2 Theory

2.1 Physiology

2.1.1 The autonomic nervous system

The vegetative nervous system innervates the smooth musculature, the heart and the glands. Its function serves for the neuronal control of the internal environment, the so-called homeostasis, and the adaptation of the internal environment to external stimuli (e.g. physical activity). Organs and organ systems without direct relation to the homeostasis as the neuronal control of the sexual organs and the inner eye muscles are also controlled vegetatively. Due to the involuntary control, the vegetative nervous system is also known as the autonomic nervous system (ANS). The ANS works together with the voluntarily controlled somatic nervous system which controls the afferent and efferent interaction with the environment.

The ANS consists of three different parts: the sympathicus, the parasympathicus and the intestinal nervous system. The intestinal nervous system, a special nervous system of the gastrointestinal tract, functions without extrinsic influence of the spinal cord and the brain stem.

The terminal neurons of sympathicus and parasympathicus are located outside the central nervous system. The anatomical part of the sympathicus, the thoracal lumbar system, originates from the thoracic medulla, the second and third segment of the lumbar medulla. The target organs of the sympathicus are the smooth vessels of every organ, the heart and the glands. The parasympathicus originates from the brain stem and the sacral medulla. The preganglionic parasympathetic fibres to the organs of the [page 5↓]abdominal and thoracal cavity originate from the nervous vagus. Therefore the expressions “nervous vagus” and “vagal” instead of “parasympathetic nerve” and “parasympathetic” are also known. The smooth musculature, the glands of the gastrointestinal tract, the excretion, the sexual organs, the salivary, the lacrimal glands, the inner eye musculature and the lungs are innervated by the parasympathetic activity beside the neuronal control of the heart.

The sympathetic and the parasympathetic nerves react in an interaction of different nervous activity levels. Due to this, the expression sympathovagal or sympathetic-parasympathetic balance is used. The quantification of the contribution of the two nervous branches remains difficult because both parts are permanently active.

The ANS supervises the internal environment of the body by receptors i.e. baroreceptors, chemo- and mechanoreceptors which compare constantly the actual values with set points in a closed control loop to keep the homeostasis as constant as possible. Internal and external influences such as physical activity, food uptake and thermic changes are immediately balanced by specific adaptation i.e. modulated respiration, heart rate and/or blood pressure induced by the ANS.

The autonomic nerves have a pivotal role in the regulation of the cardiovascular system. The study of the cardiovascular variability is mainly assumed to access the activity of the sympathetic and parasympathetic nervous activity. The heart rate variability (HRV) analysis may examine autonomic fluctuations under different physiological circumstances (e.g. work-load, body position change). The vegetative control of the heart in relation to endogenous hormonal fluctuations is part of the focus of this study. Thus, the following chapter will only focus on the cardiovascular system and the autonomic (vegetative) control of the heart.

2.1.2 The cardiovascular system

The cardiovascular system constitutes the junction of the blood vessels including arteries, capillaries and veins with the heart. The most important assignments of this transport system are the supply of cells with nutrients and oxygen as well as the removal of metabolic products. The blood circulation system comprises the systemic and the pulmonary circulation illustrated by the following figure (figure 2-1).


[page 6↓]

Figure 2-1: Illustration of the pulmonary and the systemic circulation (data modified from [77])

The circulation of the blood through the body only functions by the pump function of the heart. The right part of the heart consisting of atrium and chamber takes up the venous blood which is poor in oxygen and supplies it to the lungs where it will be enriched (arterialisation) with oxygen. After the arterialisation the blood arrives at the left part of the heart. Afterwards it will be transported to the different organs. The pump effect of the heart is caused by rhythmic series of relaxations (diastole) and contractions (systole) of the heart chambers. The cardiac muscle (myocardium) fibres are excitable structures with resting and action potentials. A stimulation of the ventricles (chambers) is spread out over all fibres of the heart; the heart reacts on stimulation completely or not. The rhythmic pulsations of the heart happen by stimulations originating in the heart itself: the so-called auto rhythm.

The sinus node generates the impulse for a heart beat. The stimulation spreads out the atrium muscles. The atrioventricular node conducts the stimulation with a delay to the bundle of His (atrioventricular bundle), to the left and the right bundle branch and the Purkinje fibers.


[page 7↓]

Finally the stimulation reaches the chamber muscles throughout the ends of the Purkinje fibers (see figure 2-2).

Figure 2-2: Conduction system of the heart (frontal tomogram modified from [77])

Due to the highest frequency, the sinus node is the primary pacemaker of the heart. The stimulation originated at the sinus node spreads out and stimulates the whole heart; the so-called sinus rhythm.

2.1.3 Electrocardiogram

An electrical field measurable at the body surface, results from the expansion and involution of the excitation (stimulation) of the heart. The electrocardiogram (ECG) represents the potential difference in relation to the time as an expression of the excitation of the heart. Changing of potential differences illustrates the time dependent changes of the direction and size of this electrical field; measurable with electrodes attached at different parts of the body.

The ECG is classified into parts characterised by letters which stand for certain segments of the excitation of the heart (figure 2-3).


[page 8↓]

Figure 2-3: Part of an ECG recording of our study which presents the single segments classified by letters

The stimulation of the atriums starts with the P-wave implying the spread out stimulation over both atriums. The PQ-segment illustrates the completed excitation of the atriums. The stimulation expansion of the chambers starts at Q and ends at the beginning of T. The QRS-segment stands for the expansion of the ventricles and the T-wave for the ventricular involution of the stimulation. Beside this, the ST-segment illustrates the total stimulation of the ventricles. Sometimes an U-wave close to the T-wave is visible; but its meaning remains unclear.

2.1.4 The vegetative control of the heart

The vegetative heart nerves of the parasympathicus and the sympathicus have a direct influence on the action of the heart. The frequency, the systolic strength formation, the contraction velocity and the velocity of the atrioventricular conduction are affected. The effect of the vegetative nerves is chemically transmitted by acetylcholine (parasympathicus) and noradrenaline (sympathicus). The permanent stimulation of the vegetative nerves is called vagal and sympathetic tone. The sympathetic and parasympathetic nervous activity still interacts in addition to this tone.

The contraction strength of the atrium myocardium diminishes with the parasympathetic influence. This results in a shortened duration of the contraction curve (from foot till top) which reduces the duration of the action potential. The sympathetic influence increases the contraction power in the atrium as well as in the chamber myocardium. Thus, the sympathicus accelerates the atrioventricular conduction and shortens the conduction between atrium and chamber action whereas the parasympathetic influence [page 9↓]decelerates the atrioventricular conduction. Thereto the vagal and sympathetic activities constantly interact; this interaction is assumed to be reciprocal.

The control of the arterial blood pressure, the cardiac output and the blood flow distribution take place in the lower brain stem. Sympathetic and parasympathetic axons are the efferences whereas the arterial baro-, chemo- and mechanoreceptors of the atriums and the ventricles of the heart are the afferences to the cardiovascular effectors. The higher brain stem and the hypothalamus supervise the medullary self-control. The control of the medulla takes place by the neuronal connections of the hypothalamus and medullary cardiovascular centre and by the direct neuronal connections from the hypothalamus to the preganglionic neurons. The hypothalamus adapts the higher neuronal control of the cardiovascular system during complex vegetative functions i.e. thermoregulation, control of food intake and of the physical activity.

