Heart Rate Variability during Night Sleep and after Awakening in Overtrained Athletes : Medicine & Science in Sports & Exercise

Journal Logo

BASIC SCIENCES: Original Investigations

Heart Rate Variability during Night Sleep and after Awakening in Overtrained Athletes


Author Information
Medicine & Science in Sports & Exercise 38(2):p 313-317, February 2006. | DOI: 10.1249/01.mss.0000184631.27641.b5
  • Free


In competitive sports, hard training is essential to improve performance. To optimize the improvement, one should be able not just to train hard, but also to recover after training. Usually, recovery after a single bout of exercise occurs in 24 h, but competitive athletes may train two to three times per day for a few days, followed by a relative recovery period of 1-2 d. An overtraining state can develop if there is an imbalance between training, competing, and recovery (29). This imbalance can occur often in high-level performance sports (14). In the early phase of overtraining state development, the term "overreaching" is used to reflect the nature of this kind of short-term overtraining. Recovery after a hard training period or overreaching can take from several days to a week or two, but after a state of overtraining, it can take from several months to a year to recover (14,21,23,29).

Autonomic nervous system (ANS) dysfunction or imbalance has been presented as one explanation for the signs and symptoms of the overtraining state (25), and they have also provided promising tools for overtraining state detection (11,20,26,28). HRV reflects the continuous oscillation of RRI around its mean value. Overall variability can be calculated in standard deviation of RRI (SDRRI), time domain method, or in total power (TP) frequency domain method. Overall variability is largely dependent on parasympathetic modulation and reduced values are considered to reflect a diminished parasympathetic and an increased sympathetic modulation of sinus node (18,25). In short recordings lasting a few minutes it is possible to use frequency domain methods, because the stationary requirements are reached easier than in longer recordings of many hours. The high-frequency power (HFP; 0.15-0.40 Hz) of HRV is known to reflect parasympathetic activity and the low-frequency power (LFP; 0.04-0.15 Hz) represents both parasympathetic and sympathetic activities (18,25). A higher level of physical fitness has been connected to higher values of HRV, especially HFP (2,4,10,19), although some studies have not found the effect of physical training on HRV (1,17,30). A single bout of exercise decreases ANS activity for up to 2 d (8). A successful training period with appropriate recovery leads to increased resources of the ANS (i.e., increased parasympathetic and decreased sympathetic modulation) (4,9,12,21). Overreaching and overtraining can lead to lower autonomic resources for responding to stress. Previous research on overreaching and overtraining have found a decrease in overall HRV, as well as shifts toward both sympathetic (12,20,21,28) and parasympathetic predominance (11,12,22).

Nocturnal stress hormone concentrations, markers of basal autonomic tone, have also been used to analyze the development of the overtraining state. In the final phase of this development, or in a parasympathetic type of overtraining, the basal urinary catecholamine excretion can be reduced by 50-70% (15,16). The findings concerning nocturnal urine catecholamines have been controversial, however, and the differences in stress hormone concentrations between a well-trained and an overtrained state are not clear (15,16,27).

This study tested the hypothesis of autonomic imbalance in overtrained athletes during night sleep and after awakening, with HRV measurements and nocturnal urine stress hormone concentrations. The hypothesis was that the overtrained athletes would exhibit exhausted autonomic resources, indicated by lower HRV and nocturnal urinary stress hormone concentrations when compared with the control athletes.



We examined 12 severely overtrained (OA, 6 men and 6 women) and 12 control (CA, 6 men and 6 women) athletes. Characteristics of the athletes are presented in Table 1. Both the overtrained and control athletes were reached via information sent to Finnish sport magazines, Finnish sport unions, Finnish sport medicine centers, and about 500 Finnish physicians taking care of athletes. Overtraining state was diagnosed by a medical doctor using the following criteria: an athlete (a) had suffered from an unexplained decrement in physical performance and fatigue even after a recovery time of at least 3 wk, which was verified by carefully interviewing the athletes and, in some cases, also their coaches; (b) had been examined to be otherwise healthy with a clinical examination and several laboratory and physiological measurements (29); and (c) had increased their training volume and intensity progressively for up to 6 months before the overtraining symptoms and had continued training with the symptoms without sufficient recovery. The training logbooks of the previous months were checked to confirm the training history of the athletes. The overtraining state was also verified with higher perceived stress (3). The athletes represented different sport events, but most of them took part in endurance sports. Besides endurance athletes, two male athletes and one female athlete in OA were ice hockey players, and one female athlete in CA was a sprinter (runner). All the athletes were fully informed of the procedures and they gave their written consent to participate in this project. The ethical committees of the Central Finland Health Care District in Jyväskylä and Kuopio University Hospital in Kuopio approved this study.

