of 400 m at the maximal aerobic running speed was converted into an 85-min run, including 10 repetitions of 400 m at the maximal aerobic running speed). The participants reproduced the same training program during each week of the overload period when considering both the content and the weekly distribution of the training sessions. The control group repeated its classic training program during each week of this period. The fourth period (IV) reproduced the same training regimen than phase II for all the subjects (i.e., 1-wk taper).
Throughout the entire experiment, all subjects were coached by the same sport scientist. The training schedule was controlled to remain similar during each week of phase III. To avoid injuries, particular attention was devoted to daily feedback obtained from triathletes. At the end of phase II (Pre), each week during phase III (W1, W2, and W3) and at the end of phase IV (recovery [R]), the subjects performed a maximal incremental running test on a 340-m indoor running track, always on the same day of the week and at the same time of the day. The performance measure was the total running distance covered during the test. The test began at 11 km·h−1, and the speed was increased by 1 km·h−1 every 3 min until volitional exhaustion (35). A rest period of 1 min was provided between each running step. Between each increment, during a 1-min rest period, blood samples were taken from the participants’ ear lobes and analyzed using a Lactate Pro system (41). The triathletes followed a cyclist traveling at the required velocity to ensure that the subjects were respecting the imposed pace. Visual marks were set at 20-m intervals along the track. The cyclist received audio cues via an mp3 player; the cue rhythm determined the speed needed to cover 20 m. To ensure that variations in performance during these tests were caused by the global training protocol rather than the training session(s) performed the day before each test, the subjects were required to respect a 24-h resting period before each maximal test. Oxygen uptake (V˙O2) and expiratory flow (V˙ E) were recorded breath by breath with a telemetric system collecting gas exchange (Cosmed K4b2, Rome, Italy), which was calibrated before each test. HR values were monitored every second using a cardiofrequencemeter (RS800sd; Polar Electro, Kempele, Finland). Expired gases and HR values were subsequently averaged every 5 s and were analyzed (i.e., mean value) on periods corresponding to the last 30 s of run sections. V˙O2max was determined according to the two following criteria: a respiratory exchange ratio value of 1.15 and an HR value higher than 90% of the predicted maximal value. The intensities and associated HR at which [La−]b increased higher than 2 mmol·L−1 and the lactate threshold calculated by the modified D-max method (LT) (2) were subsequently determined. The RPE was measured verbally using the Borg scale (3) immediately at the end of the maximal running test. The scale and its purpose were carefully explained to each triathlete before each incremental test.
Training volume and intensity during the whole protocol (phases I–IV; Fig. 1) were calculated on the basis of recordings from HR monitors (Polar RS400). For all the subjects, HR was measured (every 5 s) during each training session over the entire protocol. The distribution of training intensity was determined using three intensity zones: zone 1, low intensity, HR lower than the HR associated with the intensity at which [La−]b increased higher than 2 mmol·L−1; zone 2, moderate intensity, HR between the intensity at which [La−]b increased higher than 2 mmol·L−1 and LT; zone 3, high-intensity, HR higher than LT (14). Because the development of an F-OR state is likely to modify the relationship between HR and exercise intensity (4,10,20,26,30,31), HR values delimiting the three intensity zones were adjusted after each maximal incremental tests.
Orthostatic Test and HRV Data Analyses
An orthostatic test was performed each morning during phase II (7 d), phase III (21 d), and phase IV (7 d) (Fig. 1). All the subjects in the training groups were instructed to measure their R-R intervals at home every morning after awakening and emptying their urinary bladder. HRV was measured upon waking via R-R series recorded using the Suunto MemoryBelt HR monitor with a sampling rate of 1000 Hz (MemoryBelt; Suunto Oy®, Vantaa, Finland). Athletes were instructed to leave the HR monitor by their bedside each evening to ensure minimal physical activity when putting on the apparatus. The subjects remained supine for 8 min, after which they were asked to stand up for 7 min. R-R intervals were recorded and analyzed at rest both during the last 4 min of the supine position and during the last 4 min of the standing position. HR data were transferred to a computer using the Suunto Training Manager software and were further analyzed using specialized HRV analysis software (Nevrokard® aHRV, Izola, Slovenia) by the same individual. Data were visually inspected to identify artifacts and occasional ectopic beats, which were removed manually. Suitable normal–normal periods were then selected for analysis to determine the root-mean-square difference of successive normal R-R intervals (RMSSD). Mean HR was also analyzed. The Hanning windowing function was applied, and the Goertz algorithm was used for calculation in the frequency domain. Low-frequency (LF, between 0.05 and 0.15 Hz) and high-frequency (HF, between 0.15 and 0.50 Hz) bands were expressed in absolute units (ms2). The LF/HF ratio was also calculated. The RMSSD and the HF band (ms2) were considered as indices of parasympathetic modulation (37,42). The LF/HF ratio was used to investigate the sympathetic modulation of total HRV, although the interpretation of LF/HF is still a matter of debate (13). We decided to allow our participants to breathe spontaneously during the measurements (29). For all HRV samples, it was subsequently verified that the respiration rate, provided by the Suunto MemoryBelt, always remained in the HF range (>0.15–0.50 Hz) because the system used allowed to record this parameter during the test. When this assumption was not met, the test was not retained for subsequent analysis.
