The last 4 weeks of basic military training are very stressful. We tested the hypothesis that changes in cardiac autonomic function during this period are associated with changes in maximal oxygen uptake and/or serum hormonal concentrations in male conscripts (n22). Cardiac vagal autonomic function was assessed by measuring the high-frequency (0.150.4 Hz) spectral power of RR intervals. Maximal oxygen uptake (V ̇O2max) and basal serum testosterone concentration were measured at the beginning and end of the period. Individual changes in vagally mediated high-frequency power (range 15% to 25%) correlated (r0.73, P0.001) with changes in V ̇ O2max (range 9% to 6%) and changes in testosterone concentration (range 52% to 43%; r0.43, P0.047). The mean values of V ̇ O2max and testosterone concentration did not change during the study period. Measurements of autonomic function could be a useful tool for indicating adaptation to the highly stressful conditions associated with basic military training.
High physical fitness is necessary for military opera- tions (Kyro ̈la ̈inen et al., 2008; Nindl et al., 2006; Nindl et al., 2007). Thus, basic military training includes moderate-intensity aerobic exercises on a daily basis, with the aim of improving aerobic fitness. A common means of studying individual training responses to military training is to measure maximal oxygen uptake (V ̇ O2max) (Dyrstad, Aandstad, & Halle ́n, 2005; Dyrstad, Soltvedt, & Halle ́n, 2006; Santtila et al., 2006; Santtila, Ha ̈kkinen, Karavirta, & Kyro ̈la ̈inen, 2008; Sonna et al., 2001). However, individual responses to the same physiological stress can vary considerably, and the reasons for this are not well understood (Bouchard & Rankinen, 2001; Hautala, Kiviniemi, & Tulppo, 2009).
Previous studies have shown that changes in V ̇O2max after a training intervention are associated with autonomic nervous system regulation in healthy individuals (Al-Ani, Munir, White, Townend, & Coote, 1996; Hautala et al., 2003; Hedelin, Bjerle,
Henriksson-Larsen, 2001; Pagani et al., 1986; Tulppo et al., 2003). Cardiac autonomic regulation can be measured by means of heart rate variability, a tech- nique that offers a potential and practical tool for evaluating physiological adaptation during basic military training. Cardiac vagal activity can be ass- essed by calculating power spectra density for the high-frequency (HF: 0.150.4 Hz) band of RR intervals (Camm et al., 1996; Hayano et al., 1991). It is proposed that the low-frequency (LF: 0.04 0.15 Hz) spectral power of RR intervals is associated with both sympathetic and vagal outflow, and the LF/ HF ratio has been used as an index of sympathovagal interaction (Malliani, Pagani, Lombardi, & Cerutti, 1991; Montano et al., 1994; Pagani et al., 1986).
Measurement of circulating hormones has also been used to monitor individual training responses during military training (Friedl et al., 2000; Nindl et al., 2006, 2007; Opstad, 1992). The high stress involved in military operations has been found to decrease serum testosterone and increase cortisol concentration (Nindl et al., 2006; Opstad, 1992).
It has been suggested that changes in the testosterone- to-cortisol ratio are associated with the anabolic catabolic balance of the body (Ha ̈ kkinen, 1989; Ha ̈kkinen, Pakarinen, Alen, & Komi, 1985; Kraemer et al., 1991). Huovinen et al. (2009) showed that there is an association between cardiac autonomic regulation and the serum testosterone-to-cortisol ratio. It has been reported that testosterone produc- tion is partly influenced by pro-inflammatory cyto- kines (Jones, 2007), which can be regulated by a cholinergic anti-inflammatory pathway through the vagus nerve (Tracey, 2002). However, the associa- tion between testosterone and vagal activity is unclear.
Evaluation of individual training response by measuring V ̇ O2max or circulating hormones is practi- cally impossible for a large group of soldiers. There- fore, new methods are required that are easy to use and permit objective evaluation of individual adapta- tion during military training. We tested the hypoth- esis that changes in autonomic function, measured with the heart rate variability technique, are asso- ciated with simultaneous changes in V ̇ O2max during military training. We also hypothesized that altered autonomic function during military training might be associated with changes in anabolic and/or catabolic hormone concentration. Finally, we hypothesized that the association between heart rate variability, V ̇O2max, and testosterone could be detected more easily when two different positions (supine and standing) were assessed.