Participant characteristics (mean ± SD).

Data collection.

The measurements took place within 3-6 wk after the overtraining state diagnosis. The athletes were advised to refrain from alcohol, tobacco, caffeine, and bananas during the day before the measurements. Light exercise was allowed the day before the measurements. The athletes arrived at the laboratory between 3 p.m. and 4p.m., and a collection of RRI with an RR recorder (Polar Electro Ltd, Kempele, Finland) at an accuracy of 1 ms (24) was started. Nocturnal RRI and urine samples (∼7 p.m.-7a.m.) were collected at home. A couple of minutes after awakening (∼7 a.m.), the athletes remained still for 5 min to collect HRV during supine rest and marked the corresponding period in RRI collection by pressing a marker button on the RR recorder.

At approximately 8 a.m., the urine samples were brought to the laboratory and the RR recorder was downloaded to a computer. After the subject's arrival to the laboratory, the urine volume was determined and a sample for cortisol analysis was stored at −80°C. Another urine sample for adrenaline and noradrenaline analysis was acidified with 6n-hydrochlorid acid to a pH of 3, before storing it at −80°C. The athletes performed an incremental test to voluntary exhaustion on a treadmill or cycle ergometer after a short, low-intensity warm-up between 10 a.m. and 11 a.m. Expiratory gases were measured during the exercise test at 20-s intervals using a Sensormedics 2900 z gas analyzer (Sensormedics, Yorba Linda, CA). The analyzer was calibrated with known volumes and gas concentrations before every test and verified immediately after the subject's voluntary exhaustion. If any change in calibration was found, the measured values were corrected, assuming the change to be linear by time. The highest continuous 1-min period was selected as the V̇O2max value. The type of exercise (Nordic walking, running, or cycling) during the exercise test was selected to be as close as possible to the athlete's sport event to render the athlete's true maximal exercise level. The selection of the exercise type made it also easier to give training advice to the athletes after the tests. All these tests ended at RER of ≥1.06 and the athletes stated that they could not have continued the test any more.


All the RRI recordings were edited by visual inspection to exclude erroneous beats, such as movement artifacts. HRV during night sleep was analyzed as a continuous 4-h period between 12 a.m. and 4 a.m. HRV after awakening was analyzed as a continuous 4-min period, between 0:30 and 4:30 of the 5-min supine rest period. The HRV parameters were calculated with time and frequency domain (autoregressive) methods provided in the RR recorder's software (Polar Electro, Kempele, Finland). The following HRV parameters were analyzed: standard deviation of RRI (SDRRI), square root of the mean of the sum of the squares of differences between adjacent RRI (RMSSD), total power (TP), low-frequency power (LFP), high-frequency power (HFP), and low- to high-frequency ratio (LF:HF). In addition, to minimize the heart rate dependency of HRV and to reduce individual differences between athletes in SDRRI, the coefficients of variation (CVRRI = standard deviation of RRI/average RRI × 100) were calculated.

Nocturnal urinary stress hormone samples (cortisol, adrenaline, and noradrenaline) were thawed and analyzed later by high-performance liquid chromatography (HPLC; Yhtyneet Laboratoriot, Helsinki, Finland). Because of different lengths in collection times (from 10 to 12 h), the actual concentration of hormones in the urine sample was first multiplied by the volume of the whole urine collection, then divided by the collection time in hours, and finally multiplied by 12 to represent a 12-h collection for everyone.

Ln-transformation was used with variables that were not normally distributed, and Student t-test for nonpaired samples was used for statistical testing. To compare the responses to awakening, 2 × 2 ANOVA for repeated measurements, with t-test as the post hoc test were also used. The results are presented as mean ± SD. The level of significance was set at P < 0.05.


Cardiac autonomic modulation and urinary stress hormones during night

No differences were found in average nocturnal RRI, HRV indices, or urine stress hormone concentrations between the groups (Table 2).

HRV and urinary stress hormones during night rest in OA and CA (mean ± SD).

Cardiac autonomic modulation after awakening.