The function of the circulation is permanently controlled by receptors at different locations of the cardiovascular system. The afferent impulses of theses receptors are directed to the medulla oblongata. From there, impulses of efferent fibres proceed to the effectors in the heart and the vascular system of the central nervous system. The total peripheral resistance and the cardiac output are important adaptations of the circulation system as well as the vascular capacity and the distribution of the blood volume. The adaptations are separated into short, medium and long term regulation mechanisms. Pressoreceptor- (baroreceptors), chemo receptor- and ischemia reactions of the central nervous system are short term regulation mechanisms because of their fast (reaction) effect. Hormonal influences of adrenaline, noradrenaline and adiuretine supply the vasomotor effects and diminish the conduction. The increase of afferent impulses of the pressoreceptors cause an inhibition of sympathetic and an enhanced stimulation of the parasympathetic structures in the medulla oblongata; thereto the tonic activity of the sympathetic fibres, the heart rate, the strength and velocity of the myocardial contraction are reduced. The decrement of the total peripheral resistance results in a capacity increment of the capacitance vessels and a decrement of the systolic blood pressure. Reduced stimulation of the pressoreceptors results in the opposite reaction. Acute arterial pressure variation provokes a changing of the arterial pressoreceptors which concerns the flow resistance and the cardiac output. The aim of this reaction is the fast approach to the set point; a kind of homeostatic (self) regulation mechanism of the circulation. Due to internal and external influences on the homeostasis, specific adaptations supervised by the ANS aim to hold the set point as constant as possible.


[page 10↓]

Enhanced excitation of the pressoreceptors also inhibits the inspiratory and expiratory neurons of the respiratory centre and diminishes the respiratory frequency and the ventilation volume. The inspiration and expiration cause different pressures in the thorax which influence the blood circulation especially in the pulmonary circulation. This pressure gradient determines the volume and the velocity of the blood which thereby influences the vegetative control of the heart. The inspiration correlates with an increasing heart rate and the expiration with a decrease; this phenomenon is known as the respiratory sinus arrhythmia (RSA) due to its similarity to a sinus curve.

In summary, biological rhythms found in the vegetative nerves and conducted to the heart may also result in periodic fluctuation of the heart rate [16]. Despite of this, the sinus node remains the primary pacemaker of the heart whereas its rhythm is constantly modulated by the sympathetic and parasympathetic activity of the ANS; this activity is rhythmically modulated. Thereby a direct reflection of the autonomic activity can be expressed by the time-dependence of hemodynamic variables and therefore an insight into the cardiovascular control can be gained [2].

2.2 Physiology of the rhythm derivation

The sinus node is the pacemaker of the heart. Its rhythm is modulated by the sympathetic and parasympathetic activity. These rhythms are related to the activity of the pressoreceptors (baroreflex), the respiratory and the thermoregulatory related reflexes. Receptors are part of a closed feedback control system. They are sensitive for a certain stimulus from which they constantly compare the actual with the set point value. Finally the receptors interact with the sensory cell of a specific organ, the so-called control organ, to keep the set point as constant as possible. The feedback systems consist of non-linear elements which are mainly responsible for the development of systemic changes of the efferent nerve activity to the heart.

The pressoreceptors also known as baroreceptors control the arterial blood pressure in the arterial vessel walls. The set point in this closed control loop is the mean blood pressure in the arterial vessels. Alterations of the dilatation in the vessel walls (actual value) stimulate the pressoreceptors in the aortic arch and the carotid sinus which leads to nervous impulses to the higher brain stem (control organ). Thereby the heart rate is modulated as well.


[page 11↓]

The respiration is supervised by respiratory neurons which control the respiratory frequency and the tidal volume. The inspiration and expiration neurons which regulate the central breathing rhythm are located in the medulla oblongata. The inspiration neurons activate the inspiration muscles while the expiration neurons are simultaneously blocked by blocking neurons. The expiration starts when the effect of the blocking neurons is neutralised and the inspiration neurons diminished. Inspiration and expiration neurons permanently interact by the inhibition of each other which leads to a central rhythm of the breathing. This rhythm is additionally stabilized by peripheral influences of the stretch receptors which regulate the tidal volume of the breathing. Stretch receptors in thorax and lungs control the stretch stimulus of the inspiration and expiration. Alteration of the stimulus leads to adequate nervous impulses to the respiratory centre. This reflex control of the respiratory centre to the respiration musculature is also known as “Hering-Breuer-reflex”. The breathing rhythm is additionally modulated by the chemical breathing control. Peripheral and central chemoreceptors thereby control the CO2 and the O2 partial pressure and the pH of the arterial blood. An increasing CO2 partial pressure leads to nervous impulses to the respiratory centre which activates the respiratory musculature to enhance the tidal volume and in part the breathing frequency. The increased tidal volume finally diminishes the CO2 partial pressure to the set point value. The stretch- and the chemoreceptors modify the central breathing rhythm to stabilise the homeostasis of the breathing. The respiratory neurons and nerves for the cardiovascular control are in the medulla oblongata of the brainstem. The respiratory rhythms are mediated to the cardiorespiratory nerves, too. This is one of the sources for the modulations of the sympathetic and parasympathetic activity to the heart.

The thermoregulation in the hypothalamus keeps the core temperature at the set point by balancing the fluctuations in the heat regulation. This regulation also affects the vegetative control of the heart. Increased heat in the internal environment as enhanced core temperature, results in a dilatation of the vessels and perspiration whereas coldness provokes a constriction of the vessels and an enhanced muscle tone. The temperature is permanently controlled by cold and warm receptors in the skin which react on adapted temperature stimulus by nervous impulses to the hypothalamus.

These short term vegetative rhythms in the vegetative activity modulate the heart rate. A relation between the vegetative control system and the vegetative rhythms with rhythmic modulations are supposed to correlate with main frequencies in the HRV. [page 12↓]Main frequencies for the control systems of the pressoreceptors, of the respiratory as well as of the thermo related functions were found.

Fluctuations of the heart rate can be presented as a function of power density in relation to its frequency in a spectrum. As with the main fluctuation, the control system may be related to short term variability or long term variability which are defined by high (HF) and low frequency (LF) bands; HF 0.4-0.15 Hz and LF 0.15-0.04 Hz. HF stands for parasympathetic and LF for sympathetic activity of the vegetative control of the heart. Thus, the above mentioned control systems could be mirrored in parasympathetic or sympathetic nervous activity by its main frequency in the HRV spectrum.

The control system of the baroreflex (pressoreceptor) showed a main frequency around 0.1 Hz in the long term variability (LF band) and thereby could have a relation to the sympathetic nervous activity. The main respiratory related function was found in the short term variability (HF band) and thus could have a strong relation with the parasympathetic nervous activity which is also known as “respiratory sinus arrhythmia”. Provoked stimulations with coldness and heat indicated a thermo regulated function which fluctuated in long term variability (LF band) i.e. in the sympathetic nervous activity bands of the spectrum. Nevertheless, data concerning the thermo related control system and its frequency are inconsistent and should be handled with care [84].

In summary, the respiratory sinus arrhythmia is thought to contribute to HF component i.e. to the parasympathetic nervous activity whereas the baroreflex and thermoregulation related control systems seemed to be related to the LF component i.e. to the sympathetic nervous activity of the power spectral analysis of the HRV [16].

2.3 Measurement of HRV

The basic source of the HRV is a high quality ECG recording which requires standardized conditions to minimize artefacts and high sampling rate to avoid interferences. ECG recordings and HRV calculations should only be carried out in accordance to the guidelines of the Task Force [84].

