ISSN: 1223-1533

< Return to the list of articles


Authors: V. Moga, Sabine Jahraus, Flavian Parge, Mariana Moga, Ioana Cotet, Rodica Avram

Received for publication: 15th of June, 2013
Revised: 5th of July, 2013

Read the FULL article in .PDF format

SUMMARY: (Hide the summary)

Heart rate variability  (HRV) represents the widely  used method for the assessment of the autonomic tone. Time domain parameters, like standard deviation of all RR intervals and spectral analysis, expressed as the spectral power density (PSD) in both low frequency (LF: 0.05 – 0.15Hz) and high frequency ( HF: 0.15 – 0.50 Hz) are the most frequent parameters used in clinical  practice.  Low HRV parameters in heart failure and myocardial infarction are correlated with the risk of a high incidence for ventricular arrhythmias and sudden cardiac death.The prognostic  impact of low HRV parameters is well studied in numerous studies. Neurally mediated syncope represents one of the most frequent condition for admission and hospitalization. The aim of our study was to assess the autonomic imbalance during head-up tilt test in patients that had syncope and to compare it with healthy subjects. Spectral analysis  of the ECG signals was performed using the autoregressive analysis.

Key Words:

Syncope, autonomic imbalance, healthy subjects, tilt test.




Syncope is a symptom  characterized by a period of transient  loss of consciousness(T-LOC) that is brief in duration, self-limited,  and due to a spontaneously reversible inadequacy of cerebral nutrient flow. The most common responsible factor is cerebral hypoperfusion due to transient  hypotension. Syncope has many possible precipitating  causes. However, the principal etiologies may be  classified   into  three categories:  (1) reflex (neurally mediated) faints, (2) orthostatic faints, and (3) cardiac (cardiovascular) faints [1].

Neurally mediated syncope is a disorder  of the autonomic regulation of postural tone,which  results in hypotension, bradycardia, and loss of consciousness.

A wide variety of stimuli can trigger this reflex, the most  common stimulus being orthostatic stress. Typically,  a patient with neurally mediated syncope experiences nausea, lightheadedness, a feeling of warmth, and pallor before abruptly losing consciousness [2]. Syncope is a sudden loss of postural tone followed by rapid, spontaneous recovery.It affects all ages, from the pediatric to the elderly [2]. The human  body  has the capacity to maintain a stable blood pressure and to react at various conditions that change this steady state. Main factors are postural changes, and the main mechanisms are reflex mediated. Prolonged stand-up is followed by a shifting of blood from the thorax to the abdomen and lower extremities.   This   shift  in   blood volumecan markedly decrease the cardiac output, followed by loss of postural tone and faint.

The pathogenetic  mechanisms  of syncope  are still poorly understood. It is emphasized that the stimulation of ventricular vagal receptors would be produced by an enhancement of heart contractility  due to an exaggerated cardiac sympathetic  activation.  The possibility  that an initial cardiac sympathetic overactivity might promote vasovagal   reactions is  supported   by  the   clinical observation  of a  transient   rise in heart  rate before syncope [6]. This increase of the resting heart rate, leads to the hypothesis of an autonomic imbalance in syncope.

Spectral analysis of heart rate variability offers the opportunity to  assess the contribution of both sympathetic  and parasympathetic  components  in the modulation of the regulation mechanisms during neurally mediated syncope.

Numerous previous studies have observed different types of modulation  by the sympathetic  tone during syncope. It was noticed a heterogeneity of changes from increased sympathetic activity to unchanged contribution compared to basal conditions.

Heart rate variability, quantifies the contribution of the spectral components in the modulation of the regulation mechanisms involved in the pathogenesis of syncope. Very low frequency (VLF: 0.01 -0.05 Hz), low frequency (LF: 0.05 -0.15 Hz) and high frequency (HF: 0.15 - 0.50 Hz) are the main components of the spectral analysis. But two major oscillatory components can beidentified in the power spectrum of RR variability:  a high-frequency (HF) component at ‘0.25  Hz, considered  a marker  of vagal modulation of the sinoatrial node, and a low-frequency (LF)  component   at ‘0.10 Hz,  considered,   when normalized, a marker of sympathetic modulation [4].

