Spectral analysis of heart rate in vasovagal syncope
Research P a p e r
SPECTRAL analysis of heart rate fluctuations was used to
investigate the role of the autonomic nervous system in
the pathogenesis of vasovagal syncope. N i n e adolescents
with a history of at least three episodes of vasovagal
syncope and nine age-matched healthy controls were
studied. All subjects were tested in supine position and at
a 60 ° inclination for 60 min or less if syncope developed.
Blood pressure and heart rate were measured, while the
ECG and respiration traces were recorded on magnetic
tape for later spectral analysis.
Baseline heart rate was lower in control subjects than in
patients, increased with tilt in both groups, and remained
lower in the control subjects throughout the experiment.
Baseline systolic and diastolic blood pressure was similar
in both groups. Diastolic blood pressure initially increased
with tilt in all subjects and decreased significantly
thereafter in patients. Pulse pressure was lower in patients
throughout the experiment.
The heart rate power spectra displayed a higher baseline
level of low frequency fluctuations in the control group.
The high frequency fluctuations component was similar
in all subjects. The results of the test, regarding
haemodynamic parameters and autonomic control of the
heart rate, as expressed by low and high frequency
fluctuations, are consistent with a reduced sympathetic
reserve in the individuals with previous episodes
of syncope.
Key words: Autonomic nervous system, Fainting, Heart rate
fluctuations, Spectral analysis, Tilt, Vasovagal syncope
Introduction
Sudden transient loss of consciousness with
loss of postural tone is defined as syncope. It is a
common clinical problem with various possible
causes and has been classified as cardiovascular
(reflex and cardiac), non-cardiovascular (neuro-
logic, metabolic) and unexplained syncope. 1 The
vasovagal reflex type of syncope is commonly seen
in healthy adolescent and young persons. The
pathophysiology of the event is primarily right
heart underfilling due to peripheral pooling during
upright position. Reflex cardiovascular syncope
occurs when marked vasodilatation develops with
systemic beta-adrenergic stimulationJ '2 A para-
doxical bradycardia usually accompanies this
vasodilatation and it is commonly accepted that this
bradycardia is of vagal origin. 3'4 It appears that the
syncopal event is due to an alteration in the
cardiovascular control. <2'4-7
Short term cardiovascular control involves
cooperation of different limbs of the autonomic
nervous system that is aimed to keep an adequate
cardiac output when the blood pressure is perturbed
by a change in position. Cardiovascular homeostasis
is achieved by the dynamic response of the
cardiovascular control system to the various
© Rapid Communications of Oxford Ltd.
Clinical Autonomic Research3, 261-269 (1993)
Spectral analysis of heart rate
in vasovagal syncope" the
autonomic nervous system in
vasovagal syncope
A. B a h a r a v , MD, M . M i m o u n i , MD,
T. L e h r m a n - S a g i e , MD, S. Izraeli, MD and
S. A k s e l r o d , PhD 1"cA
Department of Pediatrics, Beilinson Medical Center,
Petach Tiqua; 1Department of Medical Physics, School
of Physics and Astronomy, Sackler Faculty of Exact
Sciences, TeI-Aviv University, Ramat-Aviv, Israel.
CACorresponding Author
perturbations in haemodynamic parameters; this
response includes an increase in the heart rate as
well as in the sympathetic vasomotor tone and in
the total peripheral resistance. The analysis of
beat-to-beat fluctuations in heart rate and/or blood
pressure may highlight the autonomic cardio-
vascular control mechanisms. 8-15
Some authors utilized spectral analysis methods
to analyse the fluctuations in haemodynamic
parameters and to achieve hereby better under-
standing of the cardiovascular control mechanisms.
Their results indicate that short term (seconds to
minutes) fluctuations in these parameters are
concentrated in two principal spectral regions: a
lower frequency range, below 0.15 Hz, which has
been often subdivided into a low (LFP; 0.02-0.09
Hz) and a mid frequency range (MFP; 0.09-0.15
Hz), and a high frequency range (HFP; around
respiratory frequency--usually between 0.2 Hz and
0.4 Hz). 9'10 From investigations of the principal
branches of the cardiovascular control and their
contribution to the power spectrum under various
pharmacological interventions, it appears that, in
man, the high frequency respiratory peak is mainly
mediated by the parasympathetic system. Fluctua-
tions due to both vasomotor and baroreceptor
control express themselves in the lower frequency
Clinical Autonomic Research.vol 3"1993 261
A. Baharav et al.
band. These lower frequency heart rate fluctuations
are both beta-sympathetically and parasympathetic-
ally mediated in the supine position, while the
sympathetic contribution increases in the upright
position, 1°'14 shifting the sympathovagal balance.