Both SDRRI and LFP after awakening were lower in overtrained than in control athletes (Table 3). In addition, when responses to awakening (night vs after awakening recordings) were calculated, CVRRI decreased more in OA than in CA (Fig. 1).

Change in the coefficient of variation of RR interval (CVRRI) from night sleep to after awakening. ***, ** indicate significant differences between night sleep and after awakening (P < 0.001, P < 0.01, respectively). # indicates a significant difference between the groups (P < 0.05).
HRV after awakening in OA and CA (mean ± SD).


This study compared the autonomic modulation during night and after awakening between overtrained and control athletes. The main finding was that the cardiac autonomic modulation was disturbed in OA after awakening but not during night sleep.

The number of overtrained athletes is higher than has usually been reported in overtraining studies, which increases the possibilities of this study to evaluate the effects of overtraining on athletes. The overtrained athletes were diagnosed to be in the overtraining state with tight criteria. Athletes suffering from the symptoms of overtraining first contacted a physician and, after careful exclusion of other possible explanations to their symptoms, they were diagnosed to be in the overtraining state. Especially the history of the OA gives good support to the diagnosis. The higher perceived stress in OA than in CA also supports the diagnosis. In a cross-sectional study design, however, it was not possible to follow the development of the overtraining state because the athletes had developed the overtraining state by themselves, via training that had been too hard and/or via poor recovery. OA also had lower V̇O2max than CA. Overtraining has been shown to decrease V̇O2max (29), but in the present study the difference may also depend on weight differences (no difference in absolute V̇O2max) or a different training status of the groups because a higher percentage of endurance athletes were in CA than in OA. In addition, the symptoms of overtraining had forced the overtrained athletes to reduce the level of their training, and at least some of them may have already been in the beginning of the recovery phase.

No differences were seen in any HR or HRV parameters during night sleep between the two groups in the present study. It has been suggested that night recordings could enhance the reliability of the measurements by ensuring some independence from environmental factors (15,21). Previous research has also presented a connection between training load and HR and HRV during nocturnal sleep (21). The athletes in the Pichot et al. (21) study were not diagnosed to be overtrained, but they were monitored during a 3-wk intensive training period, followed by a relative rest period of 1 wk. The results of Pichot et al. (21) demonstrated decreases in parasympathetic and increases in sympathetic indicators of HRV during the intensive training period. After a relative rest period of 1 wk, they found increases in parasympathetic and decreases in sympathetic HRV indicators. The findings of Pichot et al. (21) may demonstrate the acute and reversible effects of hard training or overreaching on HRV. In the present study, where the OA were in the severe overtraining state or may have started the recovery phase, acute responses such as those in the study of Pichot et al. (21), may have been weaker because of the exhaustion of resources, as proposed in the hypothesis of this study.

Autonomic modulation after awakening was disturbed in OA, as seen in lower SDRRI and LFP in OA than in CA. Interestingly, a markedly increased risk is seen of heart attack, cardiac death, and stroke in the hours around and just after awakening (7)-the same time that disturbed autonomic modulation was found in overtrained athletes. Awakening itself enhances the sympathetic tone (6), and burnout patients have shown elevated cortisol levels during the first hour after awakening (5). If burnout patients overreact to awakening by elevated cortisol levels, a similar overreaction to awakening may also be seen in the enhanced decrease in HRV of OA. Decreasing the period from longer (4 h during night) to shorter recording (4 min after awakening) usually decreases SDRRI (25). This phenomenon can be seen in Tables 2 and 3. Similar response can also be seen in CVRRI (Fig. 1), where CA may present the "normal reaction" and OA the "overreaction." Previous controlled overtraining studies concerning HRV are scarce and the results are controversial (20,28). In these studies, HRV has been measured during supine rest (awake) in the laboratory (20,28). In the present study, also the HRV after awakening was collected at home. Regardless of the methodological differences, the results of Mourot et al. (20) were almost similar to the present study in the overall HRV (SDRRI). In their study, the trained endurance athletes had approximately 50% higher values in SDRRI during supine rest (awake) than the overtrained athletes. No significant differences were noted in LFP, but the overtrained athletes had lower HFP than the trained athletes did. This led to a higher LFP:HFP ratio in overtrained athletes, which may indicate relatively higher sympathetic activation in overtrained than in trained athletes (20). Uusitalo et al. (28) found a significant increase in the LFP of RRI during supine rest induced by heavy endurance training, but found no differences between the whole training group and the subgroup of overtrained athletes.