The effect of the training regimen was recorded each week on the morning before the maximal running test; subjects assessed their perceived fatigue during the last 24 h using a visual 0–100 analog scale (from “no fatigue” to “maximum fatigue”).
Assessment of OR
To determine the reproducibility of performance during the maximal running test and to identify OR athletes in the overload group, the typical error (90% confidence interval [CI]) in performance was calculated in the control group (i.e., performance divided by
). To be diagnosed as OR, athletes of the IT group had to reveal a performance decrement higher than the lower limit of the 90% CI reported in the control group (OR threshold).
Upon waking parameters
Two analyses were performed using the HR and HRV parameters recorded both in supine and standing positions. The first analysis was performed using only the values collected once per week, each test day during the very light training week (Pre), each week of the overload program (W1, W2, and W3), and during the 1-wk taper (recovery, R). The second analysis was completed using the weekly average values for the same training periods.
To reduce the effect of interindividual differences in performance level, analyses of week-to-week HR values were achieved for three intensity levels of exercise determined during W3: low intensity running, lactate threshold, and at exhaustion. Differences in the physiological response to exercise were then evaluated for the same absolute running speed throughout the protocol. Changes in HR values were also analyzed when considering the maximal value reached during each test. For all subjects, the low-intensity running was set at 13 km·h−1 because (i) a very low coefficient of variation of running speed was reported until this intensity (3.93% at 12 km·h−1 and 2.24% at 13 km·h−1), and (ii) this running velocity was at least 2 km·h−1 lower than LT for all the subjects.
Data were assessed for practical significance using magnitude-based inferences (24). We used this qualitative approach because traditional statistical approaches often do not indicate the magnitude of an effect, which is typically more relevant to training prescription and practical detection of OR than any statistically significant effect. All data were log transformed before analysis to reduce bias arising from nonuniformity of error. To compare within-trial changes between trials, we used a modified statistical spreadsheet (22). This spreadsheet calculates the between-trial standardized differences or effect sizes (ES, 90% CI) using the pooled standard deviation (9). Threshold values for ES statistics were ≤0.2 (trivial), >0.2 (small), >0.6 (moderate), >1.2 (large), >2.0 (very large), and >4.0 (extremely large) (24). In addition, we calculated probabilities to establish whether the true (unknown) differences were lower, similar, or higher than the smallest worthwhile change or difference. Recently, Al Haddad et al. (1) showed a moderate reliability of HRV indices and argued that the interpretation of changes in these parameters should be systematically performed using a calculated minimal threshold needed to assess a meaningful difference (or the so-called “smallest worthwhile difference/change” ). This threshold was calculated for each parameter using its coefficient of variation (CV, %) during the 5-wk protocol. Half of CV was thought to represent the smallest worthwhile difference or change as proposed by Hopkins et al. (23). Quantitative chances of higher or lower differences were evaluated qualitatively as follows: >1%, almost certainly not; 1%–5%, very unlikely; 5%–25%, unlikely; 25%–75%, possible; 75%–95%, likely; 95%–99%, very likely; and >99%, almost certain. If the chance of higher or lower differences was >5%, the true difference was assessed as unclear. Otherwise, we interpreted that change as the observed chance (24). Data in text and figures are presented as mean ±90% CI.
In both experimental groups, all subjects successfully completed the prescribed training program. Changes in weekly average training volume, the distribution of the relative training time spent in the intensity zones, and the number of training sessions per week in the three disciplines during the four phases of the protocol are presented in Table 1. Fifteen of the 16 IT subjects demonstrated a decrease in performance after the overload period, followed by a systematic supercompensation effect of performance after the taper (Fig. 2). For all of them, the decrease in performance was larger than the upper limit of the 90% CI of the typical error in running performance in the control group (133 ± 65 m), and a high perceived fatigue was reported at the end of phase III (83 ± 6 on the 0–100 fatigue scale). On the basis of this analysis, these subjects were diagnosed as F-OR. The single subject whose running performance did not decrease was excluded from subsequent analyses. In addition, two other F-OR subjects were not included in the analysis due to poor signal quality during the HRV recordings. Thus, the subsequent results are presented for 13 F-OR subjects (F-OR group) and 8 control subjects (control group).
