5 Discussion

The analysis of the heart rate variability (HRV) has been applied to several examinations in different research areas as well as in clinical studies. The spectra of the power spectral analysis in particular give a non invasive insight in the vegetative control of the heart due to the classification of the sympathetic and parasympathetic activity with respective frequency bands. Therefore the sympathetic activity is classed with low (LF 0.15-0.04 Hz) and the vagal activity with high frequency (HF 0.4-0.15 Hz) fluctuations. The HF consists of pure vagal whereas the LF of a composition of sympathetic and vagal activity which may not be numerically quantized. Only the ratio between the sympathetic and parasympathetic activity may be expressed by the normalized LF and HF units as well as the LF/HF ratio which is often designated as a marker of the sympathovagal balance. Yet, the physiological significance of each HRV parameters and its correlations remains unknown so that provocation tests, e.g. the orthostatic test, are applied specifically to unbalance the autonomic nervous control of the heart which results in modulated HRV results.

Two main questions in relation to the HRV have been examined in the present study:

  1. Can the analysis of the HRV be a valuable tool to compare the vegetative control of the heart in endurance trained subjects who were included in individual daily trainings with sedentary controls?
  2. Do normal ovulatory athletic and sedentary females exhibit any effects in the autonomic nervous control of the heart investigated by the analysis of the HRV during the course of the menstrual cycle?

The results of the present study are presented in the following chapters.


[page 95↓]

5.1  Profile of mood state

In the present study, the daily mood state of trained and untrained women did not fluctuate during the menstrual cycle. The pre and the test month of all women were similar in the sub and the overall scores of the profile of mood states (POMS). No menstrual cycle influences which might have affected the POMS were noted in the pre and in the test month and the results of the POMS remained similar throughout the menstrual cycle. Therefore, our results are in line with the findings of Sato et al. [76] who also found no affected mood states in sedentary women in course of the menstrual cycle.

Additionally, the overall score of the pre and the test month in females was not significantly different compared with the overall score of men. Although differences in the sub scores of athletes compared with sedentary subjects were noted, the overall scores remained similar.

Several authors [15, 61, 67] investigated affected mood states by POMS during training periods in athletes which resulted in enhanced negative feelings. Nevertheless no difference in the overall score between trained and untrained subjects was noted and no inverted iceberg profiles (i.e. marker of enhanced training-induced negative feelings) found in the present study. Thereby, the individual training pattern in athletes did not affect the results of the daily POMS questionnaire. The overall scores of trained and untrained men and women remained in the normal range which implies unaffected mood states in course of the study. Thus, in accordance to Morgan et al. [59], no sign of staleness and/or overtraining was found in athletes during the study. Based on these findings, the requirements for the HRV analysis were met.

5.2 Blood borne parameters

5.2.1 Monitoring of the menstrual cycle

In our study, the menstrual cycle was divided into five different phases which depended on the characteristics of the hormonal fluctuations. The phases were individually determined according to the basal body temperature of the preceding month. All women had a normal ovulatory cycle including a typical course of the hormonal fluctuations despite of different individual hormonal levels. The concentrations of LH, FSH, E2 and P were at basal i.e. lowest level during the menstruation. In the follicular phase, LH, [page 96↓]FSH and E2 gradually increased with its peak at the ovulation phase (O). P only increased after O and remained enhanced till the pre menstruation phase whereas LH, FSH and E2 decline after O and approach to the basal level again in the pre menstruation phase. Therefore, cyclic changes of LH, FSH, E2 and P concentrations of women involving in this study could be shown. Moreover, no hormonal differences between trained and untrained women were noted. Based on these results, the requirements for the HRV analysis were met.

5.2.2 Monitoring of hormonal fluctuation in men

The hormonal fluctuations in men were analysed to determine whether there were training induced hormonal fluctuations in athletes compared with sedentary men and in course of one month. SHBG and the total testosterone (tT) levels were analysed in trained and untrained men during the study whereas the free androgen index (fAi) and the percentage of free testosterone (fT) were calculated. In spite of individually different hormonal levels, SHBG, tT, fAi and fT remained similar in athletes and sedentary males during the study month. These findings suggest that the hormonal levels were not affected by the individual training pattern in athletes and that there were no hormonal fluctuations in course of one month in males.