Also, the QRS complex should have sufficient amplitude and a stable baseline. The ECG signal is first analog recorded and then digitally converted. Afterwards a stable reference point has to be located by algorithms in accordance to Friesen et al. [26] to define the RR interval series. The proper interpolation of ectopic beats, arrhythmic events, missing data and noise-effects of the RR interval results in the NN interval [page 13↓]series (NN means normal to normal) [84]. The HRV parameters are now calculated by different mathematical operations. Alongside the two main HRV analyses in the time and the frequency domain, new analysis and parameters have been also investigated. The power spectral density (PSD) analysis, also known as the frequency domain analysis, provides the basic information on how power as an expression of variance distributes in the function of frequency. This can be illustrated for example by a spectrum where the power density is presented over the frequency [84]. The validity of the HRV parameters is related to the recording duration whereas only HRV segments of identical length are can be compared.

Figure 2-4 presents the step by step derivation of the time and the frequency domain analysis of the HRV.

1st Step presents a typical ECG recording.

2nd Step shows the time-event series consisting of the NN intervals. Slow and fast fluctuations can be noted by the different interval lengths.

3rd a Step presents selected parameters of the time domain analysis which are calculated by simple mathematical operations as the mean (meanNN) or the standard deviation of all NN intervals (SDNN).

3rd b Step shows a spectrum of the frequency domain analysis which presents the variability as a function of power density in relation to the frequency.


[page 14↓]

Figure 2-4: Steps in the analysis of the HRV

2.3.1 Time Domain

In the time domain analysis the intervals between successive NN interval series or at any point in time are determined. Thus the calculation of the time as well as of the [page 15↓]frequency domain parameters is based on the NN interval series or the instantaneous heart rate which leads to different results due to the non-linear relation between the fluctuation of the NN intervals and the heart rate.

The main time domain variables for short term recordings include the mean of all consecutive NN intervals (meanNN), the standard deviation of the NN intervals (SDNN), the square root of the mean squared differences of the successive NN intervals (RMSSD) and the number of intervals greater than 50 ms (NN50) and the proportion which derives by dividing NN50 by the total number of NN intervals (pNN50). The meanNN can be used as a marker of the heart rate and the SDNN as an expression of the total variability of the heart rate. Thus, the high correlation of the RMSSD, NN50 and pNN50 estimates a relation with the short term variation i.e. the HF component of the heart rate [62, 84]. Still the time domain analysis gives only information about the amount of variability without regarding the period length of the oscillations.

2.3.2 Frequency Domain

The spectral analysis of the frequency domain decomposes the series of continuous NN intervals or instantaneous heart rates into its sum of sinusoidal functions which differ in the amplitudes and the frequencies. Thus, the main advantage of this analysis is that not only the amount of variability but also the frequency specific oscillation can be obtained [16]. The power of each component is plotted as a function of its frequency and is computed in defined frequency bands; ergo the magnitude of variability is implied as a function of frequency. Therefore the power spectral density (PSD) provides basic information of how power (variance) distributes as a function of frequency [84].

In relation to the data evaluation, the PDS can be calculated by nonparametric or parametric methods which both provide comparable results. In most cases the Fast Fourier Transform (FFT) is used because of the algorithm simplicity, the high processing speed and its stability [26]. However, in accordance to the Task Force [84], the use of nonparametric methods, e.g. FFT, is only suggested with a regularly sampled interpolation of the discrete event series (DES). Standards for the nonparametric methods (e.g. the FFT) include the formula of the DES, the frequency of the sampling DES interpolation, the number of samples used for the spectrum and the spectral window used. Hann, Hamming and the triangular window are the most frequently used spectral windows [84].


[page 16↓]

The FFT decomposes the series of sequential NN intervals into a sum of sinusoidal functions resulting in the magnitude of variability as a function of frequency. The power density in relation to the frequency can be shown best by spectral analysis which reflects the amplitude of the heart rate fluctuations at different oscillation frequencies [16].

2.3.2.1 Frequency domain parameters in absolute values

The total variability of the spectral power analysis is expressed by the total power [ms²] whereas further parameters depend on the recording time. Short-term recordings only include three main spectral components; the very low frequency (VLF), the low frequency (LF) and the high frequency (HF) power in [ms²]. These components express the absolute value of the power [ms²] in a defined frequency band. Therefore the expression HF, LF and VLF power is used. The Task Force of European Society of Cardiology and the North American Society of Pacing and Electrophysiology laid down the bands of the specific frequencies in a special report in 1996 [84]. The three main frequencies of the short term recordings are defined by the following bands:

VLF ranges from <0.04 Hz to 0 Hz
(Including the ultra low frequency (ULF) which is only accepted for long term recordings <0.003 – 0 Hz)

LF ranges from 0.04-0.15 Hz with central frequency around 0.1 Hz

HF from 0.15-0.4 Hz with central frequency at respiratory rate around 0.25 Hz

The distribution of the LF and HF power and its central frequencies are modulated by fluctuations of the cardiovascular system. Therefore main periodic fluctuations of the respiratory sinus arrhythmia (RSA), the baroreflex (pressoreceptor) and thermoregulation related reflexes were mirrored in the spectra. Due to their main frequencies, strong relations between the HF power and the RSA and the LF power and the baroreflex were found. Whereas the relation between the thermoregulation and the VLF remains disputable and specific physiological process attributable are still in doubt.


[page 17↓]

2.3.2.2  Frequency domain parameters in normalized units

HF and LF may also be expressed in normalized units which represent the relative value of each power component in proportion to the total power minus the VLF component. To avoid a mix-up of the parameters, the HF and LF in normalized units are expressed as HFnu and LFnu. The representation of HFnu and LFnu illustrates more clearly the control and the balance of the two nervous branches, i.e. the sympathetic and the parasympathetic nerves of the autonomic nervous control. The normalization of the HF and LF also tends to diminish the effect of the total power on the values of the HF and LF power. Nevertheless, the Task Force [84] recommends a quotation of the HFnu and LFnu with the absolute values of the HF and LF power in order to describe completely the distribution of power in spectral components.

2.4 Physiological interpretation of HRV

Each cardiac cycle is modulated by the integrated efferent sympathetic and vagal activities which are directed to the sinus node. Additionally, frequent small adjustments in the heart rate are made by periodic modulations of this activity, which are related to cardiovascular control mechanisms i.e. the respiratory sinus arrhythmia, the baroreflex (pressoreceptor) and the thermoregulation related HRV. Fluctuations can be illustrated in a spectrum which may help to find physiological interpretations related to the vegetative control of the heart.

Strong relations between the HF and the LF component and the sympathetic and the parasympathetic nervous activity were established by clinical and experimental observations of autonomic manoeuvres, e.g. electrical stimulation, pharmacological receptor blockades and animal research. Nevertheless the autonomic control of the heart and its relation to the VLF remains disputable. Thus, only the physiological interpretation of the HF and the LF component are described in the following chapters.

2.4.1 HF power and parasympathetic activity

The respiratory frequency approximates to 0.2-0.4 Hz at rest in general and a respiratory related modulation of the heart rate has been found. This phenomenon is known as the respiratory sinus arrhythmia (RSA). The RSA is highly correlated with the HF component of the HRV. In a spectrum, the peak of the respiratory related modulation can be found at the respiratory frequency. Controlled respiration i.e. metronome [page 18↓]breathing at different frequencies showed a higher increase of the RSA with approaching respiratory frequency to the intrinsic baroreflex related heart rate fluctuations. The maximum was found at 0.1 Hz (i.e. 6 breaths/min) whereas the RSA greater than 0.1 Hz negatively correlated with the amount of RSA [16, 43]. This implies a positive correlation of tidal volume and the amplitude of the respiratory related heart rate oscillations [62].