A total of 38 Finnish military conscripts performed pre- and post-V ̇ O2max tests. However, due to the high percentage (90%) of acceptable RR intervals required for each recording, data from only 22 of the 38 conscripts were used for further analysis. The participants (age 19.090.3 years, height 1.789 0.08 m, body mass 80.0920.3 kg; mean9s) pro- vided written informed consent after receiving an explanation of the risks and benefits of the study. The study was approved by the ethics committees of the University of Jyva ̈skyla ̈ and the Kainuu region, as well as the Headquarters of the Finnish Defence Forces.
The participants served as conscripts in their own platoons during basic military training. In the Finnish Defence Forces, the first 4 weeks are not as physically demanding as the rest of basic military training. During the 8 weeks of basic training, the intensity of training is increased gradually, and the
most stressful time is during weeks 58 (Santtila et al., 2008). During that time, most activities are performed outside, and the conscripts learn indivi- dual practical skills that will be required on the battlefield. The V ̇ O2max and hormone assessments were performed at the beginning and end of this highly stressful period. Weekly averages of daily heart rate variability were calculated. Training load was measured by calculating the weekly training impulse (TRIMP) from daily heart rate measurements.
Maximal oxygen uptake was assessed at the start of weeks 5 and 8 of basic military training. The parti- cipants performed a graded V ̇O2max test on a tread- mill. The test started at 4.6 km × h1, followed by increases in speed and/or inclination every 3 min to induce an increase of 6 ml × kg1 × min1 in the theoretical V ̇ O2max demand of running, until volun- tary exhaustion (American College of Sports Medicine, 2001). Pulmonary ventilation and respira- tory gas exchange (Jaeger Oxygen Pro, VIASYS Healthcare GmbH, Hoechberg, Germany), together with heart rate (Polar s810i, Polar Electro, Kempele, Finland) were monitored continuously. The criteria used to determine V ̇O2max were: (1) a lack of in- crease in V ̇O2max and heart rate despite an increase in the grade and/or speed of the treadmill; (2) a respiratory exchange ratio (RER) higher than 1.1; and (3) a post-exercise blood lactate value (deter- mined 1 min after exercise from a fingertip blood sample using a lactate analyser; LactatePro†, Arkray, Japan) that was higher than 8 mmol × l1 (American College of Sports Medicine, 2001). Every partici- pant met at least two of these criteria.
Basal blood samples were collected at the end of weeks 4 and 7 starting at 06.00 h. One day before collection of blood samples, the participants under- went only light physical stress and ate standardized meals. They went to sleep at 22.00 h and were ordered to be quiet and stay in their beds thereafter. On the morning of blood sampling, the participants lay on their beds until a blood samples had been taken from an antecubital vein (Venosafe, Terumo Medical Co., Leuven, Belgium). The blood samples were centrifuged at 1500 g for 10 min. The serum samples were then stored at 808C for further analysis. Serum total testosterone and cortisol con- centrations were assessed with the chemilumines- cence method using an IMMULITE† 1000 analyser (DPC Diagnostics Corporation, Los Angeles, CA, USA). The sensitivites (and intra-assay coefficients
of variance) for these assays were 0.5 nmol × (5.8%) and 5.5 nmol × l1 (4.8%), respectively.
Measurement and analysis of heart rate variability
solid percentages of heart rate reserve, we used in- dividually calibrated heart rate zones based on the V ̇O2max test in the laboratory. The zones corre- sponded to 30100% V ̇ O2max at 10% intervals. The individual heart rate zones were interpolated for each day based on the V ̇O2max tests in weeks 4 and 7. We calculated the number of minutes that each participant trained in each of the heart rate zones, scaled the number according to successful measure- ment time, and used the average TRIMP factor of each zone to accumulate the total weekly TRIMP value. Because of holiday weekends during basic training, some of the training weeks included more training days than others (range 46 days). How- ever, in a comparison between the first and second periods, there were 21 and 20 military service days, respectively.