5.2.3 Glucose and insulin concentration

The blood glucose (BG) and the insulin (INS) concentration were supposed to have an influence on the vegetative control of the heart i.e. the HRV. Therefore, subjects were requested to refrain from eating and drinking for at least 8 hours before the ECG recordings. The determination of the BG and the INS served to control the fasting value of subjects´ BG and INS level. The results of the present study showed that the BG levels remained low during the study and were similar in athletes and sedentary subjects without any difference between males and females. Based on these findings it could be assumed that the subjects were fasting 8h prior to the study days. The insulin (INS) levels were also at fasting values and stable throughout the study month. Still lower INS concentrations were found in male and female athletes compared to the sedentary subjects. Nevertheless, no relation between the BG and INS levels as a marker of the metabolic supply was found in athletes or sedentary subjects. Furthermore, diabetes mellitus could be excluded and fasting metabolic supply controlled during the study.


[page 97↓]

5.2.4  Electrolytes and blood count

The electrolytes sodium (Na+), potassium (K+), magnesium (Mg²+), calcium (Ca²+), chloride (Cl-) were determined in the blood serum to control its homeostasis during the study. Affected Na+, K+, Mg²+, Ca²+ and Cl- concentrations may lead to modulation of the HRV because of its effect on the conduction system of the heart and therefore the control of electrolytes. During the five study days, the electrolytes remained stable and in the normal range in athletes and sedentary subjects. No intergroup difference was noted.

The haemoglobin (Hb) and the hematocrit (Hc) concentration also remained stable and in the normal range in trained and untrained subjects. Differences between men and women were only noted in the Hc level which was lower in females than in males. Still, the Hc was similar throughout the study month in all subjects. Based on these findings, any kind of anaemia as well as affected balance of the fluid concentration could be excluded in athletes and sedentary volunteers and the requirements for the HRV analysis were thereby met.

5.3 Heart rate variability at rest

In this study, the heart rate variability (HRV) of endurance trained athletes, who were supposed to show an improvement of the vegetative control expressed by enhanced vagal and/or reduced sympathetic activity due to the lower resting heart rate were compared with sedentary i.e. moderately active people. The HRV of male and female subjects were investigated by short time ECG recordings at rest in the time and the frequency domain.

The present results show that male and female athletes had a significantly lower resting heart rate as well as increased HRV parameters in the time domain compared with sedentary subjects. The enhanced variability was expressed by an increased SDNN, by higher RMSSD and pNN50 values in athletes which reflect augmented vagal activity. In the frequency domain, the total power and the LF power were significantly enhanced in athletes whereas the HF power was similar in trained and untrained subjects.

Although the results of the time domain indicated an increased vagal activity in trained subjects, the power spectral analysis did not because the HF power as a measure of short term variability and vagal activity was not increased. On the basis of these findings, the spontaneous breathing frequency and the individual training pattern were [page 98↓]supposed to affect the HRV results in the frequency domain whereas any kind of overtraining e.g. enhanced negative feeling and fatigue could primary be excluded because of the POMS evaluation.

5.3.1 Heart rate variability affected by the breathing frequency

The breathing frequency (BF), which modulates the HF power by its strong relation with the respiratory sinus arrhythmia (RSA), was significantly lower in trained than in untrained men whereas the opposite was noted in the female groups. The main BF of male athletes was found inside the LF instead of the HF power band whereas the other groups did not show any main respiratory frequencies in the LF area.

The slow BF in male athletes was supposed to induce a roll-off of the respiratory-linked oscillations into the area below 0.15 Hz. That means that the HF power peak shifted into the LF power band, which resulted in an augmented LF and a reduced HF power. For the first time, Melanson et al. [57] observed such a shift in a pilot study with endurance trained athletes who had an extremely low spontaneous BF. They supposed an overlap of the HF in the LF power band, which had affected the results in the power spectral analysis.

In the present study, male athletes showed such a shift of the respiratory-linked oscillations from the HF into the LF power band which can be illustrated best by two different power spectra of the same male athlete. Primary spectrum 5-1 presents a HF power peak inside the HF bands and secondary spectrum 5-2 illustrates a shift of the HF peak inside the LH power band due to slower BF.