Beside the controlled respiration, the following investigations also showed a relation between the vagal activity (the HF component) and the RSA. Pharmacological blockades by high doses of atropine abolished the short term variability and the RSA [16]. Subjects with paraplegia, i.e. completely interrupted sympathetic nervous innervations, but with preserved nervous vagus, showed a decreased LF component but normal fluctuations in the respiratory related frequencies i.e. the HF component. Finally a nervous vagus cut in animal research models reduced the vagal activity and the RSA which illustrated a strong correlation of the vagal activity and the respiratory related heart rate fluctuations [62]. This findings support, that the parasympathetic activity is the major marker of the HF component and responsible for the respiration linked oscillation of the HRV [2]. Additionally, the RMSSA and pNN50 are also found to highly correlate with the HF component which demonstrates that both parameters of the time domain are vagally mediated [84].

2.4.2 LF power and sympathetic activity

The LF component is mediated by both branches of the vegetative control of the heart, which can be demonstrated in several conditions. High doses of atropine which block the vagal activity abolished the short term variability (HF component) and also diminished the long term variability (LF component). Beside this, propranolol, which blocks the sympathetic nervous activity attenuated, the LF without affecting the HF component. Patients with paraplegia of totally interrupted sympathetic innervations of the heart showed unaffected high fluctuations and reduced but still measurable low fluctuations of the HRV. Thus, the LF component is considered not only as marker of the sympathetic but also as a marker of the parasympathetic nervous activity [42, 62, 65, 84]. Ergo, the LF component is mediated by two parts; the sympathetic and vagal nerves with a predominant sympathetic activity. Still the precise quantification of vagal and sympathetic parts is conflicting due to the unknown distribution of the activity. Additionally, correlation between the fluctuations in the peripheral vascular resistance i.e. fluctuations of the blood pressure (baroreflex) and the LF component were found. [page 19↓]This implies that the baroreflex related fluctuations with a spectral peak around 0.1 Hz are mediated by the sympathetic nervous activity [16]. Furthermore, the LF component was noted to be increased in certain conditions as 90° tilt, standing, mental stress and moderate exercise [84].

2.4.3 Orthostatic test

The orthostatic test is the best known test to provoke an increase in the sympathetic activity. The active and the passive (i.e. tilt up table) posture change from supine to standing causes a shift of the blood volume induced by the hydrostatic pressure change. This results in reduced arterial blood and central venous pressure. Thereby, the diminished venous supply to the heart leads to a decrease of the arterial blood pressure. The activation of the sympathicus induced by the stretch- and pressoreceptors of the ANS causes a constriction of the vessels, an increase of the heart rate as well as a normalization of the arterial blood pressure.

In supine position, the vagal activity is predominant whereas a shift towards the LF component leads to an increase of the sympathetic nervous activity while standing [44, 84, 90]. The following table illustrates the physiological difference of the vegetative control of the heart between supine and standing position with the help of three different illustrations. Still, no difference was observed by Ryan et al. [73] while lying in left or right lateral position compared with the supine position.


[page 20↓]

Furthermore, the LF/HF ratio which consists of the LFnu and HFnu is used to illustrate the interaction of the sympathetic and parasympathetic activity and can be seen as a mirror of the sympathovagal balance [84]. During the orthostatic test, this balance is shifted towards the sympathetic predominance because of the LFnu increase and the HFnu decrease which directly interact. In accordance to the Task Force [84], the LF/HF ratio and the normalized units minimize the effect of the total power changes and therefore must be quoted with absolute values to describe completely the spectral power components.


[page 21↓]

Table 2-1: The orthostatic test presented in three different analysis steps based on data of the present study


[page 22↓]

2.4.4  Influence of respiration

Respiratory rate and tidal volume are known to exert major influences on RR interval or heart rate fluctuations. The spontaneous breathing is individually different and is modulated by physiological, psychological and external influences. For instance, women are known to have increased breathing frequencies with increasing progesterone level during the menstrual cycle whereas endurance trained athletes are known to have slower breathing frequencies than sedentary men. Therefore, ways to minimize the physiological differences and the external influences during the HRV measurements are needed. Thus, the controlled breathing to the rhythm of an auditory signal at a fixed rate was thought to be a valuable tool to standardize the respiratory influences on HRV. Different studies [6, 9, 53] investigated the difference between the spontaneous and the controlled breathing on the vegetative control of the heart.

Bloomfield et al. [6] compared spontaneous breathing with different metronome-set breathing rates. The HF component was determined during each respiratory cycle. They discovered that the HF power during spontaneous compared to metronome-guided breathing was significantly different at the same breathing rate. Mental concentration and subjects discomfort tended to decrease the HF power. The metronome breathing caused disturbances of the ANS and thereby the HRV was affected. Increasing sympathetic activation was noted when the manoeuvre of the metronome breathing tended to be stressful; ergo the physiological process was affected.

However, controlled respiration at frequencies within the range of resting physiological range provided to be a convenient tool to provoke an enhancement of the sympathetic modulation of the heart period [53].

In addition, Brown et al. [9] compared the influence of different metronome breathing rates on the RR interval power. Breathing rates (BR) of about 10 breaths per min. or less yielded the maximum RR interval power. BR of more than 10 breaths/min reduced the level of RR interval power. That means that the RR interval variability decreased with increasing breathing frequency (BF). At the slowest BR, maximum RR intervals were longer and minimum RR intervals shorter which also provoked major increases of LF power. Ergo, the respiratory frequency overlaps in the LF band which resulted in an increase of the LF component due to the slow breathing. Based on these findings, Brown et al. [9] supposed that slow BR only augments the LF power by a shift of the [page 23↓]power without affecting the vegetative control of the heart. Therefore, slow BR does not reflect an increased sympathetic neural outflow.

Nevertheless the best way to measure HRV under standardised conditions and without any respiratory induced influences has not yet been found. Not only metronome breathing which modifies the ANS but also the spontaneous breathing causes problems with the evaluation of the HRV. It still remains essential to control the respiration to have a feedback of the respiratory induced influence on the HRV [62, 44, 2, 43]. The way of the respiratory control only depends on the investigation of the physiological phenomena modulating the HRV.

2.5 Effect of physical activity on HRV

The effect of physical activity on the autonomic nervous control of the heart was investigated in animal models. Dogs with documented higher risks of acute myocardial ischemia were trained by running for six weeks. After the training intervention they showed an enhancement of the HRV of 74% [84]. These findings indicated the positive physiological effect of training especially endurance training on the cardiovascular system. Based on these findings the modification of the autonomic balance by physical activity was more and more investigated.

2.5.1 Physical activity affecting the cardiovascular system

Physical activity is associated with hemodynamic changes and alterations of the loading conditions of the heart [72]. The cardiovascular responses to physical activity vary in dependence of the type, the intensity and the volume load of the training. Endurance training leads to the most adaptive changes of the cardiovascular function due to the increased volume load compared with strength trainings. Based on the subject of the thesis, only the cardiovascular responses in relation to endurance training are described in this chapter.