The data were analysed with SPSS software (SPSS 13.0) and the results are expressed as means9 standard deviations. Gaussian data distribution was verified with the Shapiro-Wilk test. As the spectral values of heart rate variability (HRV) were not normally distributed, the values were transformed by taking the natural logarithms of their absolute values. Weekly averages were calculated from the daily HRV variables. The average weekly TRIMP for the first and second periods was calculated by dividing the total TRIMP by the number of weeks. A repeated-measures analysis of variance (ANOVA) was conducted with Bonferroni correction to exam- ine changes in the heart rate and HRV indices between weeks. A dependent t-test was used to analyse changes in serum hormone concentrations, V ̇ O2max, and the TRIMP. Pearson’s correlation ana- lysis was performed between hormone concentra- tions, HRV parameters, V ̇O2max, and their changes, and the TRIMPs. Furthermore, the participants were divided into tertiles according to changes in high-frequency power from weeks 58 (range of 15.5% to 3.1%, of 1.5% to 2.1%, and of 2.3% to 25.5% for group 1, 2, and 3, respectively). Differences in V ̇O2max and hormones between the subgroups were analysed by one-way ANOVA fol- lowed by Bonferroni post-hoc tests. Statistical sig- nificance was set at PB0.05.
Association between heart rate dynamics and V ̇O2max
Changes in heart rate and heart rate variability, measured at rest, correlated significantly with changes in V ̇ O2max during the training period (Figures 1A, 1C, 1E). However, the changes in
The participants were woken for measurement of heart rate variability 15 min before the other con- scripts so that the measurements could be performed in a quiet environment. They were housed in a dormitory in the garrison with 10 participants in each room. The room temperature was maintained at 228C by an air-conditioning system. Autonomic regulation of the heart was assessed by measuring RR intervals using Polar s810i heart rate monitors (Polar Electro, Kempele, Finland) with an accuracy of 1 ms, allowing natural breathing. The measure- ments began each morning at 05.45 h while the participants lay in their beds. The group leaders used a CD player to communicate the test instructions 1 min before the test. After 5-min recordings of RR intervals during supine rest, the participants stood up and the measurements were repeated in a con- trolled standing position.
After measurement, the RR intervals were trans- ferred to a computer for further analysis (Heart Signal Co., Oulu, Finland). The RR intervals were edited based on visual inspection, and artifacts were deleted. The data were accepted only if the number of accepted beats was 90% (Salo, Huikuri, & Seppa ̈nen, 2001). An autoregressive model (model order 20) was used to assess power spectrum den- sities of the low-frequency (0.04BLFB0.15 Hz) and high-frequency (HF: 0.150.4 Hz) spectral components of RR interval variability (Camm et al., 1996). The LF/HF ratio was also calculated to determine sympathovagal interaction (Malliani et al., 1991; Pagani et al., 1986). In the military en- vironment, short-term measurements can be easily introduced to daily schedules and, therefore, we assessed heart rate variability by means of 5-min re- cordings every morning (Camm et al., 1996). Since heart rate variability has been shown to be suscep- tible to saturation when measured at a low heart rate (Goldberger, Challapalli, Tung, Parker, & Kadish, 2001; Kiviniemi et al., 2004), we measured heart rate variability both during orthostatic stress in the standing position and at rest.
Daily heart rate measurements and training load
Heart rate was recorded every day (from 06.00 to 21.30 h) of basic military training using Polar e600 heart rate monitors (Polar Electro, Kempele, Finland). The weekly training impulse (TRIMP) was calculated as A × B × C (Banister, 1991), where A is exercise time (min), B is heart rate (as a proportion of heart rate reserve), and C is 0.64e1.92B. Instead of heart rate and heart rate variability measured while standing did not correlate with changes in V ̇O2max (Figures 1B, 1D, 1F). Changes in the LF/HF ratio did not correlate with V ̇O2max in either position. Mean values of heart rate and heart rate variability as well as V ̇ O2max did not change during the training period (Table I).
Association between heart rate dynamics and hormones
The changes in testosterone concentration corre- lated significantly with changes in high-frequency and low-frequency power measured at rest (Figures 2A, 2C, 2E). In contrast, changes in testosterone concentration while standing correlated with changes in heart rate but not with changes in high-frequency or low-frequency power (Figures 2B, 2D, 2F).