At home in normal life, the response to awakening provided the enhanced sensitivity of HRV measurements to differentiate OA from CA (Table 3 and Fig. 1). The effect sizes between the groups were approximately 1 SD in both SDRRI and LFP after awakening (Table 3). This encourages further investigation of the phenomenon. No good HRV parameters indicate purely sympathetic activation and, therefore, the autonomic balance is estimated from the power of parasympathetic indicators. Findings of lower SDRRI may suggest that the autonomic balance has changed to sympathetic predominance by parasympathetic withdrawal. LFP has been reported to reflect both sympathetic and parasympathetic activity (25). The findings of lower LFP with almost similar but not significant difference on average HFP may support slight parasympathetic withdrawal. Previous research has found lower autonomic responses to the head-up tilt in overtrained than in trained athletes (20,28). With respect to these results, the response to some kind of challenge (awakening in this study) seems to provide better sensitivity than simple resting measurements.

No differences in nocturnal urinary stress hormones were found between the groups in the present study. This indicates that basal sympathetic tone was similar in these groups, which was against our hypothesis. Previous literature has found decreased nocturnal urine catecholamine concentrations after high-volume training in overtrained endurance athletes (16), but the results have been controversial (27). Some methodological differences exist in nocturnal urinary stress hormone collections and study setups (cross-sectional vs follow-up) between the different studies, which may explain the different findings. Lehmann et al. (16) have reported that athletes successfully training and competing had low nocturnal urinary stress hormone concentrations, but the lowest values (a decrease by at least 50%) were found among the group of overtrained athletes. Again, if recovery after the overtraining state had already started in the OA of the present study, it may have been first visible during night sleep.

In conclusion, cardiac autonomic modulation after awakening was disturbed in overtrained athletes, whereas the autonomic modulation during night sleep did not differ between the overtrained and control athletes. The results suggest that parasympathetic cardiac modulation after awakening was slightly diminished in the overtraining state. Further investigations should concentrate on ANS reactivity to different challenges, because the reactivity may provide a more sensitive index of overtraining, as seen in the differences during nocturnal sleep and supine resting after awakening.

This work was supported by grants from The National Technology Agency of Finland (TEKES), Polar Electro Ltd, Alpha Wave Ltd, and Ministry of Education, Finland. The authors would like to thank MSc. Tiina Hoffman for checking the language.