The “athlete’s heart” is a well known phenomenon which describes the increase of the left ventricular diastolic cavity dimensions, the wall thickness and the mass after long term athletic exercise [2]. The lower heart rate leads to longer periods of the diastole which enhances the stroke volume. The higher stroke volume is induced by the larger blood volume of athletes and the Frank Starling Mechanism (FSM). The FSM describes that a prolongation of the diastole results in an enhanced cardiac work load. Endurance [page 24↓]training leads to an improvement of the heart’s ability to pump blood by an increased stroke volume which is possibly due to an increase in the end-diastolic volume and a slight enhancement of the left ventricular mass. This results in a more efficient pressure to time relationship i.e. the stroke volume increases and the heart rate decreases. At rest and at sub maximal exercise intensity, the metabolic load on the heart is thereby decreased. Increased heart rate (>100 beats/min) is called tachycardia which can occur either by neural stimulation or by an elevation in circulating catecholamines. While exercising, the increasing heart rate raises the cardiac output [2] whereas the increased heart volume and enhanced contractility lead to higher stroke volume at rest and during exercise. In addition to these responses, endurance training reduces the resting and sub maximal systolic, diastolic and mean arterial blood pressure.

The cardiovascular control of the heart is modulated by physical activity and thereby the vegetative control of the heart may also be affected. HRV measurements are used to investigate the difference of the sympathetic and parasympathetic activity in sedentary subjects and athletes. Studies involving female athletes remain rarely and further investigations concerning gender differences are needed.

2.5.2 Exercise induced HRV changes

Athletes have lower resting heart rate (HR) compared with sedentary people. This bradycardia leads to longer average RR interval length in athletes [74] because of the enhanced diastole. A lower resting heart rate can be induced by higher vagal activity and/or diminished sympathetic activity. Thus, exercise training is thought to modulate the sympathovagal control of the heart resulting in a predominance of the vagal activity by an increased vagal and/or a decrease sympathetic nervous activity. This modulation is supposed to lead to an enhancement of the vegetative control of the heart which has been quantified by an increased HRV. Selected studies investigating the difference between trained and untrained subjects illustrate the difficulties to prove the relation between bradycardia and increased HRV.

2.5.2.1 HRV in sedentary subjects and athletes

Sacknoff et al. [74] compared the HRV (15 min supine) of aerobic trained (3h/week) with sedentary subjects in the time and the frequency domain. In the time domain, the trained subjects showed significantly increased meanNN, SDNN and pNN50. Therefore a resting bradycardia i.e. lower resting heart rate was noted in athletes whereas the LF, the HF and the total power were significantly lower. The results of the time domain [page 25↓]indicate an increased HRV which could not be proven in the frequency domain. Unfortunately Sacknoff et al. [74] failed to control the respiration of the subjects which could have affected the results of the power spectral analysis of the HRV.

Perini et al. [63] compared parameters of the frequency domain of high endurance trained cyclists and inactive men in sitting position by a short time recording. The cyclists showed similar HF and reduced LF power and thus significantly lower LF/HF ratio than their sedentary counterparts. Similar results were also found by Perini et al. [63] comparing tri-athletes and untrained men in the same way. In neither study was the respiration controlled.

Kouidi et al. [46] investigated the resting HRV (24h recording) of endurance (A), sprint (B) and weight lifting (C) trained athletes compared to sedentary (D) subjects. Parameters of the time domain were evaluated including the HRV triangular index (HRVI), which is the integral of the density distribution. All results had the following ranking, starting with the best results according to the HRV: Group A, group B, group C and finally group D. Therefore, group A showed the most increased meanNN, SDNN and HRVI and the lowest resting heart rate. A relation between the maximal oxygen uptake and the HRVI was only found in group A. Still, athletes showed enhanced HRV compared to their sedentary counterparts. The fact that the endurance trained athletes had the most significantly increased HRV indicates that the exercise training pattern contributes significantly to the modulation of the HRV.

Goldsmith et al. [29] investigated the HRV (24h recording) in aerobically trained and untrained young men in the frequency domain. Athletes had significantly higher HF power during the day, the night and over the entire 24h recording duration. Based on this finding they concluded a substantially greater parasympathetic activity in athletes.

Melanson et al. [57] compared the HRV by short time recordings (10 min) at rest in low (L), moderate (M) and high (H) endurance trained men. The classification of the groups was related to a self reported habitual physical activity level of the subjects. While recording the ECG, volunteers had to match their breathing to an auditory signal (metronome) set to 10 breaths/min. Parameters of the time domain were significantly but similarly increased in M and H compared to L. Ergo, the heart rate was lower in M and H than in L. In the frequency domain, L had lower LF, HF and total power compared with M and H. Still, no difference was found between moderately (M) and highly (H) active men in the HRV. Although the time and frequency domain measures of HRV were greater in active than sedentary (L) individuals, Melanson et al. [57] failed [page 26↓]to demonstrate a dose-dependent manner of HRV with increasing level of physical activity.

Finally, Rennie et al. [70] investigated the effects of moderate and vigorous activity on the HRV (5 min at rest) in more than 3000 British Civil Servants during two years. Subjects were classified into 4 groups in accordance to their metabolic rate based on the self reported activity level: inactive (1), light (2), moderate (3) and vigorous (4) active group. The resting heart rate was lowest in group 4 and highest in group 1 whereas the SDNN was highest in group 4 and lowest in group 1. No gender differences were noted in the time but the frequency domain. LF and HF power were highest in group 4 and lowest in group 1 considering the males whereas the females did not have any differences between groups 1 to 4. Rennie et al. [70] still concluded that vigorous activity is associated with higher HRV. Unfortunately, neither respiration nor menstrual cycle related modulations of the vegetative control of the heart which are known to affect the HRV were controlled.

2.5.3 Training interventions and HRV

Endurance training is assumed to modulate strongest the cardiovascular system compared with other types of training [46]. Nevertheless the exact mechanism induced by physical activity which affects the vegetative control of the heart, remains unknown. Therefore, training interventions of short (5-9 weeks) and long time (3-6 months) duration have been carried out to investigate the direct effect of training on the cardiovascular system and its adaptation due to the autonomic nervous system. Trained and untrained men and women were included in different training intervention programs, which are presented exemplary separated into short and long time interventions, to illustrate the difficulties of the HRV measurements and its physiological interpretation.


[page 27↓]

2.5.3.1  Short time interventions

Pigozzi et al. [64] investigated the effect of 5-weeks exercise training on the autonomic regulation of the heart in untrained women under daily conditions. The sedentary control group was daily active (e.g. housekeeping and cleaning) whereas the intervention group completed a 5-week endurance training of 1h duration 3 times per week. No significant difference was found in the time and in the frequency domain after the training intervention. Therefore, Pigozzi et al. [64] concluded that the training intervention might have been too short to modulate the cardiovascular control of the heart. Besides, nor respiratory neither menstrual cycle related modulations were controlled in this study.

The aim of Uusitalo et al. [85] was to examine the HRV under influence of heavy (70-90% of VO2max and 130% volume) and moderate (<70% VO2max and 5-10% volume) exercise intensity according to the subjects´ individual anaerobic threshold in endurance trained women. The first HRV measurement (20 min supine) took place after 4 weeks and the second after 6-9 weeks of training intervention. The average RR interval length significantly increased and the resting heart rate decreased in the heavy training group whereas the blood pressure remained unaffected in both groups. Nevertheless the SDNN, the LF and the total power significantly decreased despite of increased average RR interval length in the heavy trained group. Uusitalo et al. [85] attributed these findings to a possible sign of impending fatigue due to the heavy training intervention. Additionally, they failed to take account of to control the menstrual cycle and its hormonal fluctuations in athletes which might have been affected by the heavy endurance training intervention in this study.