Neither changes in cortisol concentration nor in the testosterone-to-cortisol ratio correlated with changes in heart rate or heart rate variability. Changes in the LF/HF ratio and V ̇O2max did not correlate with changes in serum hormone concen- tration in either position. The mean values of hormones did not change during the training period (Table I).
Daily heart rate measurements were successfully (7099% of the original data) collected from 14 of 22 participants. The weekly TRIMPs showed that the second period of basic military training was more stressful than the first (11459582 vs. 14569496, average weekly TRIMP for the first and second periods, respectively; P 0.016). The weekly TRIMPs
did not correlate with changes in V ̇O2max or with HRV variables.
Subgroup analysis based on changes in high-frequency power
Groups 1 and 3 (change in high-frequency power: 6.196.3% vs. 8.898.3%, respectively, n7 in both groups) differed from each other in terms of changes in V ̇ O2max (3.493.2% vs. 4.194.5%, respectively; P0.040) and testosterone concentra- tion (24.1918.8% vs. 24.9912.7%, respectively; P0.029) (Figures 3A, 3B). Furthermore, changes in the testosterone-to-cortisol ratio showed the same trend, although the differences between the groups were not statistically significant (Group 1 vs. Group 3: 14.2930.9% vs. 26.9916.4%; P0.055) (Figure 3C). The average weekly TRIMPs for these subgroups were 16649237, 12549227, and 15909 866 (3, 7, and 4 participants in Groups 1, 2, and 3, respectively). Groups 1 and 3 did not differ from each other at baseline in any variables, including heart rate, high- and low-frequency power, LF/HF ratio, age, body mass index, percent body fat, testo- sterone and cortisol concentrations, and V ̇ O2max.
There are two novel findings in this study based on daily measurements of heart rate variability per- formed in a military environment. First, altered cardiac autonomic function, documented by indivi- dual changes in heart rate and vagally mediated high- frequency power of RR intervals, was strongly associated with individual changes in aerobic fitness during the stressful military training period. Second, there was an association between changes in cardiac autonomic function and changes in anabolic hor- mone concentrations.
Changes in V ̇O2max and heart rate dynamics
It is well known that heart rate variability in- creases after aerobic exercise (Al-Ani et al., 1996; Goldsmith, Bigger, Steinman, & Fleiss, 1992; Hautala et al., 2003; Hedelin et al., 2001; Kiviniemi, Hautala, Kinnunen, & Tulppo, 2007; Pichot et al., 2002; Tulppo, Ma ̈kikallio, Seppa ̈nen, Laukkanen, & Huikuri, 1998; Tulppo et al., 2003). However, our military training resulted in wide negative and positive changes in aerobic fitness, hormone concen- trations, and autonomic function due to training in a military environment.
At the beginning of basic military training, the average aerobic fitness of the conscripts (aged 20 years) in a 12-min running test corresponded to a V ̇O2max of 43.2 ml × kg1 × min1 (Santtila et al., 2006). In the present study, the participants were an accurate representation of Finnish conscripts in terms of aerobic fitness, as their mean V ̇O2max was 4397 ml × kg1 × min1. Although there were no changes in mean values, individual changes in V ̇ O2max showed large variation, even during the relatively short-term military training period. Indivi- dual decrements in V ̇ O2max, despite continuous (military) exercise training, could be related to overreaching or overtraining in those participants (Halson & Jeukendrup 2004). The weekly TRIMP during the second period of basic military training was 14569496, which indicates a rather stressful training load. Kiviniemi et al. (2007) reported a weekly TRIMP of 500 to be suitable for recrea- tional male runners (age 32 years). In addition, Iwasaki and colleagues (Iwasaki, Zhang, Zuckerman, & Levine, 2003) suggested that a monthly TRIMP of 1500 is enough for sedentary males and females (age 29 years). Furthermore, increments in V ̇ O2max could be related to positive adaptation to military training. As a main finding of the present study, these changes in V ̇O2max were strongly associated with changes in autonomic function measured by the HRV technique.