[page 99↓]

Spectrum 5-1:Power spectrum with a BF of 11.2 breaths/min and its main frequency around 0.19 Hz; the HF power peak can be noted between 0.15-0.25 Hz inside the HF band

Spectrum 5-2: Power spectrum with a BF of 5.2 breaths/min and its main frequency around 0.09 Hz; the HF power peak shifted inside the LF band between 0.15-0.04 Hz where several peaks can be noted

In summary, male athletes did not show significantly increased HF power due to an overlap of the HF into the LF power area which affected the HRV results in the frequency domain. However, we found significantly enhanced LF power which resulted from the shift of the HF inside the LF power in trained compared with untrained subjects. This indicates that absent HF power not necessary correlates with reduced vagal activity in male athletes. Lower BF which is often accompanied with a higher [page 100↓]level of endurance capacity could therefore prevent from the application of power spectral analysis to determine the autonomic nervous activity to the heart.

Although female athletes did not cause a power shift induced by slow BF, a permanent increase of the BF was noted in course of the menstrual cycle. Trained women showed significantly enhanced BF in the middle of the luteal (MidL) and the pre menstruation (PreM) compared with the menstruation (M) phase whereas the untrained females failed to demonstrate this. The main respiratory frequency remained in the HF power band i.e. beyond 0.15 Hz in athletic women. Nevertheless, a shift of the HF power could be found in female athletes due to the permanently enhanced BF during the five phases. Thereby the main respiratory frequency changed from phase to phase. Some authors [6, 9, 43, 44, 86] investigated the HRV at different BF and noted a decreased HRV, i.e. HF power, in an inverse relationship to the BF. Still we failed to demonstrate a diminished HF power at increased BF in female athletes. On the one hand, simultaneously increased tidal volume which results in higher HRV, i.e. HF power might have neutralized the effect of the enhanced BF. On the other hand, the individually different training pattern of the female athletes might have affected the HF power and thus the HRV results in the frequency domain. And finally, other mechanisms as a modulated sensitivity of the respiratory as well as cardio circulatory control, which includes the neurons to PCO2 and to the pH during the menstrual cycle, have to be assumed.

Yet, the HF power in female athletes remained nearly unaffected in the present study. Nevertheless the enhanced BF caused a shift within the HF band which can be illustrated best by the two different spectra. Spectrum 5-3 presents the power spectrum of an untrained man with similar main respiratory frequencies which resulted in stable HF power peaks around 0.2 Hz. In comparison, spectrum 5-4 presents the power spectrum of a woman who showed permanently enhanced BF which caused a change of the main respiratory frequencies in course of the menstrual cycle. This change resulted in a shift of the HF power peak from phase to phase however remaining within the HF band.


[page 101↓]

Spectrum 5-3: Five power spectra of an untrained man with HF peaks around 0.2 Hz; unaffected HF power due to the BF

Spectrum 5-4: Power spectra of a woman with HF peaks shifting between 0.15-0.3 Hz in course of the menstrual cycle; still the HF power peaks remained in the HF area


[page 102↓]

5.3.2  Heart rate variability and individual training pattern

In the present study, LF power was significantly increased, but HF power was not changed in trained subjects. Significantly lower resting heart rate (HR) as well as increased HRV was noted in the time domain. The reason for the discrepancy between low HR and missing increased HF power, which is an indicator for the vagal activity, is due to the low breathing frequency in athletes as pointed out in the previous chapter.