2.5.3.2 Long time interventions

Catai et al. [13] investigated the effects of a 3-months walking and jogging intervention in young (mean 21 years) and middle-aged (mean 53 years) men on the HRV (24h recording). The average RR interval length was significantly increased and the resting heart rate decreased in both groups after the training intervention. The SDNN, the HF and the total power remained unaffected in both groups whereas only the young men showed a significantly increased LF power. Catai et al. [13] missed to demonstrate a HRV change after 3-month of aerobic exercise training which increased the aerobic capacity in both groups. Thus they explained the increased LF power in young men with the age dependent changing of the autonomic nervous system.


[page 28↓]

Loimaala et al. [51] compared two exercise regimes of varying intensities on HRV and the baroreflex sensitivity (BRS) in middle-aged men. Thereby an inactive control group was compared with exercise 1 (E1), which had to walk or run 4-6 times a week at an intensity of 55% of VO2max, and exercise 2 (E2) group which trained similar but at an intensity of 75% of VO2max. The resting heart rate decreased in E2, but remained similar in E1. No significant changes occurred in the time or the frequency domain measures of the HRV or the BRS in either of the exercise groups. Loimaala et al. [51] finally concluded that their intervention training program despite enhanced VO2max was not able to modify the cardiac vagal outflow in middle-aged men.

Furthermore, Wayne et al. [89] investigated the effect of 6-month endurance training on the HRV (2 min supine) at rest in older and younger men. The training program consisted of walking, jogging and bicycling 4-5 times a week with increasing work-load (from 50 to 85%) according to the subjects’ maximal heart rate. The VO2max increased in both groups after the intervention. Additionally, the resting heart rate significantly decreased whereas the SDNN significantly increased in both groups. No other HRV parameter was evaluated in this study. Still Wayne et al. [89] concluded that exercise training increases the parasympathetic activity and the HRV which was expressed by the lowered resting heart rate and the enhanced SDNN.

2.5.4 Summary

Most studies [13, 46, 51, 57, 70, 74, 85, 89] noted a lower resting heart rate (bradycardia) in athletes and in sedentary individuals after the training interventions whereas two studies [13, 85] additionally noted an increased average RR interval length. These findings are thought to be based on enhanced vagal activity due to the training. Still, uniform HRV results in the frequency domain in relation to physical activity could not be demonstrated because of missing methodological standards which might have affected the results.

Further studies must be in accordance with the guidelines of the Task Force [84], which include the selection of the subjects, training intervention programs and the control of respiration related modulations on the HRV. Additionally, enhanced HRV in athletes has to be quoted not only by the lower resting heart rate and the increased average RR interval but by the SDNN and the LF, HF and total power to describe in detail and to prove completely the training induced effect on the vegetative control of the heart. [page 29↓]Studies including females should also observe hormonal fluctuations and physiological differences in relation to the menstrual cycle.

2.6 Physiological differences of genders

Physiological differences between males and females are based on gender, i.e. sexual characteristics with different internal and external reproductive organs. Gender specific hormonal concentrations, receptor sensitivity and fluctuation pattern result in different modulation of the body. Women are known to have a fluctuation of the endogenous hormones related to the menstrual cycle whereas men do not have any cycle related hormonal fluctuations.

2.6.1 Menstrual cycle

The menstrual cycle starts with the first day of bleeding. The cycle length varies from 21 to 32 days with a mean time of 28 days. The menstrual cycle is a feedback control system of the hypothalamus, the pituitary glands and the ovary. Neurons in hypothalamus produce the hypothalamic releasing hormone (LHRH) known also as gonadotrophic releasing hormone (GnRH). LHRH, a decapeptide, stimulates the secretion of gonadotrophic hormones; the luteinizing hormone (LH) and the follicle-stimulating hormone (FSH). Furthermore FSH stimulates the maturation of several follicles of which only one follicle will get to the end of maturation.

The maturing follicle produces a rising amount of natural estradiol, which consists of estrone, estriol and the predominantly estradiol-17β (E2). Natural estradiol mostly consists of E2 and therefore only E2 is described acting for the natural estradiol.

E2 affects the proliferation of the endometrium. The feedback of E2 to the pituitary glands and the hypothalamus provokes a midcyclic enhanced LHRH production and a stronger sensitivity of hypophysis cells to LHRH.

LH stimulates the ovulation and the luteinizing of follicular granulose cells (corpus luteum). The granulose cells start the production and secretion of progesterone (P). The feedback of E2 and P to the pituitary glands and the hypothalamus reduces the FSH and LH secretion. This results in a reduction of the P level, which leads to start of the luteolisis. At least the menstruation bleeding results from an abrupt P withdrawal (i.e. the withdrawal bleeding) and with it a new menstrual cycle starts again.


[page 30↓]

The following figure (figure 2-5) illustrates the feedback control system and its hormonal interaction including the maturation of the follicle during one menstrual cycle.

Figure 2- 5 : Feedback control system of the hypothalamus, the hypophysis and the ovary (modified from [77])

2.6.2 Basal body temperature

The basal body temperature (BBT) changes rhythmically throughout the menstrual cycle. The BBT rises 0.5 °C up after ovulation and remains elevated during the luteal phase till the onset of the menstrual bleeding. Faber et al. [24] found strong evidence in individual regulation of body temperature at different set points in women. The rate of heat production was positively related to the temperature of the body.Additionally the elevated BBT was associated with the increase of the P level which was observed during the ovulatory cycle and the pregnancy in women. Today an increase of the thermoregulatory set point is suggested to elevate the BBT during the luteal phase in women. Furthermore, Janse de Jorge [34] supposes that all thresholds for all [page 31↓]thermoregulatory effector responses are shifted in a similar direction during the luteal phase.

In animal models the P administration decreased the activity of warm sensitive and increased the activity of cold sensitive neurons in pre-optic area [60]. On the other hand, E2 administration increased the activity of warm sensitive neurons and decreased the body temperature. But there was no indication of a strong specificity between thermo sensitive and steroid-sensitive neurons [79]. So the increase of the thermoregulatory set point is suggested to be related to the ratio between E2 and P in women [12].

2.6.3 Respiratory response

Research on animals has suggested that progesterone increases the ventilation (VE) by means of a central effect in the brain stem. This respiratory response to P is likewise modulated by estrogen (E2). In addition the VE has been shown to be affected by the body temperature [34]. Thus, elevated P level and core temperature during the luteal phase of the menstrual cycle suggest an increased VE as shown by White et al [92].

Supposing a training induced influence, Schoene et al. [78] compared the respiratory response of trained and untrained women at rest and during exercise. The ventilation drive at rest and the exercise ventilation were significantly increased in the luteal phase in both groups. In spite of this increase, no decrease in the exercise performance was noted. The same results were found by Beidleman et al. [3] who compared the VE at rest and during exercise at sea level and at altitude in trained women. The VE at restas well as the exerciseVE were increased with increasing P level during the menstrual cycle. Nevertheless, the menstrual cycle could not affect the maximal and the sub maximal exercise performance at sea level as well as at altitude in women.

Menstrual cycle induced influences on the respiratory response were also thought to affect the balance between the CO2 controlling system and the CO2 exchange. Therefore, Dutton et al. [22] did a CO2 rebreathing test based on Rebuck’s outlines [69] at different phases of the menstrual cycle. They noted that the sensitivity of the respiratory centre to CO2 was significantly increased in the luteal compared to follicular phase; still a significant decrease of the CO2 threshold was not found. The same results concerning the CO2 sensitivity were found by Schoene et al. [78].