The most obvious correlation was observed be- tween the changes in aerobic fitness and the vagally mediated resting heart rate and high-frequency power of RR intervals. In addition, changes in low- frequency power at rest also correlated with changes in V ̇ O2max. As documented in pharmacological studies (Hayano et al., 1991; Tulppo, Ma ̈kikallio, Takala, Seppa ̈nen, & Huikuri, 1996), a significant part of the oscillation in low-fequency power (80%) is mediated by vagal outflow. Therefore, vagal out- flow may also explain the association between low-frequency power and V ̇ O2max. Second, all corre- lations between fitness and autonomic function vanished when the measurements of autonomic function were performed during orthostatic stress caused by standing intervention. Third, change in the LF/HF ratio, which was used as an index of
sympathetic outflow, did not correlate with changes in fitness in either position. Taken together, our data support the concept that vagal outflow, rather than sympathetic activity, is associated with changes in V ̇ O 2 m a x .
In previous studies, various explanations for the association between heart rate variability and V ̇ O2max have been suggested, including genetic factors, changes in blood pressure/volume, and adaptation at the central nervous system level (Hautala et al., 2003). In addition, the different individual responses to physiological stress are well recognized (Bouchard & Rankinen, 2001). However, hormonal regulation, expressed as the anabolic or catabolic balance of the body, has received much less attention. The association between individual changes in training responses and heart rate dy- namics could be the result of a complex interaction between neural and humoral regulatory mechanisms, as described by Tracey (2002). Indeed, changes in testosterone concentration were associated with changes in heart rate dynamics in the present study.
Heart rate dynamics and hormone concentrations
The normal ranges of serum testosterone and cortisol at rest in healthy young males vary from 14 to28nmol×l1andfrom110to520nmol×l1, respectively (Young, 1993). In response to combined short- (days) and long-term (weeks) mental and physical stress, these hormone concentrations have been shown to change significantly. Opstad (1992)
reported that testosterone concentration decreased (from 25.9 to 6.0 nmol × l1) and cortisol concen- tration increased (from 550 to 698 nmol × l1) significantly during a 5-day military training course. Similar changes were observed in testosterone and cortisol concentrations during an 8-week US Ranger training course (Nindl et al., 2007). In the present study, mean serum testosterone and cortisol con- centrations were within the normal range, and significant changes in their mean values were not observed (range of change from 52 to 43% and from 33% to 45% for testosterone and cortisol, respectively; Table I). However, the serum
testosterone concentration of some individuals de-
creased, while their cortisol concentration increased (from 23.8 to 15.5 nmol × l1 and from 403 to 585 nmol × l1 for testosterone and cortisol, respec- tively). Thus, in some participants these individual changes could be indicators of catabolic metabolism in the body.
In the present study, a correlation between changes in high-frequency power at rest and testos- terone concentration was observed, which suggests that there is an association between cardiac vagal activity and serum concentration of this anabolic hormone. The increase in vagal activity was greatest in conscripts with the greatest increase in V ̇O2max, testosterone concentration, and the testosterone-to- cortisol ratio. In addition, participants exhibiting a decrease in vagal activity also showed decrements in all of these variables. To clarify this, Figure 3 demonstrates that Groups 1 and 3 (determined by changes in high-frequency power at rest) exhibited dissimilar changes in V ̇O2max, testosterone concen- tration, and the testosterone-to-cortisol ratio during the intervention. In addition, Figure 4 demonstrates two individual cases, one a ‘‘positive responder’’ and the other a ‘‘negative responder’’ to the stress of basic military training. Our results suggest that indivi- dual TRIMPs were not associated with measured physiological changes in the positive or negative responders. However, there was a trend of a higher TRIMP in the negative responders (16649237 vs. 15909866 for Group 1 and 3, respectively). In addition, the TRIMP trend in the two individual cases (Figure 4) was in line with that of the sub- groups (16999191 vs. 14859188 for negative and positive responders, respectively). However, because the TRIMP data were successfully collected from only 14 of 22 participants, these results should be interpreted with caution.