The present findings can be compared to other authors. Primarily the study of Furlan et al. [28] noted a depressed HF power for 48h in untrained subjects after maximal exercise training. They found greater HF power in trained swimmers during a break in the yearly training compared with those swimmers who were involved in the training. Moreover, the HF power was noted to be reduced when the swim training resumed. Based on the findings of Furlan et al. [28], the HVR was supposed to be affected by maximal exercise in untrained as well as by yearly training periods in athletes. This affection was found to result in a depressed HF power up to 48h. Additionally, Melanson et al. [57] investigated the HRV in three different kinds of training groups (high, moderate, low) which were separated in relation to their self reported physical activity level. They found a significantly increased HRV in the time and the frequency domain in the low compared with the moderately and highly active subjects. But they missed to demonstrate any differences in the HRV i.e. the HF power between the moderately and highly active groups. Due to this, Melanson et al. [57] supposed that a depressed HF power was noted in highly active subjects because their training happened more often and with higher intensity than the training of the moderate active group. Furthermore, Janssen et al. [35] compared cyclists and sedentary subjects and finally found a persistent sympathetic activation in athletes up to 24h after exercise in cyclists. Based on these three studies [28, 35, 57] it may be argued that exercise could be supposed to diminish the vagal activity shown by depressed HF power up to 48h and to increase the sympathetic activation up to 24h in athletes [35].

The athletes included in this study were all long term endurance trained with at least 2 years of training experience. Male and female athletes did their individual training but none were involved in competitions at the time of the study. Still, the aerobic capacity of athletes was examined by a maximal ergometric test. The median of the maximal oxygen uptake in relation to the body weight (VO2rel) was 62.8 ml/min/kg for men and 50.9 ml/min/kg for women. During the course of the study, athletes were allowed to maintain their habitual training pattern which was recorded daily in a training diary [page 103↓]including the training duration and its intensity. Based on these self-reported data, men and women showed an average of 4 training sessions per week consisting of 8 hours/week in male and 6.5 hours/week in female athletes at comparable intensity. The different individual range of the sessions lasted up from 3-7 sessions/week. This would mean that athletes were not able to rest 24-48h prior to the HRV measurements. In consequence, the vegetative control of the heart may have been affected by the training intensity in athletes. Finally, this affection contributes to enhanced LH and diminished HF power during the present study which would be consistent with the above mentioned findings [28, 35, 57].

5.3.3 Summary

The results of the present study showed a significantly enhanced HRV in endurance trained athletes in the time domain whereas in the frequency domain the LF rather than the HF power was increased in athletes compared with sedentary controls.

These findings can be explained primarily by the low breathing frequencies noted only in male athletes which caused a shift of the HF into the LF power band. Due to the overlapping of both frequencies in one, an enhanced LF power was found in athletes as well as similar HF power in trained and untrained groups.

The BF nearly unaffected the HF power of female athletes, which implied to examine other mechanisms affecting the spectral power analysis. Therefore, the different individual training in course of the study was additionally supposed to affect the HRV results. Several studies [28, 35, 57] noted an increased sympathetic activity up to 24h and a depressed vagal activity up to 48h after the training intervention. Considering the training patterns during the study which were led by a protocol, the above mentioned rest between 24-48h after the training was not given prior to the HRV measurements. In consequence, the results of the power spectral analysis were supposed to be affected in trained due to the regular training.

In summary, the HRV was modulated primarily by the training influence in athletes and additionally by the low breathing frequency in trained men.

5.4 Heart rate variability during the menstrual cycle

An additional aim of this study was to determine whether the vegetative control of the heart was modulated by the hormonal fluctuations in normal ovulatory women during [page 104↓]one menstrual cycle. Thus, endurance trained and sedentary females were investigated in five different phases which had been individually determined based on the basal body temperature of one month prior to the study.

Several studies [30, 50, 75, 76, 97] already compared the HRV in different menstrual cycle phases in sedentary females but with methodological differences. Most investigators [30, 75, 76, 97] suggested a modulated vegetative control based on some selected HRV results whereas one author [50] did not find any HRV modulations in the time and the frequency domain in course of the menstrual cycle. Due to these disagreements and the differences between the studies, the HRV was investigated by short time ECG recording at rest in the present study. The parameters of the time and the frequency domain were evaluated in two female groups; trained and untrained women. The monitoring of the menstrual cycle was done by hormonal analysis which represented hormonal fluctuations of LH, FSH, E2 and P of normal ovulatory females. In addition to this, no affected mood state were noted in women based on the evaluation and the comparison between pre and test month of the daily POMS questionnaire. In summary, menstrual disturbances and/or dysfunctions as well as any affected mood states could be definitely excluded as having a modulating effect on the HRV in athletic and sedentary females in the present study.