[page 32↓]

2.6.4  Hormonal fluctuation in men

Men have a feedback control system consisting of the hypothalamus, the anterior lobe of pituitary gland and testicles. The LHRH neurons of hypothalamus secrete a decapeptide into the portal system every 2 to 4 hours. The LHRH pulses as well as the LH and FSH pulses mostly occur at night and in the early morning hours.

LH stimulates the Leydig cells, which results in an augmentation of the androgen production. The Leydig cells give feedback to the hypothalamus and the anterior lobe of pituitary gland, which closes the feedback control loop. Testosterone (T) of which only soluble T in plasma has a biological effect, is the most important testicular androgen. It reaches the anterior lobe of pituitary gland and the central nervous system via the bloodstream and has a negative feedback to the LH and LHRH production. Moreover, T is coupled to the protein sexual hormone binding globulin (SHBG).

In men, the influences of FSH as well as the intra testicular androgen result in the spermatogenesis. Androgens are additionally known to stimulate the protein synthesis, the bone and muscle growth and the masculine hair growth. Men have diurnal fluctuation of the hormones whereas the LH and T production depends on sleeping habits, and not on the time of day.

The BBT of men only shows small fluctuation and is more or less stable throughout the month.


[page 33↓]

2.7  Effect of menstrual cycle on HRV

The menstrual cycle is characterized by phases with different endogenous hormones and consequently by different neurotransmitter concentrations. The cyclic changes in estradiol and progesterone levels modulate physiological functions including e.g. the basal body temperature and the respiration of women. However the relation between the menstrual cycle and the vegetative control of the heart remains disputable due to the lack of studies. Four studies exist which investigated the HRV in relation to endogenous hormonal fluctuations in sedentary women [30, 50, 76, 97] whereas one study [75] investigated the reflex control of autonomic function induced by an orthostatic test during the menstrual cycle. These studies are described detailed in the following chapter. The menstruation phases are verified in days or described in phases according to the authors’ data to compare the study intervention phases/days. Therefore the first day of bleeding means the first day of the menstrual cycle, e.g. day 5 stands for the 5th day of the menstrual cycle whereas the cycle length is generally between 28-32 days.

2.7.1 HRV in relation to the menstrual cycle

Guasti et al. [30] compared the HRV (25 min supine) of the follicular (day 5±1) with the luteal phase (day 23±3) in 13 women. The blood pressure and the heart rate remained similar throughout the menstrual cycle. In the frequency domain, the LF and the HF power were not significantly different whereas the LFnu was significantly increased and the HFnu decreased in the luteal phase. Consequently, the LF/HF ratio was significantly increased in the luteal compared to the follicular phase. Guasti et al. [30] suggested an increased sympathetic activity in the luteal phase.

Yildirir et al. [97] investigated the HRV (5 min supine) during the follicular (day 11±1) and luteal phase (day 21±2) in 43 women. The heart rate, the LF and HF power were similar in both phases. A significant increase was only noted in the LFnu and consequently in the LF/HF ratio in the luteal phase whereas the HFnu was not significantly different. Based on these findings, Yildirir et al. [97] still concluded that the sympathetic activity was enhanced in the luteal compared with the follicular phase.

Sato et al. [76] compared the HRV (20 min supine, 10 min sitting) of the follicular (day 7-10) and the luteal phase (3-7 days prior to the next bleeding) at 3 consecutive days in 20 females. Each day of the study was repeated in the following month and thus, the [page 34↓]study lasted for two menstrual cycles. The resting heart rate and the blood pressure remained similar. Still, the LFnu significantly increased, the HFnu decreased and the LF/HF ratio consequently increased in the luteal compared with the follicular phase. The absolute values of the LF and HF were not mentioned. Sato et al. [76] suggested a predominant sympathetic activity in the luteal phase.

Saeki et al. [75] investigated the autonomic reflex control by an orthostatic test (5 min supine, 5 min sitting) in 5 different phases which are abbreviated by the first letter of the phase: menstruation (day 1-3), follicular (between M and O), ovulation (3 prior and 4 days after O), luteal (between O and P) and the premenstrual phase (7 days prior to next bleeding). Subjects´ (n=10) breathing frequency was controlled by a metronome (15 breaths/min) during the orthostatic test. In supine and sitting position, the HF power was significantly higher in F than in M. The LF/HF ratio was similar in both positions whereas the LF power was not calculated. The LF/HF ratio increased significantly by changing the position from supine to sitting in M, F and P. Saeki et al. [75] concluded an increased vagal activity in F at rest and an enhanced sympathetic activity in M, F and P during the orthostatic test. Based on these findings they supposed a different modulation of the vegetative control of the heart at rest and at the orthostatic reflex control during the menstrual cycle.

At least, Leicht et al. [50] examined the HRV (20 min supine) of 10 women in 3 different phases M (day 3.8±0.5), O (day 15.8±0.7) and L (day 22.1±0.4). The resting heart rate was significantly greater at O compared with M and L whereas the LF, HF and total power as well as the LFnu, HFnu and the LF/HF ratio remained similar. Leicht et al. [50] concluded that there was no association between the cyclic variation of the endogenous hormones and the cardiac autonomic control.

In summary, Guasti et al. [30], Sato et al. [76], and Yildirir et al. [97] suggested an enhanced sympathetic activity in the luteal compared with the follicular phase. Their findings are based on the LFnu, HFnu and the LF/HF ratio whereas the absolute power values of LF and HF were not evaluated [76] or remained unaffected [30, 97] during the menstrual cycle. In accordance to the guidelines of the Task Force [84] the normalized units have to be quoted with the absolute values of LF and HF power in order to describe completely the distribution of power in spectral components. However the authors [30, 76, 97] did not follow these guidelines. Saeki et al. [75] concluded an enhanced vagal activity in the follicular phase at rest and an increased sympathetic [page 35↓]activity during the orthostatic test in M, F and P based on the results of the HF power and the LF/HF ratio. They supposed an enhanced sympathetic activity in M, F and P due to an increased LH/HF ratio, but they failed to validate these findings by using the absolute power values of LF as demanded by the Task Force [84].

Solely Leicht et al. [50] could not find any modulations of the vegetative control of the heart in relation to the menstrual cycle or in the time or the frequency domain of the HRV. Based on this finding and due to the shortcomings of the above mentioned studies, further investigation is needed to clarify the cardiac autonomic control during the menstrual cycle in women.

Further studies on female subjects should only be done in consideration of special guidelines, which are not yet been laid down. Until then, at least the following points should be considered in HRV studies in relation to the menstrual cycle:

  1. Monitoring of the menstrual cycle by hormonal analysis.
  2. HRV measurements in accordance to the guidelines of the Task Force [84],which include the respiration control.
  3. Control of the mental changes, e.g. individual symptoms and feelings during the menstrual cycle with a daily questionnaire.


[page 36↓]

2.8  Mood state control

2.8.1 Mood state during the menstrual cycle

During the menstrual cycle women mentioned individual different symptoms and feelings, which might affect the mood state. The changes of the cardiovascular control could be modulated by mental fluctuations in women. Therefore, the profile of mood state (POMS) might be a useful tool to investigate the subjects’ well-being during the intervention days of the study.