Performance changes during the protocol in the control group and in the OR group are depicted in Figure 2. At the end of the overload period, running performance was almost certainly decreased in the F-OR group compared with its Pre value (within-group change ± 90% CI = −9.0% ± 2.0% of Pre value, chances that the true difference was higher/trivial/lower, 0%/0%/100%; ES ± 90% CI = −0.62 ± 0.15). The difference in change of running performance was almost certainly greater in the F-OR group than that in the control group at the end of the overload period (0%/1%/99%, ES = −0.39 ± 0.19). Changes in performance in the control group was likely trivial during the same period. At the end of the taper, performance was almost certainly improved in the F-OR subjects (+18.6% ± 3.6% of W3 value, 100%/0%/0%, ES = +1.17 ± 0.22), when compared with W3. The change of performance during the same period was unclear in the control group (+2.6% ± 4.8% of W3 value, 63%/30%/7%, ES = −0.13 ± 0.24). When considering the change in performance during the whole protocol (taper vs Pre), the performance increased likely in the control group (80%/17%/3%, +3.9% ± 4.6% of Pre value, ES = +0.20 ± 0.24) and almost certainly in the F-OR group (100%/0%/0%, +7.9% ± 2.4% of Pre value, ES = +0.52 ± 0.15). This change in performance during the 5-wk program was likely greater in F-OR group than that in the control group (79%/18%/3%, +3.9% ± 4.0% of Pre value, ES = 0.25 ± 0.32).
Differences in perceived fatigue at rest between the two experimental groups at baseline were trivial (32 ± 11 for F-OR vs 24 ± 9 for control subjects). However, the OR group perceived a very likely greater fatigue at rest than the control group at the end of the overload period (83 ± 6 vs 38 ± 15, ES = 1.25 ± 0.88). After the taper, the perceived exertion at rest decreased almost certainly in the F-OR group (40 ± 20, ES = −1.28 ± 0.65). Within-group and between-group differences in RPE at exhaustion were trivial throughout the protocol; all the triathletes always ranged their RPE between very difficult and very, very difficult at exercise cessation during the whole experiment (range = 17–20).
The time course of resting HR and HRV variables using either isolated single-day values recorded each seventh-day (A) or weekly average values (B) are detailed in Figure 3.
When analyses were performed using isolated seventh-day values, the decrease in HR was likely greater in the F-OR group than that in the control subjects during phase III (3%/15%/82%, ES = −0.46 ± 0.55). The decrease of HR in the F-OR group was likely during the overload period (5%/7%/88%, ES = −0.30 ± 0.35). In contrast, HR likely increased during the taper in this group (88%/7%/5%, ES = +0.25 ± 0.19). These outcomes were strengthened when considering weekly average values. Between-group difference in the change in HR was almost certain during the overload period (1%/0%/99%, ES = −0.45 ± 0.30), with an almost certain decrease of HR in F-OR subjects at W3 (0%/0%/100%, ES = −0.60 ± 0.23). HR increase during the taper was possible (63%/36%/1%, +0.13 ± 0.16). For all HRV parameters, using isolated values, the differences in changes between groups were unclear. In contrast, weekly average values of HRV parameters demonstrated a very likely greater increase in Ln RMSSD in the F-OR group when compared with the control subjects (96%/4%/0%, ES = 0.40 ± 0.28). Similarly, a likely greater change in Ln HF (77%/22%/1%, ES = 0.29 ± 0.32) and LF/HF (0%/7%/93%, ES = −1.07 ± 0.77) was reported in the OR subjects. For Ln (LF + HF) (50%/48%/2%, ES = 0.18 ± 0.27), the differences in change between groups were possible. When considering the within-group change in the F-OR group, increases between Pre and W3 were almost certain in Ln RMSSD (99%/1%/0%, ES = 0.38 ± 0.23) and very likely in Ln (LF + HF) (92%/4%/4%, 0.18 ± 0.19). During the taper, Ln (LF + HF) likely decreased [4%/4%/92%, ES = −0.15 ± 0.15 for Ln (LF + HF)], whereas the changes in Ln RMSSD were unclear (13%/13%/74%, ES = −0.10 ± 0.19).