Nevertheless, no significant difference was noted in the HRV parameters between five different phases i.e. menstruation, middle of follicular, ovulation, middle of luteal and pre menstruation phase. The results remained similar in the time and the frequency domain in course of one menstrual cycle. No significance was noted between the athletic and the sedentary females; both groups reacted comparably throughout the five study days. That means that the individual training pattern of the endurance trained women did not affect the vegetative control of the heart in the course of the menstrual cycle.

The present findings are in line with Leicht et al. [50], who also did not find any affected HRV during the menstrual cycle. Due to this consistency and based on our entire HRV investigations we conclude that the autonomic nervous control of the heart is not directly modulated by the hormonal fluctuations in normal ovulatory women. Additionally, no difference was noted between athletes and sedentary women; this implies that the physical activity level did not affect the course of the HRV throughout five menstruation phases. Thus, athletes and sedentary females showed an unaffected HRV course throughout one menstrual cycle.


[page 105↓]

5.5  HRV during the orthostatic test

The orthostatic test is the best known test to provoke an increase of the sympathetic and/or a decrease of the vagal activity by an active or passive body position change from supine to standing. The vagal predominance i.e. the enhanced HF power in supine position and the shift towards the increased sympathetic and/or the decreased vagal activity i.e. the enhanced LF power while standing are well described [44, 84, 90]. However, the different reaction of trained and untrained subjects during the orthostatic test was rarely investigated. Janssen et al. [35] compared cyclists and sedentary controls during supine rest and while standing. They found significantly decreased meanNN while standing and suggested a reduced parasympathetic activity in cyclists in this position. Nevertheless while lying supine, a vagal predominance was noted in cyclists.

In the present study the orthostatic test was done primarily to compare the provocation in athletes and sedentary subjects. The active standing was considered as a physical work-load and therefore differences between endurance trained and moderately active subjects were supposed while standing as well as in the second supine position. Furthermore, the orthostatic test was investigated in women to examine the provocation in five different menstrual cycle phases. The orthostatic tests were assumed to be affected by the hormonal fluctuations in females as shown by Saeki et al. [75], who noted differences between the orthostatic tests in five menstrual cycle phases and suggested a modulated reflex control of the autonomic functions during the menstrual cycle in sedentary women.

5.5.1 Orthostatic test in athletes and sedentary subjects

In the time domain, the orthostatic reaction was similar in males and females and without any differences between athletes and sedentary subjects. While standing the HRV parameters of the time domain were reduced compared with the supine position. Based on these findings, we could prove the assumed vagal predominance at rest as well as the increased sympathetic and/or decreased parasympathetic activity while standing. In the frequency domain the reaction of the total power (TP) was comparable with the results of the time domain due to the reduced TP while standing and its enhancement at rest. Also an increased LF power was only noted in trained men whereas the LH power remained similarly or slightly decreased in the three other groups while standing. Nevertheless, the HF power was similarly reduced while standing and thus a comparable reaction of the HF power was noted in all subjects.


[page 106↓]

The interplay of LF and HF components during the orthostatic provocation can be also demonstrated by the normalized units (LFnu and HFnu) as well as by the LF/HF ratio. But it has to be taken into consideration that the normalized units only represent the relative value of each power component in proportion to the total power minus the VLF component. This representation only emphasizes the controlled and balanced behaviour of the sympathetic and parasympathetic nervous branches which tends to minimize the effect of the TP changes on the LF and HF values. Therefore, the normalized units have to be quoted by the absolute power values which were described before.

Athletes and sedentary subjects showed a reduced HFnu and an increased LFnu in the standing position whereas the opposite reaction was noted at rest which resulted in an increased LF/HF ratio while standing and a decreased one in supine position. No difference was noted in the orthostatic reaction between men and women and between trained and untrained subjects. Based on these results we supposed that the orthostatic test was not affected by enhanced physical activity e.g. endurance training in male and female athletes. The present findings are still not consistent with the study of Janssen et al. [35] who noted reduced vagal activity while standing in cyclists compared to sedentary controls during the orthostatic test.