1. Boutcher, S. H., and P. Stein. Association between heart rate variability and training response in sedentary middle-aged men. Eur. J. Appl. Physiol. 70:75-80, 1995.
2. Carter, J. B., E. W. Banister, and A. P. Blaber. The effect of age and gender on heart rate variability after endurance training. Med. Sci. Sports Exerc. 35:1333-1340, 2003.
3. Cohen, S., T. Kamarck, and R. Mermelstein. A global measure of perceived stress. J. Health Soc. Behav. 24:385-396, 1983.
4. De Meersman, R. E. Heart rate variability and aerobic fitness. Am. Heart J. 125:726-731, 1993.
5. De Vente, W., M. Olff, J. G. C. Van Amsterdam, J. H. Kamphuis, and P. M. G. Emmelkamp. Physiological differences between burnout patients and healthy controls: blood pressure, heart rate, and cortisol responses. Occup. Environ. Med. 60(Suppl I): i54-i61, 2003.
6. Dodt, C., U. Brechling, I. Derad, H. L. Fehm, and J. Born. Plasma epinephrine and norepinephrine concentrations of healthy human associated with nighttime sleep and morning arousal. Hypertension 30:71-76, 1997.
7. Elliot, W. J. Cyclic and circadian variations in cardiovascular events. Am. J. Hypertens. 14:291S-295S, 2001.
8. Furlan, R., S. Piazza, S. Dell'Orto, et al. Early and late effects of exercise and athletic training on neural mechanisms controlling heart rate. Cardiovasc. Res. 27:482-488, 1994.
9. Garet, M., N. Tournaire, F. Roche, et al. Individual interdependence between nocturnal ANS activity and performance in swimmers. Med. Sci. Sports Exerc. 36:2112-2118, 2004.
10. Goldsmith, R. L., J. T. Bigger, D. M. Bloomfield, and R. C. Steinman. Physical fitness as a determinant of vagal modulation. Med. Sci. Sports Exerc. 29:812-817, 1997.
11. Hedelin, R., U. Wiklund, P. Bjerle, and K. Henriksson-Larsen. Cardiac autonomic imbalance in an overtrained athlete. Med. Sci. Sports Exerc. 32:1531-1533, 2000.
12. Iellamo, F., J. M. Legramante, F. Pigozzi, et al. Conversion from vagal to sympathetic predominance with strenuous training in high-performance world class athletes. Circulation 105:2719-2724, 2002.
13. Israel, S. The problems of overtraining in a medical and sports physiological point of view (Zur Problematik des Übertrainings aus internisticher und leistungsphysiologischer Sicht). Medizin und Sport. 16(1):1-12, 1976.
    14. Kelmann, M. Underrecovery and overtraining: different concepts-similar impact? In: Enhancing Recovery, M. Kelmann, (Ed.). Champaign, IL: Human Kinetics, 2002, pp. 3-24.
    15. Lehmann, M., H-H. Dickhuth, G. Gendrich, et al. Training-overtraining: a prospective, experimental study with experienced middle- and long-distance runners. Int. J. Sports Med. 12:444-452, 1991.
    16. Lehmann, M., W. Schnee, R. Scheu, W. Stockhausen, and N. Bachl. Decreased nocturnal catecholamine excretion: parameter for an overtraining syndrome in athletes? Int. J. Sports Med. 13:236-242, 1992.
    17. Loimaala, A., H. Huikuri, P. Oja, M. Pasanen, and I. Vuori. Controlled 5-mo aerobic training improves heart rate but not heart rate variability or baroreflex sensitivity. J. Appl. Physiol. 89:1825-1829, 2000.
    18. Lombardi, F. Clinical implications of present physiological understanding of HRV components. Card. Electrophysiol. Rev. 6:245-249, 2002.
    19. Melanson, E. L. Resting heart rate variability in men varying in habitual physical activity. Med. Sci. Sports Exerc. 32:1894-1901, 2000.
    20. Mourot, L., M. Bouhaddi, S. Perrey, et al. Decrease in heart rate variability with overtraining: assessment by the Poincaré plot analysis. Clin. Physiol. Funct. Imaging 24:10-18, 2004.
    21. Pichot, V., F. Roche, F. E. Gaspoz. Relation between heart rate variability and training load in middle-distance runners. Med. Sci. Sports Exerc. 32:1729-1736, 2000.
    22. Portier, H., F. Louisy, D. Laude, M. Berthelot, and C.-Y. Guezennec. Intense endurance training on heart rate and blood pressure variability in runners. Med. Sci. Sports Exerc. 33:1120-1125, 2001.
    23. Raglin, J., and A. Barzdukas. Overtraining in athletes: the challenge of prevention. A consensus statement. ACSM's Health and Fitness Journal 3:27-31, 1999.
    24. Ruha, A., S. Sallinen, and S. Nissila¨. A real-time microprocessor QRS detector system with a 1 ms timing accuracy for the measurement of ambulatory HRV. IEEE Trans. Biomed. Eng. 44:159-167, 1997.
    25. Task Force of the european society of cardiology and the north american society of pacing and electrophysiology. Heart rate variability: Standards of measurement, physiological interpretation, and clinical use. Circulation 93:1043-1065, 1996.
    26. Uusitalo, A. Ability of non-invasive and invasive methods of autonomic function measurements and stress hormones to indicate endurance training-induced stress. Doctoral Dissertation, University of Tampere, Department of Clinical Physiology, Tampere, Finland. October 1998.
    27. Uusitalo, A. L. T., P. Huttunen, Y. Hanin, A. Uusitalo, and H. Rusko. Hormonal responses to endurance training and overtraining in female athletes. Clin. J. Sport. Med. 8:178-186, 1998.
    28. Uusitalo, A. L. T., A. Uusitalo, and H. Rusko. Heart rate and blood pressure variability during heavy endurance training and overtraining in female athletes. Int. J. Sports Med. 21:45-53, 2000.
    29. Uusitalo. A. L. T. Overtraining-a big challenge for top athletics. Phys. Sportsmed. 29:35-50, 2001.
    30. Uusitalo, A. L. T., T. Laitinen, S. B. Va¨isa¨nen, E. La¨nsimies, and R. Rauramaa. Physical training and heart rate and blood pressure variability: a 5-yr randomized trial. Am. J. Physiol. Heart. Circ. Physiol. 286:H1821-H1826, 2004.


    ©2006The American College of Sports Medicine