Using isolated values, the between-group difference in the change in HR was very likely greater in the F-OR group than that in the control subjects during the overload period (2%/2%/96%, ES = −0.72 ± 0.64). At the end of this phase, the decrease of HR was very likely in the F-OR group (1%/1%/98%, ES = −0.76 ± 0.57). During the taper, F-OR subjects demonstrated a likely increase in HR (96%/2%/2%, ES = 0.59 ± 0.48) with a likely between-group difference in change, when compared with the control group during this period (91%/4%/5%, ES = 0.46 ± 0.50). Between-group difference in change in Ln HF (91%/8%/1%, ES = 0.57 ± 0.47) and Ln (LF + HF) (97%/1%/2%, ES = 0.50 ± 0.43) were likely to very likely greater in the OR subjects than that in the control group during phase III. For Ln RMSSD (75%/9%/16%, ES = 0.30 ± 0.64) and LF/HF (10%/17%/73%, ES = −0.36 ± 0.63), between-group differences in change were unclear. When considering within change in F-OR group during the overload period, HR very likely decreased (1%/1%/98%, ES = −0.76 ± 0.57), whereas Ln RMSSD (95%/4%/1%, ES = +0.34 ± 0.28), Ln HF (100%/0%/0%, ES = 0.49 ± 0.21), and Ln (LF + HF) (99%/1%/0%, ES = 0.48 ± 0.30) increased very likely to almost certainly. During the taper, between-group differences in change were only possible or unclear for all HRV parameters. When analyses were performed using weekly average values, these results were strengthened both for the overload period and the taper. Between-group differences in change during the overload period were very likely to almost certainly greater in the F-OR group than that in the control group for HR (0%/0%/100%, ES = −0.78 ± 0.44), Ln RMSSD (98%/1%/1%, 0.38 ± 0.27), Ln HF (99%/0%/1%, 0.56 ± 0.36), and Ln (LF + HF) (95%/1%/4%, ES = 0.31 ± 0.29). The within-group changes in these parameters were systematically almost certain (HR, ES = −1.01 ± 0.32; Ln RMSSD, ES = 0.62 ± 0.24; Ln HF, ES = 0.91 ± 0.38 and Ln [LF + HF], ES = 0.37 ± 0.25). During the taper, between-group differences in change were likely greater in the F-OR group than that in the control group in Ln RMSSD (5%/12%/83%, ES = −0.21 ± 0.25), Ln HF (4%/12%/84%, ES = −0.29 ± 0.35), and Ln (LF + HF) (4%/6%/90%, ES = −0.24 ± 0.26). These parameters decreased very likely to almost certainly in the F-OR group during this period [Ln RMSSD (0%/1%/99%, ES = −0.38 ± 0.21); Ln HF (1%/1%/98%, ES = −0.68 ± 0.48); Ln (LF + HF) (3%/1%/96%, ES = −0.27 ± 0.24)].
Changes in HR during exercise are presented in Figure 4. Examination of HR values revealed very likely to almost certainly greater between-group differences in change in the F-OR group compared with the control subjects for all the running intensities during both the overload and the taper periods.
In the present study, we studied the morning resting HRV responses and the cardiac response to exercise in a group of trained triathletes who completed an overload training program followed by a 1-wk taper and developed signs of F-OR in comparison with a control group. The most important finding was that the analysis of weekly average values of HRV parameters upon waking indicated a progressive increase in the parasympathetic modulation of HR during the overload period, which was reversed during the subsequent taper in athletes who developed F-OR. The hypothesis of a parasympathetic hyperactivity in the F-OR group was strengthened by lower HR values during exercise for all running intensities. Further, our data suggested that an alternate analysis approach using mean weekly values obtained from daily HRV recordings, rather than isolated HRV assessments, may improve the diagnostic utility of HRV indices in endurance-trained athletes to assess training-induced adaptations of the ANS. The present results suggest that the day-to-day variability of HRV values is too high to allow clear detection of autonomic modulations associated with F-OR using single-day values.
Assessment of F-OR in the overload group
At baseline, subjects of the F-OR group reported low perceived fatigue at rest (24 ± 9 on the 0–100 scale at Pre), confirming that they were not already in an OR state at the beginning of the protocol. In contrast, all of them reported high perceived fatigue at the end of the overload training (83 ± 6), while no clear variations were observed in the control group during the same period. At the end of the overload training period, a 9% decline in performance was observed in the F-OR group with a 100% chance to be lower at W3 than at baseline. After completion of a recovery week, all the subjects of the F-OR group demonstrated an increase in performance (+7.9% of Pre value). This brief reduction in performance followed by a rapid recovery and a performance gain allowed us to confirm that the subjects of the overload group were in an F-OR and not an NF-OR state, which represents the long-term form of OR. This allowed us to conduct further comparisons with the control group to investigate the potential cardiac autonomic modulations associated with F-OR. In addition, it validates the trainers’ and coaches’ use of this kind of training program: if carefully managed, this leads to transient F-OR with the acceptable follow up to improve performance in athletes.