In summary, the orthostatic provocation led to a comparable shift of the sympathovagal balance in trained and untrained subjects. The only exception of the provocation was found in the LF power which reacted different in male athletes compared with the other groups. On one hand, the different course of the LF power during the orthostatic test was assumed to be due to the similar affection of the slow breathing frequency in male athletes as seen in chapter 5.5.3. On the other hand, the diminished TP while standing was also supposed to affect the LF and the HF power composition in the spectrum. These possible explanations were additionally confirmed by the similar orthostatic reaction in normalized units as well as by the LF/HF ratio seen in all groups. The orthostatic provocation can be illustrated best with the help of a power spectrum. Spectrum 5-5 presents the orthostatic provocation of a trained woman whereas spectrum 5-6 presents the similar reaction of a sedentary male. Although the female athlete shows higher power density peaks than the untrained men, the orthostatic reaction remained comparable.


[page 107↓]

Spectrum 5-5: The orthostatic test presented in three different parts (i.e. supine, standing and supine) of a trained woman

Spectrum 5-6: The orthostatic test presented in three different parts (i.e. supine, standing and supine) of a trained man

5.5.2 Orthostatic test during the menstrual cycle

In normal ovulatory women, the orthostatic test was investigated in five different phases i.e. menstruation, middle of follicular, ovulation, middle of luteal and pre menstruation phase. Despite different hormonal fluctuations which have been demonstrated in chapter 5.2.1, similar orthostatic reactions were noted in athletic and sedentary women. No [page 108↓]significant difference was found in the HRV parameters of the time and the frequency domain during the course of the menstrual cycle. Thus, the orthostatic provocation remained comparable in trained and untrained women over the five phases. This implies that the provocation of the orthostatic test was not directly affected by the menstrual cycle and its hormonal fluctuations in females. These findings are not in line with the results of Saeki et al. [75] who found modulated orthostatic reactions in the luteal and pre menstruation compared with the menstruation phase in sedentary women.

5.5.3 Summary

In the present study the orthostatic test has been proven in all groups. Due to the reduced TP while standing, the vagal predominance at rest as well as the enhanced and/or diminished sympathetic activity could be shown by results of the normalized units. Nevertheless we failed to demonstrate a different orthostatic reaction between trained and untrained subjects. Moreover, the orthostatic reaction was also not affected by the hormonal fluctuations of normal ovulatory females. No relation between the orthostatic reaction and the menstrual cycle in athletic and sedentary females could be found in this study.

5.6 Reliability

In the present study the reliability of the repeated measurements was controlled by the intragroup correlations coefficient (ICC), the coefficient of variation (CV) and the Bland-Altman plot which tested the intra- as well as the intergroup reliability. Although the data of this study were not normally distributed, the ICC, the CV and the Bland-Altman plots were analysed because equally good tests for non parametric data do not exist. In addition to this, universal standards for the ICC and the CV are missing. Thereby the ICC was determined in accordance to Lee et al. [49] whereas the CV was subjectively discussed considering the present data.

The reliability was determined from the ICC, the CV and the Bland-Altman plots ranged from acceptable to good in the time domain. However the reliability was poor in the frequency domain. Due to these findings, we suggest individual biological variance and their effects leading to the debased reliability in the frequency domain. Thus, further investigations are needed to examine these biological variances and their effects on the heart rate variability.


[page 109↓]

The HRV reproducibility has been extensively assessed by other authors and found to be acceptable for measurement durations as short as 2.5 min in resting position in the time as well as the frequency domain [49]. Nevertheless, the reproducibility has not been investigated in the present study because the comparison of the algorithms used in this study with algorithms of a so-called golden standard is missing in the HRV investigation. An additional problem is that the hard- and software as well as the combination of algorithms, used to detect and evaluate the HRV, are mostly proprietary, not well described and therefore not always duplicated.

In accordance to the Task Force [84] comparable accuracy of the HRV measurements, reported by different commercial equipment, may only be achieved by devices, which have been tested independently of the manufacturer. Several short and long time inventions with precisely known HRV parameters and with different morphological characteristics of the ECG signal should be involved. In particular, both the recording and the analytical part of the device should be tested. Alongside, an appropriate technology should be used to record fully reproducible signal with precisely known HRV parameters, which have to be done by generated computer software and/or hardware. In summary, technical reports of the testing should be prepared and established for further HRV measurements [84].