Very low frequencies are modulated by thermoregulation and other complex mechanisms described in other studies (fig. 1). The guidelines for heart rate variability have been established in the mid 90’s, as emphasized by the Task Force of The European Society of Cardiology and The North American  Society  of Pacing and Electrophysiology [7].

Various  spectral  methods  have  been  used  for the analysis  of the HRV, and  they may be  classified  as nonparametric (fast Fourier transform) and parametric.

The main advantage of parametric methods such as autoregressive (AR) method is that the smoother spectral components can be distinguished independently of the selected frequency bands. Consequently, they allow an accurate  estimation  of power spectral  density  with automatic calculation of  the  LF  and   HF  power components [6].



The aim of our study was to highlight the complex heart rate modulation in patients with syncope and to analyze the behavior  of RR intervals dynamics compared to young healthy subjects.

The study was performed at the Cardiology Clinic of the Emergency County Hospital Timis, University of Medicine  “Victor Babes”   Timisoara, Romania. The patients group ( 12 men, 8 women) with mean age 57.4 years, described   syncope   or  pre-syncope  sympto- matology. A control group consisted of 12 young health subjects (5 men and 7 women) with mean age of 30.3 years was used to compare data and to record data in a healthy group. All subjects have been in sinus rhythm.

Subjects were placed on an electrically driven tilt table,  and the ECG was monitored by a Cardiax V 3.50.4 ECG system by IMED Co Ltd, Hungary.



Click on the image to zoom in


Figure  1. Tachogram of 256 selected  R-R interval  time series after detrending  and HRV representations of non parametric (FFT) and parametric spectra (AR) for one representative subject during tilt test. RR, R-R interval; PSD, power spectral density.


During the procedure subjects were been asked to breathe at a constant rate of 15 breathe/min (0.25 Hz). After 15 minutes in the supine position, the table was rotated for 15 minutes to a 900 up-right position that was maintained for 30 minutes. None of the healthy subjects experienced syncope or any symptoms.

ECG signals  and RR intervals  were  recorded  and the spectral  analysis  was performed  using fast Fourier transform  (FFT) and autoregressive analysis (AR).

The R-R interval  series corresponding  to supine and standing position were selected, respectively, just before and just after the change of position.  The FFT and AR spectra were then calculated from this interpolated and detrended 512 seconds window  width  R-R interval with HRV Analysis Software 1.1 for Windows (The Biomedical Signal Analysis  Group, Department of Applied Physics University of Kuopio, Finland.

Two frequency bands were  considered: (1) LF band from 0.045 to 0.15 Hz and (2) HF band from 0.15 to 0.4 Hz ( fig. 1).

The AR  spectrum was calculated by fitting a 32th-order AR model, into  the R-R data  [6]. Autoregressive analysis furnished both the power and the central frequency of the oscillatory components. The LF and HF components  were defined as those with central  frequencies within  band limits. Central frequencies were around 0.1 Hz and 0.25 Hz for the LF and HF components,  respectively  [4, 6]. Low-frequency and HF oscillatory  components  arepresented in absolute (square milliseconds) units.


Statistical analysis

For the statistical  analysis we have used Graph Pad Prism. All numeric variables were expressed as mean and   the  statistical analysis was performed  using Student’s  t-test and  correlation  analysis  by Pearson method. A p value < 0.05 was considered statistically significant.



The resting mean heart rate in the syncope patients group was 67 b/min compared to the healthy subjects, 76 beats/min (p: 0.01). During the 900up-right position the mean  heart  rate in the syncope  patients  group was 82b/min vs. 92 b/min in the young healthy group (p: 0.05). Statistically  significant  differences of the mean heart rate have been measured also between  resting conditions and the 900 up-right position in both groups (table 1). The  mean  RR  interval   (ms.) didn’t show significant differences at 900 up-right position. The main resting parameters in resting conditions are summarized in table 1.



Click on the image to zoom in


Table 1. Clinical data in both study groups during resting conditions.