Evidences of autonomic modulations in functionally OR athletes
A progressive decrease in resting HR values was reported in the F-OR group during the overload period, whatever the resting position tested (supine or standing) or the sampling period taken into consideration (isolated day or weekly average values). This finding was reported in 11 of the 13 F-OR athletes in the supine position and in all of them in the standing position, when considering weekly average values. During the same period, mean HR values remained unchanged in the control group. The results suggested that these changes were linked to an increased parasympathetic activity in the F-OR triathletes. Time-varying analysis and spectral analysis of HRV weekly average values confirmed this hypothesis by showing a likely to almost certain increase in Ln RMSSD and Ln (LF + HF) in both supine and standing positions and an increase in Ln HF in the standing position, three parameters that are accepted as reflecting the parasympathetic modulation of HR (42). The potential influence of this parasympathetic hyperactivity to explain the transient decrease of performance in the F-OR subjects at the end of the overload period was strengthened by the reversed response of these parameters during the taper. Although all F-OR subjects demonstrated a performance supercompensation after a week of moderate training load, a possible to almost certain increase in HR was reported in supine and standing positions, respectively. This progressive restoration of HR values recorded at baseline was associated with a concomitant decrease of HRV indices of parasympathetic tone during the taper. It should be noted that HR and all weekly HRV averages did not return to their baseline level during the subsequent taper period despite performance supercompensation. This result may be explained by the residual effect of the parasympathetic hyperactivity induced by the overload period during tapering in the F-OR athletes. Moreover, these results could also reflect the adaptive response of the cardiovascular system to the prescribed training program with higher vagal tone (28) and/or remodeling of the sinoatrial node (6) leading to reduced resting HR.
The signs of a potential increase in parasympathetic hyperactivity to explain F-OR reported by the present investigation were in contrast with past studies who reported a reduced parasympathetic activity or a shift of the cardiovascular autonomic modulation from a parasympathetic toward a sympathetic predominance at rest after an overload training period (25,47). Uusitalo et al. (47) showed higher LF values in the supine position associated with a lower HRV in the standing position in five trained athletes, who demonstrated a decrease in performance after a 6-wk overload period, when compared with a control group. These between-group differences were interpreted as a sign of increased sympathetic modulation and decreased parasympathetic activity in the OR group. Nevertheless, no significant change in any HRV parameters within the OR group was reported during the overload, making these conclusions controversial. Similarly, Iellamo et al. (25) studied the entire Italian junior national team of rowing (n = 7) at increasing training loads up to 75% and 100% of maximum, the latter approximately 20 d before the Rowing World Championship. Their results showed that very intensive endurance training shifted the cardiovascular autonomic modulation from a parasympathetic toward a sympathetic predominance. However, performance was not assessed during the protocol, making it difficult to determine whether these athletes truly experienced F-OR at the end of the prescribed overload period or not. Overall, these observations confirm the need to include measures showing the decline of sport-specific performance and its subsequent restoration in the short term (>3 wk) to study F-OR and to understand its etiology (33).
The hypothesis of a parasympathetic hyperactivity in the F-OR group upon waking was coherent with the moderate to large decrease in HR values during exercise for all running intensities at the end of the overload period. Interestingly, this transient “chronotropic incompetence” observed during the maximal running test was partially reversed after the 1-wk taper. While the decrease of maximal HR value at exhaustion remained almost certain after the recovery week when compared with baseline, the magnitude of this difference decreased from a large effect size at the end of the overload period (ES = −1.20; −2 ± 2 bpm of the Pre value) to a small one at the end of the taper (ES = −0.24; −2 ± 2 bpm of the Pre value). This result suggests that vagal control was not withdrawn completely during exercise in F-OR and may have significantly limited performance capacity at the end of the overload period. We hypothesized that the small reduction in maximal HR after the taper, potentially associated with the persistence of a slight parasympathetic hyperactivity, was no longer sufficient to limit oxygen transport to the actives muscles at exhaustion. Further studies investigating the cardiac output evolution at exercise both during an overload period leading to F-OR and during its subsequent taper are required to confirm this hypothesis. In addition, other mechanisms could contribute to the decreased HR values observed during exercise in F-OR subjects by the present study and others (4,10,20,26,30,31). A decreased chronotropic response to exercise could point to (i) a remodeling of the sinoatrial node (6), (ii) a down-regulation of the sympathetic nervous system, and/or (iii) a down-regulation of sinus node β receptors as a result of chronic sympathetic activation, which can occur during prolonged periods of intense training (20,30,31,44). In favor of a central origin for decreased sympathetic activity, Lehmann et al. (31) reported a decrease in nocturnal urinary norepinephrine and epinephrine excretion after an increase in training volume. Moreover, submaximal and maximal HR values significantly declined during exercise along with the changes in resting catecholamine levels (31). In favor of a peripheral origin, decreases in HR and/or blood lactate concentrations in the absence of changes in the catecholamine response have been reported by previous studies (18,21,44). Prolonged exposure to elevated catecholamine concentrations as a result of intensified training and/or psychological stress would be sufficient to down-regulate the sensitivity of β-adrenergic receptors and/or decrease their number (32,49). This was observed after exhaustive dynamic exercise (7), chronic exposure to hypoxia (15), and during a prolonged long-term period of heavy endurance training (27) or after infusion of adrenergic agonists (43). Further investigations involving the concomitant recording of HR and blood concentration of catecholamines during exercise are required to test these two hypotheses and delineate the relative influence of central and peripheral mechanisms on the development of F-OR.