Prolonged head-up tilt is currently being used as
an aid to detect the propensity to vasovagal reaction
in patients with recurrent episodes of unexplained
syncope, ls-2° It is applied in the present study in
combination with spectral analysis of heart rate
fluctuations to obtain a better understanding of the
role of the autonomic nervous system in the
pathogenesis of vasovagal syncope. This clinical
phenomenon is of great interest as it is thought to
be the result of a certain 'autonomic imbalance' in
patients who suffer recurrent episodes of 'vasova-
gal' syncope.
Methods
Syncope was defined as a sudden transient loss
of consciousness with inability to maintain postural
tone, which was not compatible with cardiac,
metabolic or convulsive-neurologic disorder. The
fainting events were in upright position and
patients remembered the whole episode.
Subjects." Eleven healthy individuals aged 9-16 years
(mean + SD, 13.5 ___2.3) with a previous history of
at least three episodes of unexplained syncope,
clinically compatible with vasovagal reflex fainting,
were enrolled in the study, in agreement with
criteria by Sheldon. 21 All episodes accounted for
inclusion criteria were due to orthostatic in-
tolerance. Ten healthy youngsters aged 10-16.5
years (mean ___ SD, 12.4 -t- 1.6) with no history of
syncope were enrolled in a control group. Two
subjects, one from each group, were excluded
before data analysis due to technical problems in
the ECG recording. One teenage girl from the first
group was also excluded because of an evident
episode of hyperventilation fainting with no change
in heart rate or blood pressure during the episode.
The study was approved by the Committee for
Human Experiments at the Beilinson Medical
Center. All subjects and their parents gave their
informed consent before enrolment.
Basic evaluation: Detailed history was obtained from
all subjects and their parents. All subjects were
submitted to complete physical and neurological
examination. The blood pressure was measured
during supine and standing positions. All subjects
had their haemoglobin and blood glucose levels
tested. A twelve-lead ECG was also performed
before the test. All of the above were normal. Most
of the subjects in the patient group were submitted
to some more specific investigations as part of
previous unrelated work-up of the fainting
262 Clinical Autonomic Research. vol 3 . 1 9 9 3
episodes: 24 h Holter (six subjects); echocardio-
graphy (six subjects); fundus examination (one
subject); E E G (fiVe subjects); controlled physical
exercise under continuous ECG recording (ergo-
metry) (one subject); brain computerized tomo-
graphy (one subject), see Table 1. All of the above
were normal. No patient was submitted to
electrophysiological studies.
Study protocol: All subjects were tested in the late
morning hours, neither in a postprandial nor in a
fasting state. The test began only after a resting
period of 45min. The subject lay down on a
horizontal tilt-table, the inclination of which could
be smoothly and rapidly changed to any pre-
determined angle. All patients were supported by
belts at knees and waist, and the table had a foot
plate. The two-lead ECG trace was continuously
monitored with an Horizon 2000 (Mennen) monitor
and recorded from standard lead II on an FM tape
recorder (TEAC). Respiration was monitored and
recorded with the same instruments. Blood pressure
was measured by cuff sphygmomanometer at 10 min
intervals while supine, and at 30 s, 3 min, 5 min and
every other 5 min during head-up tilt position.
Additional measurements of the blood pressure
were performed if the subject complained of any
symptoms compatible with pre-syncope or syncope
or if signs of imminent fainting developed.
For each subject, following the 45 min of supine
rest, we recorded 20 min in supine position. Then
the tilt-table was inclined to 60 ° head-up tilt
position during a time interval of 10 to 15 s. The
subjects were kept in this position for 60 min or less
if they developed symptoms of syncope (a drop in
systolic blood pressure of at least 20 mmHg and/or
a drop in heart rate below 50 beats per minute, or
subjective feeling of imminent fainting episode)
before the end of this time period. At that moment
the table was rapidly returned to the horizontal
position.
Heart rate spectral ana~sis: We analysed data from
256 s long traces taken sequentially during the last
15 min in supine position and throughout the entire
period in head-up tilt position. We chose to focus
on the following epochs:
(0) During the supine position prior to the change
in inclination.
(1) Immediately after the change in inclination.
(2) 20 rain after the tilt or the 256 s interval just
before syncope, if this took place prior to the end
of the first 20 min period in the inclined position.
In this instance we focused on only two epochs in
tilt position instead of three.
(3) 55 rain after the tilt or the 256 s interval just
prior to the syncope, if this happened before the
end of the test time.