Measurement of long-term changes (e.g. over several weeks) in vagally mediated high-frequency power of RR intervals may be a useful tool for fol- lowing individual changes in anabolic or catabolic balance of the body. This finding is in line with pre- vious studies that have shown a positive association between testosterone concentration and parasympa- thetic activity in patients with coronary artery disease (Wranicz et al., 2004). The possible mechanism for this could be related to a cholinergic anti-inflammatory pathway that regulates pro-inflammatory cytokines through the vagus nerve (Tracey, 2002). In addition, an increment in pro-inflammatory cytokines is asso- ciated with testosterone production through a re- duced pituitary luteinizing hormone pulse (Jones, 2007). Thus, changes in high-frequency power could be associated with altered regulation of both neural and hormonal mechanisms.
Limitations of the study
In the present study, we used a single-time-point analysis of blood samples (Opstad, 1992), and circadian rhythms and the pulsatile nature of hor- mone secretions were not studied (Nindl et al., 2006). Salivary free cortisol concentration has been shown to peak 30 min after waking (Clow et al., 2006), whereas the participants in the present study were tested 45 min after waking. Furthermore, although the conscripts were asked to go to bed at 22.00 h, the actual quality of sleep is unknown. Similarly, although the conscripts ate the same food at the same time of day, their actual nutritional intake was not controlled. It is therefore possible that increased psychological loading due to sleep and energy deprivation during basic military training may have affected some participants. To observe possible overreaching in some individuals, a follow-up of recovery should have been undertaken and, there- fore, the present results should be interpreted with caution. In the present study, high-frequency power was measured at rest and during orthostatic stress in a standing position. In the military environment, the supine position was easier to standardize than the standing position, and thus the results for orthostatic stress may have been stronger if, for example, a tilt bench had been used. Finally, it should be empha- sized that a quasi-experimental pre- to post-test design (Cook & Campbell, 1979) was used in the
present study performed in an isolated military environment.
The present study provides novel information about the association between changes in cardiac auto- nomic function, aerobic fitness, and anabolic hor- mone concentrations during basic military training in Finnish conscripts. These observations may have important practical implications for following indi- vidual physiological levels of stress based on mea- surement of heart rate variability.
The authors gratefully acknowledge the Finnish Defence Forces, the Ministry of Education, the Scientific Advisory Board for Defence, and Polar Electro for financially supporting the study. We also thank all of the subjects who volunteered for the study.
Al-Ani, M., Munir, S. M., White, M., Townend, J., & Coote, J. H. (1996). Changes in RR variability before and after endurance training measured by power spectral analysis and by the effect of isometric muscle contraction. European Journal of Applied Physiology, 74, 397403.
American College of Sports Medicine (2001). Guidelines for exercise testing and prescription (6th edn.). Baltimore, MD: Lippincott Williams & Wilkins.
Banister, E. W. (1991). Modeling elite athletic performance. In H. J. Green, J. D. McDougal, & H. Wenger (Eds.), Physio- logical testing of elite athletes (pp. 403424). Champaign, IL: Human Kinetics.
Bouchard, C., & Rankinen, T. (2001). Individual differences in response to regular physical activity. Medicine and Science in Sports and Exercise, 33, 446451.
Camm, A., Malik, M., Bigger, J. T., Breithardt, G., Cerutti, S., Cohen, R. J., et al. (1996). Heart rate variability: Standards of measurement, physiological interpretation, and clinical use. Circulation, 93, 10431065.
Clow, A., Edwards, S., Owen, G., Evans, G., Evans, P., Hucklebridge, F., et al. (2006). Post-awakening cortisol secre- tion during basic military training. International Journal of Psychophysiology, 60, 8894.
Cook, T. D., & Campbell, D. T. (1979). Quasi-experimentation: Design and analysis issues for field settings. Chicago, IL: Rand McNally College Publishing.
Dyrstad, S. M., Aandstad, A., & Halle ́n, J. (2005). Aerobic fitness in young Norwegian men: A comparison between 1980 and 2002. Journal of Medicine and Science in Sports, 15, 298303.
Dyrstad, S. M., Soltvedt, R., & Halle ́n, J. (2006). Physical fitness and physical training during Norwegian military service. Military Medicine, 171, 736741.