Using weekly means of HRV values as a potential method to detect autonomic modulations associated with functional OR
The data showed that analyzing weekly means of HRV values provided a more meaningful assessment of any consistent change in cardiac autonomic balance than relying upon values obtained from a single day for analyzing the chronic ANS response associated with a prolonged overload training program. Indeed, no clear change in HRV parameter was reported using daily values, whereas HR decreased likely to very likely. In contrast, the parasympathetic hyperactivity associated with F-OR was confirmed by HRV parameters with weekly mean values in both supine and standing positions (Fig. 3). This finding was coherent with previous studies, which reported decreased HR values at rest and/or during exercise without any significant concomitant modulations in HRV parameters using isolated measurements before and after an overload training program (12,20,46). This has been previously demonstrated in 14 moderately trained runners, where changes in weekly averaged Ln RMSSD had a very large correlation with changes in the 10-km running performance and likely provides superior methodological validity (8). In addition, Plews et al. (40) showed recently that the diagnosis of NF-OR in an elite triathlete was facilitated by using weekly averaged values instead of single isolated assessments. Altogether, these results suggest that the day-to-day variability of HRV values remained too high to allow clear detection of autonomic modulations using single-day values, when considering the chronic autonomic response to a prolonged training program. This methodological concern may explain why previous studies investigating the HRV response in trained (12,47) and elite athletes (20) led to F-OR reported significantly reduced HR values at rest and during exercise without observing any concomitant significant changes in HRV parameters. In all of these experiments, HRV recordings were indeed performed only once before and after the overload period. On the other hand, we cannot exclude that the discrepancy between the isolated seventh-day and the weekly average analyses may have been partially induced by the 24-h rest period respected before each seventh-day test. Although the rest period was initially imposed to limit the potential influence of training session(s) performed the day prior each running test, it may also have partially reversed the chronic effect of the overload program on the autonomic activity recorded the morning of each exercising test (seventh-day analysis). Nevertheless, we assume that this process was unlikely sufficient to totally abolish the vagal hyperactivity associated with the F-OR development given that all HRV weekly average values progressively increased throughout the overload period. This finding strengthens the necessity for daily HRV measurements, when investigating the development of OR (F-OR and NF-OR), especially when measurements are performed at home by the participants.
Measurements performed at home may not be as highly standardized as those performed in laboratory conditions. Our aim was to test the applicability of HRV in training prescription in a real-life scenario that is more convenient for the athletes. In this manner, we also avoided possible confounding psychophysiological effects of laboratory conditions on cardiovascular measurements (17). However, uncontrolled factors such as the quality and quantity of nighttime sleep or other psychological and physiological stressors may contribute to daily HRV dispersion, and their significance in HRV-based training prescription remains unclear. Moreover, daily HR recordings could be difficult to follow during a long-term (pluriannual) follow-up, and this methodology needs to be used only punctually for baseline values and during critical periods when performance is decreased during more than 2 wk.
In conclusion, the present study supports the concept of autonomic modulations toward a parasympathetic predominance in the functionally overreached endurance athlete. A significant decrease in HR was observed in triathletes led to F-OR upon waking in supine and standing positions as well as during exercise over a large spectrum of intensities. Further investigation involving nonfunctional OR and OT athletes is needed to determine whether this HRV response could be differentiated between functional OR, nonfunctional OR, or OT athletes. It also revealed that due to a wide day-to-day variability, isolated, once per week HRV recordings may not undoubtedly detect training-induced autonomic modulations in F-OR athletes.
This study was supported by the grant of the Agence Française de Lutte contre le Dopage (AFLD, Paris), the French Institute of Sport (INSEP, Paris), and the CNRS, Paris Descartes University. It was made possible by technical support from the French Federation of Triathlon. The authors are grateful to Delphine Saint Laurent for managing the project and to Prof. M. Rieu, Prof. S. Bouisset, and Prof. X.B. Bigard for their constructive help and advices during the study.
The authors also thank Dr. Martin Buchheit for his help concerning the use of qualitative statistical analysis and especially acknowledge the athletes for their help and cooperation. They also thank the reviewers for helpful suggestions on revisions made to the manuscript.
The authors report no conflict of interest.
The results of the present study do not constitute endorsement by the American College of Sports Medicine.
1. Al Haddad H, Laursen PB, Chollet D, Ahmaidi S, Buchheit M. Reliability of resting and postexercise heart rate measures. Int J Sports Med
. 2011; 32 (8): 598–605.
2. Bishop D, Jenkins DG, McEniery M, Carey MF. Relationship between plasma lactate parameters and muscle characteristics in female cyclists. Med Sci Sports Exerc
. 2000; 32 (6): 1088–93.
3. Borg G. Perceived exertion as an indicator of somatic stress. Scand J Rehabil Med
. 1970; 2 (2): 92–8.