Click on the image to zoom in


Fig. 2. Spectral analysis of heart rate variability at 90o. Cardiax v.3.50.4



Click on the image to zoom in


Table 2. Heart rate  variability  parameters  during  the  90o up-right position

SDNN – standard deviation  of all RR intervals,

LF/HF – autonomic index




The head up tilt test and the 90o up-right position has induced changes in both clinical data and in the spectral analysis parameters of heart variability in the syncope patients group but also in the healthy subjects. In the syncope group the 900 up-right position is followed by an important increase of the mean heart rate, 67.6 b/min vs 82.3 b/min ( p: 0.0002). This sympathetic over activity is also noticed in the healthy group even if not with the same magnitude, 76.2 b/min vs 92.8 b/min.

The spectral analysis of heart rate variability in both study groups, reflects the intensity of the sympathetic activity during the 90o up-right position, and the reduced vagal modulation of the heart rate. All this aspects are observed when the measurements are made in absolute (square milliseconds) units or reflected by the LF/HF ratio as a measure of the autonomic imbalance during head-up tilt test (Fig.2).

The changes induced by 90o up-right position on   the heart rate variability parameters, in time domain and in frequency  domain are summarized in table 2. The spectral pattern of the heart  ratevariability are reflected in fig. 2.

The main spectral  power  is in the 0.01 – 0.15 Hz, reflecting a high sympathetic tone.

The AR analysis in some cases can be performed using higher model order to obtain a better representation of the spectral components of heart rate variability. In the case above, we have used an AR model of 30.



High  resting heart rate and an increase of the sympathetic tone at 900 up-right position characterize the pattern of healthy subjects during the head-up tilt test, even  in free  of symptomatology  subjects.  This behavior  reflects the autonomic   modulation   during postural changes.

In syncope patients, a lower mean heart rate in resting conditions will prove a significant increase of the values during head-up tilt test. In both situations we deal with a reduced vagal activity  and a higher sympathetic tone activity.  These changes are reflected independently of the method that is  used   to  assess   the spectral components  of heart  rate variability.  Fast  Fourier transform  or autoregressive  analysis   reflects these aspects.

The specificity and sensitivity of the HUTT test are hard to determine because of methodologic differences in the  test’s performance. The  absence of a “gold standard”  also makes it difficult to determine  normal versus abnormal results.  Nevertheless, with “normal” volunteers and with patients who have a history typical of NMS, the reported specificity  is about  90%, and the reported sensitivity  ranges from 32% to 85% [ 6 ]. Other studies [ 5 ] highlight the fact that the main effect of tilt test seems to be the reduction of the vagal activity rather the high sympathetic activity.

Our  study is a focus  in the  mechanisms  that are involved in the neutrally mediated syncope and highlight the contribution of the autonomic tone in both syncope patients  and healthy subjects.  Further studies  will be focused on the correlation between heart rate and blood pressure during neutrally mediated syncope.




  1. Ali Massumi et all - Tex Heart Institute J 2000;27:268-72
  2. Michele Brignole ·David G. Benditt SyncopeAn Evidence-Based Approach – Springer Verlag 2011
  3. Moya A, et al. Guidelines for the diagnosis and management of syncope (version 2009). The task force for the diagnosis and management of syncope of the European Society of Cardiology (ESC). Eur Heart J. 2009;30:2631–2671.
  4. Raffaello Furlan et all - Cardiac Autonomic Patterns Preceding Occasional Vasovagal Reactions in Healthy Humans - Circulation. 1998;98:1756-1761
  5. Fabio Badilini et all – Heart Rate Variability  in Passive Tilt Test: Comparative Evaluation of Autoregressive and FFT Spectral Analysis - Pace Vol. 21, 1998
  6. Aurelien Pichon, et all -  Spectral analysis of heart rate variability: interchangeability between autoregressive analysis and fast Fourier transform- Journal of Electrocardiology 39 (2006) 31– 37
  7. Guidelines: Heart rate variability, Standards of measurement, physiological interpretation, and clinical use. Task Force of The European Society of Cardiology and The North American Society of Pacing and Electrophysiology. European Heart Journal (1996) 17, 354–381.

Correspondence to:
Victor Moga, MD, PhD - Clinic of Cardiology, Timișoara County Hospital, 10 Iosif Bulbuca St., 30073