Friedl, K. E., Moore, R. J., Hoyt, R. W., Marchitelli, L. J., Martinez-Lopez, L. E., & Askew, E. W. (2000). Endocrine markers of semi starvation in healthy men in multistressor environment. Journal of Applied Physiology, 88, 18201830.
Goldberger, J. J., Challapalli, S., Tung, R., Parker, M. A., & Kadish, A. H. (2001). Relationship of heart rate variability to parasympathetic effect. Circulation, 103, 19771983.
Goldsmith, R. L., Bigger, J. T., Steinman, R. C., & Fleiss, J. L. (1992). Comparison of 24-hour parasympathetic activity in endurance trained and untrained young men. Journal of the American College of Cardiology, 20, 552558.
Ha ̈kkinen, K. (1989). Neuromuscular and hormonal adaptations during strength and power training: A review. Journal of Sports Medicine and Physical Fitness, 29, 926.
Ha ̈kkinen, K., Pakarinen, A., Alen, M., & Komi, P. V. (1985). Serum hormones during prolonged training of neuromus- cular performance. European Journal of Applied Physiology, 53, 287293.
Halson, S., & Jeukendrup, A. (2004). Does overtraining exist? An analysis of overreaching and overtraining research (review). Sports Medicine, 34, 967981.
Hautala, A., Kiviniemi, M., & Tulppo, M. (2009). Individual responses to aerobic exercise: The role of the autonomic nervous system (review). Neuroscience and Biobehavioral Re- views, 33, 107115.
Hautala, A., Ma ̈ kikallio, T., Kiviniemi, A., Laukkanen, R., Nissila ̈, S., Huikuri, H., et al. (2003). Cardiovascular auto- nomic function correlates with the response to aerobic training in healthy sedentary subjects. American Journal of Physiology: Heart and Circulatory Physiology, 285, 17471752.
Hayano, J. Y., Sakakibara, Y., Yamada, A., Yamada, M., Mukai, S., Fujinami, T., et al. (1991). Accuracy of assessment of cardiac vagal tone by heart rate variability in normal subjects. American Journal of Cardiology, 67, 199204.
Hedelin, R., Bjerle, P., & Henriksson-Larsen, K. (2001). Heart rate variability in athletes: Relationship with central and peripheral performance. Medicine and Science in Sports and Exercise, 33, 13941398.
Huovinen, J., Tulppo, M., Nissila ̈, J., Linnamo, V., Ha ̈kkinen, K., & Kyro ̈la ̈inen, H. (2009). Relationship between heart rate variability and the serum testosterone-to-cortisol ratio during military service. European Journal of Sport Science, 9, 277284.
Iwasaki, K., Zhang, R., Zuckerman, J., & Levine, B. (2003). Doseresponse relationship of the cardiovascular adaptation to endurance training in healthy adults: How much training for what benefit? Journal of Applied Physiology, 95, 15751583.
Jones, T. (2007). Testosterone associations with erectile dysfunc- tion, diabetes and the metabolic syndrome. European Urology, 6 (suppl.), 143167.
Kiviniemi, A. M., Hautala, A. J., Kinnunen, H., & Tulppo, M. P. (2007). Endurance training guided individually by daily heart rate variability measurements. European Journal of Applied Physiology, 101, 743751.
Kiviniemi, A. M., Hautala, A. J., Seppa ̈nen, T., Ma ̈kikallio, T. H., Huikuri, H. V., & Tulppo, M. P. (2004). Saturation of high- frequency oscillations of RR intervals in healthy subjects and patients after acute myocardial infarction during ambulatory conditions. American Journal of Physiology: Heart and Circula- tory Physiology, 287, 19211927.
Kraemer, W. J., Gordon, S. J., Fleck, S. J., Marchitelli, L. J., Mello, R., Dziados, J. E., et al. (1991). Endogenous anabolic hormonal and growth factor responses to heavy resistance exercise in males and females. International Journal of Sports Medicine, 12, 228235.
Kyro ̈ la ̈ inen, H., Karinkanta, J., Santtila, M., Koski, H., Ma ̈ntysaari, M., & Pullinen, T. (2008). Hormonal responses during a prolonged military field exercise with variable exercise intensity. European Journal of Applied Physiology, 102, 539546.