4. Bosquet L, Leger L, Legros P. Blood lactate response to overtraining
in male endurance athletes. Eur J Appl Physiol
. 2001; 84 (1–2): 107–14.
5. Bosquet L, Montpetit J, Arvisais D, Mujika I. Effects of tapering on performance: a meta-analysis. Med Sci Sports Exerc
. 2007; 39 (8): 1358–65.
6. Boyett MR. ‘And the beat goes on.’ The cardiac conduction system: the wiring system of the heart. Exp Physiol
. 2009; 94 (10): 1035–49.
7. Brodde OE, Daul A, O’Hara N. Beta-adrenoceptor changes in human lymphocytes, induced by dynamic exercise. Naunyn Schmiedebergs Arch Pharmacol
. 1984; 325 (2): 190–2.
8. 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.
9. Cohen J. Statistical Power Analysis for Behavorial Sciences
. Hillsdale (NJ): Lawrence Erlbaum Associates; 1988. p. 567.
10. Coutts AJ, Slattery KM, Wallace LK. Practical tests for monitoring performance, fatigue
and recovery in triathletes. J Sci Med Sport
. 2007; 10 (6): 372–81.
11. De Pauw K, Roelands B, Cheung SS, de Geus B, Rietjens G, Meeusen R. Guidelines to classify subject groups in sport-science research. Int J Sports Physiol Perform
. 2013; 8 (2): 111–22.
12. Dupuy O, Lussier M, Fraser S, Bherer L, Audiffren M, Bosquet L. Effect of overreaching on cognitive performance and related cardiac autonomic control. Scand J Med Sci Sports
. 2012.doi: 10.1111/j.1600-0838.2012.01465.x
13. Eckberg DL. Sympathovagal balance: a critical appraisal. Circulation
. 1997; 96 (9): 3224–32.
14. Esteve-Lanao J, San Juan AF, Earnest CP, Foster C, Lucia A. How do endurance runners actually train? Relationship with competition performance. Med Sci Sports Exerc
. 2005; 37 (3): 496–504.
15. Favret F, Richalet JP. Exercise and hypoxia: the role of the autonomic nervous system
. Respir Physiol Neurobiol
. 2007; 158 (2–3): 280–6.
16. Garet M, Tournaire N, Roche F, et al. Individual Interdependence between nocturnal ANS activity and performance in swimmers. Med Sci Sports Exerc
. 2004; 36 (12): 2112–8.
17. Grassi G, Turri C, Vailati S, Dell’Oro R, Mancia G. Muscle and skin sympathetic nerve traffic during the “white-coat” effect. Circulation
. 1999; 100 (3): 222–5.
18. Halson SL, Bridge MW, Meeusen R, et al. Time course of performance changes and fatigue
markers during intensified training in trained cyclists. J Appl Physiol
. 2002; 93 (3): 947–56.
19. Halson SL, Jeukendrup AE. Does overtraining
exist? An analysis of overreaching and overtraining
research. Sports Med
. 2004; 34 (14): 967–81.
20. Hedelin R, Kentta G, Wiklund U, Bjerle P, Henriksson-Larsen K. Short-term overtraining
: effects on performance, circulatory responses, and heart rate variability. Med Sci Sports Exerc
. 2000; 32 (8): 1480–4.
21. Hedelin R, Wiklund U, Bjerle P, Henriksson-Larsen K. Cardiac autonomic imbalance in an overtrained athlete. Med Sci Sports Exerc
. 2000; 32 (9): 1531–3.
22. Hopkins WG. Spreadsheets for analysis of controlled trials with adjustment for a subject characteristic. Sportscience
. 2006; 10: 46–50.
23. Hopkins WG, Hawley JA, Burke LM. Design and analysis of research on sport performance enhancement. Med Sci Sports Exerc
. 1999; 31 (3): 472–85.
24. Hopkins WG, Marshall SW, Batterham AM, Hanin J. Progressive statistics for studies in sports medicine and exercise science. Med Sci Sports Exerc
. 2009; 41 (1): 3–13.
25. Iellamo F, Legramante JM, Pigozzi F, et al. Conversion from vagal to sympathetic predominance with strenuous training in high-performance world class athletes. Circulation
. 2002; 105 (23): 2719–24.
26. Jeukendrup AE, Hesselink MK, Snyder AC, Kuipers H, Keizer HA. Physiological changes in male competitive cyclists after two weeks of intensified training. Int J Sports Med
. 1992; 13 (7): 534–41.
27. Jost J, Weiss M, Weicker H. Comparison of sympatho-adrenergic regulation at rest and of the adrenoceptor system in swimmers, long-distance runners, weight lifters, wrestlers and untrained men. Eur J Appl Physiol Occup Physiol
. 1989; 58 (6): 596–604.