[page 110↓]

5.7  Critics of the method

A possible limitation of this study was the density of the different menstrual cycle phases in women. Alternatively it might be possible, that the examination of more than five phases could demonstrate other HRV results because of a menstrual cycle monitoring with close-meshed control. At least, the influence of the menstrual cycle and its hormonal fluctuation of the vegetative control of the heart in women might only be completely excluded by studies which investigate daily HRV measurements in course of the menstrual cycle. Beside this, the duration of the study which consisted of a pre and a test month in women might have been too short to examine the influence of the menstrual cycle on the HRV in women.

It was of interest in this study to determine whether a short time recording of the HRV might be a practicable tool to investigate physiological mechanisms during individual training interventions. Because of this, a short time instead of a 24h ECG recording was used. On point of critic could be that the recording period of 10 min used in the present study might have limited the examination of the vegetative control of the heart in daily active athletes. Still it is not proven whether a 24h ECG recording really contains more detailed information of the HRV. In addition to this, the intraindividual differences of the subjects´ pre analysis between the single study days which were found beside standardized study conditions can also be mentioned, here. But even under more controlled conditions, the differences between biological systems as humans would remain consistent. And thus, the total standardisation of the pre analytics in human beings keeps impossible. The individual differences between subjects will always remain.

Finally, the HRV in the frequency domain was analysed by the Fast Fourier transform (FFT) which requires stationary ECG signals. Although biological systems may not reach conditions producing stationary signals, the FFT is used to investigate the power spectral analysis of the HRV because of poor available alternatives in mathematical calculations. On this account, other mathematical possibilities which might be applicable in further studies have to be assessed for non stationary ECG signals.


[page 111↓]

5.8  Applicable consequences in sports medicine

Based on the reliability of the present data, the applicability of the HRV analysis is given in the time but not in the frequency domain. Repeated measurements as well as short time recordings are applicable in relation to training interventions of male and female athletes. Unfortunately, data of comparable work-loads and/or training interventions in sports medicine are still missing. Additionally, the diagnostic significance in relation to sensitivity and specification of the HRV analysis remains unclear. Thus, the HRV analysis is not yet a practicable tool to supervise and/or control the training process in athletes which might prevent a developing overstrain and overtraining. Further investigations are needed.

However, we found that the vegetative control of the heart in normal ovulatory females is not affected during the menstrual cycle; trained and untrained women reacted similar. Thus, female athletes do not have any affected autonomic nervous control of the heart by endogenous hormonal fluctuations. Based on these findings, we suggest that the HRV can be analysed in normal ovulatory females irrespective of the menstrual cycle.

5.9 Future prospective

Future prospective should involve the standardisation of the HRV procedures and its methods which include the testing of the measurement equipment as well as the hard- and software used to detect and calculate the HRV. Besides, new mathematical calculations (i.e. algorithms) have to be developed to analyse primary non stationary ECG signals and secondary ECG signal affected by slow breathing frequencies found mostly in athletes. Such mathematical possibilities would finally allow improving the investigation of the HRV i.e. the vegetative control of the heart in athletes. And thus, mechanisms and parameters which control the physical work load might be better examined in athletes. Unfortunately, the examination of the HRV in athletes is limited till today due to the missing mathematical alternatives in sports medicine. At least the available concepts used to examine the sympathovagal control of the heart should also have been taken under consideration. What, if these would be the limitations to improve the determination of the vegetative control of the heart in biological systems?


© Die inhaltliche Zusammenstellung und Aufmachung dieser Publikation sowie die elektronische Verarbeitung sind urheberrechtlich geschützt. Jede Verwertung, die nicht ausdrücklich vom Urheberrechtsgesetz zugelassen ist, bedarf der vorherigen Zustimmung. Das gilt insbesondere für die Vervielfältigung, die Bearbeitung und Einspeicherung und Verarbeitung in elektronische Systeme.
DiML DTD Version 3.0Zertifizierter Dokumentenserver
der Humboldt-Universität zu Berlin
HTML generated:
06.01.2005