Malliani, A., Pagani, M., Lombardi, F., & Cerutti, S. (1991). Cardiovascular neural regulation explored in the frequency domain. Circulation, 84, 482492.
Montano, N., Ruscone, T. G., Porta, A., Lombardi, F., Pagani, M., & Malliani, A. (1994). Power spectrum analysis of heart rate variability to assess the changes in sympathovagal balance during graded orthostatic tilt. Circulation, 90, 18261831.
Nindl, B., Barnes, B., Alemany, J., Frykman, P., Shippee, R., & Friedl, K. (2007). Physiological consequences of U.S. army ranger training. Medicine and Science in Sports and Exercise, 39, 13801387.
Nindl, B., Rarick, K., Castellani, J., Tuckow, A., Patton, J., Young, A., et al. (2006). Altered secretion of growth hormone and luteinizing hormone after 84 h of sustained physical exertion superimposed on caloric and sleep restriction. Journal of Applied Physiology, 100, 120128.
Opstad, P. (1992). Androgenic hormones during prolonged physical stress, sleep, and energy deficiency. Journal of Clinical Endocrinology and Metabolism, 74, 11761183.
Pagani, M., Lompardi, F., Guzzetti, S., Rimoldi, O., Furlan, R., Pizzinelli, P., et al. (1986). Power spectral analysis of heart rate and arterial pressure variabilities as a marker of sympatho-vagal interaction in man and conscious dog. Circulation Research, 59, 178193.
Pichot, V., Busso, F., Roche, F., Garet, M., Costes, D., Duverney, D., et al. (2002). Autonomic adaptations to intensive and overload training periods: A laboratory study. Medicine and Science in Sports and Exercise, 34, 16601666.
Salo, M. A., Huikuri, H. V., & Seppa ̈nen, T. (2001). Ectopic beats in heart rate variability analysis: Effects of editing on time and frequency domain measures. Annals of Noninvasive Electrocar- diology, 6, 517.
Santtila, M., Ha ̈kkinen, K., Karavirta, L., & Kyro ̈la ̈inen, H. (2008). Changes in cardiovascular performance during an 8-week military basic training period combined with added
endurance or strength training. Military Medicine, 173,
Santtila, M., Kyro ̈ la ̈ inen, H., Vasankari, T., Tiainen, S., Palvalin,
K., Ha ̈kkinen, A., et al. (2006). Physical fitness profiles in young Finnish men during the years 19752004. Medicine and Science in Sports and Exercise, 38, 19901994.
Sonna, L., Sharp, M., Knapik, J., Cullivan, C., Angel, K., Patton, J., et al. (2001). Angiotensin-converting enzyme genotype and physical performance during US Army basic training. Journal of Applied Physiology, 91, 13551363.
Tracey, K. J. (2002). The inflammatory reflex. Nature, 420, 853859.
Tulppo, M., Hautala, A., Ma ̈kikallio, T., Laukkanen, R., Nissila ̈, S., Hugson, R., et al. (2003). Effects of aerobic training on heart rate dynamics in sedentary subjects. Journal of Applied Physiology, 95, 364372.
Tulppo, M., Ma ̈kikallio, T., Seppa ̈nen, T., Laukkanen, R., & Huikuri, H. (1998). Vagal modulation of heart rate during exercise: Effects of age and physical fitness. American Journal of Physiology: Heart and Circulatory Physiology, 274, 424429.
Tulppo, M., Ma ̈kikallio, T., Takala, T., Seppa ̈nen, T., & Huikuri, H. (1996). Quantitative beat-to-beat analysis of heart rate dynamics during exercise. American Journal of Physiology: Heart and Circulatory Physiology, 271, 244252.
Wranicz, J., Rosiak, M., Cygankiewicz, I., Kula, P., Kula, K., & Zareba, W. (2004). Sex steroids and heart rate variability in patients after myocardial infarction. Annals of Noninvansive Electrocardiology, 9, 156161.
Young, D. S. (1993). Implementation of SI units for clinical laboratory data. American Journal of Clinical Nutrition, 57, 98113.