2.8.2 Profile of mood state (POMS)

Profile of mood states (POMS) is a self measured confidential scale which records individual change in mood states. The original version created by McNair et al. [56] consists of 65 items; the short version of 35 items. Items are adjectives characterizing the mood state in an answer scale of 5 to 7 possibilities, ranging from “absolutely not” to “very strong”. The original scale was classified into six and the short scale into the following four subscales: depression, fatigue, vigor and anger. Each subscale includes certain adjectives expressing the mood state. These adjectives are awarded points based on the value of the subscale on which the evaluation of POMS is based. Reliability of POMS was proved by a coefficient of internal consistency.

The mood response to exercise has been investigated. Agreement exists that exercise has a positive influence of the individual mood state. But the effects of maximal exercise differ from sub maximal exercise in sports. Differences are also related to the level of physical fitness in human beings [82]. The main findings of Pronk et al. [67] indicated acute increase of fatigue and depression as well as a decrease of anger and vigor after maximal exercise. This implies that the mental well-being of individuals can be influenced by physical activity. Differences between men and women are suggested because of the menstrual cycle.

In addition to this, the POMS questionnaire is considered as a valuable tool to predict overtraining and staleness in elite sports. Coker et al. [15] compared female starters and non-starters of a softball team prior to playing. Significant mean differences were found between starters and non-starters on constructs of anger, confusion, tension and depression; prior to the play non-starters presented higher fatigue. The author suggested [page 37↓]that in spite of the same training conditions the non-starters did not share the same psychological profile as their starting peers.

The mood state measured by POMS and the resting salivary Cortisol levels were examined in 14 female college swimmers in 1989 by O’Connor et al.[61]. Significant alterations in tension, depression, anger, vigor, fatigue and global mood across the training season were noted in swimmers compared to controls. A correlation between salivary Cortisol level and a depressed mood during overtraining could be found. Swimmers classified as stale had significantly higher global mood expressed by the overall score, enhanced depression and increased salivary Cortisol levels than swimmers without performance decrements. The valence of salivary Cortisol is disputed and thus these findings have to be considered with care.

Morgan et al. [59] collected data of mood states during a ten year research effort in 400 male and female competitive swimmers. The results indicated that mood disturbances increased in a dose response manner as training stimulus increased. These mood disturbances fell back to baseline levels with a reduction of the training load.

The study of Raglin et al. [68] compared successful and unsuccessful women’s rowing teams. In baseline mood state, no difference was noted and mood disturbances increased in both groups during the training season. At the end of the season, the mood of the successful rowers returned to baseline whereas the mood of the unsuccessful females still showed significantly elevated mood disturbances.

Williams et al. [94] noted in moderately trained runners that more economical values of running were associated with more positive mental health profiles. The graphical representation of the POMS scores resulted in a mountain profile or in an inverted one; thereby the expression Iceberg profile was used to mention the positive POMS results (less negative feelings, better mood state), whereas the inverted Iceberg profiles illustrated the negative POMS scores (enhanced negative feelings, worse mood state) [59]. Soccer players also presented iceberg profiles during successful performance and with a high percentage of winning in a study of Filaire et al. [25] whereas decreased performance resulted in a decrease in vigor and an increase in tension and depression.


[page 38↓]

Based on these findings the notion is, that successful elite athletes tended to produce so-called iceberg profiles of mood but less successful ones not. Thereto mood could be considered as an emotional state influenced by personality and environmental factors [14].


[page 39↓]

2.9  Derivation of the question

The measurements of the HRV, which are non-invasive and easily handled, could be a valuable tool for sports medicine, because of the insight into the autonomic nervous system (ANS) and its status. Training causes short and long term disturbances in the ANS, which are rarely described. The physiological reactions to different training interventions, which cause autonomic disturbances and acute or chronically adaptations could be investigated by the HRV. Thereby the diagnoses of approaching overstrain and overtraining in athletes is of interest which could be investigated using the orthostatic test, i.e. reflex control of the ANS. Still data of athletes, particularly female athletes, are inconsistent and the physiological significance of the HRV is not yet practicable in relation to training interventions. Due to this, the first purpose of this study was to investigate whether the short time recording of the HRV including the orthostatic test could be a valuable tool for male and female athletes who were involved in individual training patterns during the study. The guidelines of the Task Force [84] were observed in this study.

The influence of the hormonal fluctuations (P, E2, LH and FSH) during the menstrual cycle on the vegetative control of the heart has been investigated in several studies whereas the difference between sedentary and physically active women, i.e. highly endurance trained females is missing. Acute or chronic adaptations induced by training could affect the modulation of estrogen and progesterone on the HRV. Such training induced adaptations do not have to go along with menstrual cycle disturbances and dysfunctions. Female athletes may have a normal ovulatory cycle. Based on this, the second aim of this study was to investigate the HRV at rest and during the orthostatic test in relation to the fluctuations of P, E2, LH and FSH in course of the menstrual cycle in trained (i.e. highly endurance trained) and untrained (i.e. sedentary or moderately active) women.

Progesterone is known to affect the breathing during the menstrual cycle in sedentary women. The oxygen uptake, the respiratory frequency and the tidal volume are individually modulated by the progesterone fluctuations, but data concerning female athletes are missing. The respiratory frequency and the tidal volume are known to affect the HRV by afferent reflexes in the baroreceptors and the pulmonary stretch receptors. Ergo the modulated breathing is expected to affect the HRV in the second part of the [page 40↓]menstrual cycle with increasing progesterone level. However the respiratory related modulations of the HRV in endurance trained women has not yet been investigated. The investigation of the respiration on the HRV requires spontaneous breathing instead of metronome controlled breathing. Therefore the fourth aim of the study was to investigate the influence of the spontaneous breathing which might be modulated by progesterone on the HRV in trained and untrained women.

Men do not have any cycle related hormonal fluctuations but individually different levels of testosterone. Therefore, the influence of testosterone especially free testosterone on the vegetative control of the heart was investigated by the HRV in athletes and sedentary men. Additionally, male athletes served as controls for the athletic women and untrained men for the sedentary females. Finally the repeated HRV measurements in the time and frequency domain in the male groups served as a test for the reliability of our study.

Furthermore, the blood glucose and the insulin concentration are supposed to have an influence on the vegetative control of the heart based on data of diabetic patients. The relation of the blood glucose and the insulin on the HRV in healthy and/or trained individuals as well as gender differences are missing. Thus, an additional aim of this study was to investigate the relation between the HRV and the metabolic supply by the blood glucose and insulin concentration in trained and untrained men and women.

2.9.1 Control parameters

Due to the high sensitivity of the ANS to internal and external influences, several parameters were controlled during the study i.e. at each study day.

The conduction and the impulse transmission of the nerves and the muscle cells depend on the homeostasis of sodium (Na+), potassium (K+), magnesium (Mg²+), calcium (Ca²+) and chloride (Cl-) concentration. An affected homeostasis of the electrolytes may lead to an affection of the conduction system of the heart which would modulate the HRV. Due to this, the electrolytes Na+, K+, Mg²+, Ca²+ and Cl- were controlled in the blood serum during the study.

Haemoglobin concentration was controlled because of anaemia which is often diagnosed in highly endurance trained athletes whereas the hematocrit level served to ensure the balance of the fluid.

At least, the POMS questionnaire was filled in daily by the subjects to evaluate the mood state during the study which might have been affected by the menstrual cycle, the [page 41↓]individual training interventions or other influences. Additionally the alcohol uptake, the basal body temperature, the resting heart rate, the sleep quality and the training (duration, intensity and type) were noted daily by the volunteers.


© 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