28. Kiviniemi AM, Hautala AJ, Kinnunen H, Tulppo MP. Endurance training
guided individually by daily heart rate variability measurements. Eur J Appl Physiol
. 2007; 101 (6): 743–51.
29. Larsen PD, Tzeng YC, Sin PY, Galletly DC. Respiratory sinus arrhythmia in conscious humans during spontaneous respiration. Respir Physiol Neurobiol
. 2010; 174 (1–2): 111–8.
30. Le Meur Y, Hausswirth C, Natta F, Couturier A, Bignet F, Vidal PP. A multidisciplinary approach to overreaching detection in endurance trained athletes. J Appl Physiol
. 2013; 114 (3): 411–20.
31. Lehmann M, Dickhuth HH, Gendrisch G, et al. Training-overtraining
. A prospective, experimental study with experienced middle- and long-distance runners. Int J Sports Med
. 1991; 12 (5): 444–52.
32. Lehmann M, Foster C, Dickhuth HH, Gastmann U. Autonomic imbalance hypothesis and overtraining
syndrome. Med Sci Sports Exerc
. 1998; 30 (7): 1140–5.
33. Meeusen R, Duclos M, Foster C, et al. Prevention, diagnosis, and treatment of the overtraining
syndrome: joint consensus statement of the European College of Sport Science and the American College of Sports Medicine. Med Sci Sports Exerc
. 2013; 45 (1): 186–205.
34. Meeusen R, Duclos M, Gleeson M, Rietjens G, Steinacker J, Urhausen A. Prevention, diagnosis and treatment of the overtraining
syndrome. Eur J Sport Sci
. 2006; 6: 1–14.
35. Mercier D, Gignac E, Léger L. Advantages and validity of discontinuous field test to evaluate runners. In: Proceedings of the CSEP Annual Meeting
. Quebec: 1995.
36. Nederhof E, Lemmink KA, Visscher C, Meeusen R, Mulder T. Psychomotor speed: possibly a new marker for overtraining
syndrome. Sports Med
. 2006; 36 (10): 817–28.
37. Pagani M, Lombardi F, Guzzetti S, et al. Power spectral analysis of heart rate and arterial pressure variabilities as a marker of sympatho-vagal interaction in man and conscious dog. Circ Res
. 1986; 59 (2): 178–93.
38. Pichot V, Busso T, Roche F, et al. Autonomic adaptations to intensive and overload training periods: a laboratory study. Med Sci Sports Exerc
. 2002; 34 (10): 1660–6.
39. Pichot V, Roche F, Gaspoz JM, et al. Relation between heart rate variability and training load in middle-distance runners. Med Sci Sports Exerc
. 2000; 32 (10): 1729–36.
40. Plews DJ, Laursen PB, Kilding AE, Buchheit M. Heart rate variability in elite triathletes, is variation in variability the key to effective training? A case comparison. Eur J Appl Physiol
. 2012; 112 (11): 3729–41.
41. Pyne DB, Boston T, Martin DT, Logan A. Evaluation of the Lactate Pro blood lactate analyser. Eur J Appl Physiol
. 2000; 82 (1–2): 112–6.
42. TaskForce. Heart rate variability. Eur Heart J
. 1996; 17: 354–81.
43. Tohmeh JF, Cryer PE. Biphasic adrenergic modulation of beta-adrenergic receptors in man. Agonist-induced early increment and late decrement in beta-adrenergic receptor number. J Clin Invest
. 1980; 65 (4): 836–40.
44. Urhausen A, Gabriel HH, Kindermann W. Impaired pituitary hormonal response to exhaustive exercise in overtrained endurance athletes. Med Sci Sports Exerc
. 1998; 30 (3): 407–14.
45. Urhausen A, Kindermann W. Diagnosis of overtraining
: what tools do we have? Sports Med
. 2002; 32 (2): 95–102.
46. Uusitalo AL, Uusitalo AJ, Rusko HK. Exhaustive endurance training
for 6–9 weeks did not induce changes in intrinsic heart rate and cardiac autonomic modulation in female athletes. Int J Sports Med
. 1998; 19 (8): 532–40.
47. Uusitalo AL, Uusitalo AJ, Rusko HK. Heart rate and blood pressure variability during heavy training and overtraining
in the female athlete. Int J Sports Med
. 2000; 21 (1): 45–53.
48. Winsley RJ, Battersby GL, Cockle HC. Heart rate variability assessment of overreaching in active and sedentary females. Int J Sports Med
. 2005; 26 (9): 768–73.
49. Zavorsky GS. Evidence and possible mechanisms of altered maximum heart rate with endurance training
and tapering. Sports Med
. 2000; 29 (1): 13–26.
Keywords:© 2013 American College of Sports Medicine
HR VARIABILITY; OVERTRAINING; FATIGUE; AUTONOMIC NERVOUS SYSTEM; ENDURANCE TRAINING