The vasovagal response

575
Clinical Science (1991) 81,575-586
Editorial Review
The vasovagal response
JOHANNES J.
VAN
LIESHOUT’.*, WOUTER WIELING‘, JOHN M. KAREMAKER3 AND
DWAIN L. ECKE3ERG4*5
Departments of ‘Medicine, 21ntensiveCare and 3Physiology,Academic Medical Centre, University of Amsterdam, The Netherlands,
and Departments of 4Medicine and 5Physiology,Hunter Holmes McGuire Department of Veterans, Affairs Medical Center and Medical
College of Virginia, Richmond, Virginia, U.S.A.
INTRODUCTION
PHYSIOLOGY OF THE VASOVAGAL RESPONSE
The vasovagal response is the development of arteriolar
dilatation (‘vaso’) and inappropriate cardiac slowing
(‘vagal’)leading to arterial hypotension with loss of consciousness [ l , 21. The syndrome is called vasovagal
(vasodepressor) syncope or fainting. Although it may
occur in response to a relative or absolute loss of blood, it
also may develop in response to strong emotions. The
surgeon John Hunter (1728-1793) wrote [3]: “I bled a
lady but she fainted and while she continued in the fit the
colour of the blood that came from the vein was a fine
scarlet. The circulation was then very languid”. We now
think that Hunter observed the effects of vasodilatation
during fainting: the scarlet appearance of the blood can be
ascribed to the small arteriovenous oxygen difference [4].
In 1895, Leonard Hill [ 5 ] suggested that emotional fainting results from withdrawal of vasomotor neural traffic.
In 1932, Thomas Lewis introduced the term ‘vasovagal’
to stress that both blood vessels and the heart are
involved. He showed that the abrupt slowing of the heart
rate is vagally mediated; however, although atropine prevented bradycardia, it had little o r no effect on hypotension [6]. The finding in Chagas’s heart disease that
vasodepressor syncope occurs in patients with defective
cardiac vagal control supports the view that hypotension
in vasovagal fainting is not a heart-rate-dependent phenomenon [7].
The first section of this review summarizes much of the
physiology of the cardiovascular events in vasovagal fainting, and the second part focuses on the diagnostic and
therapeutic approach to the patient with recurrent vasovagal syncope.
Neural control of the circulation and vasovagal fainting
Correspondence: Dr J. J. van Lieshout, Academic Medical
Centre, Department of Intensive Care - C3, Meibergdreef 9,
1105 AZ Amsterdam, The Netherlands.
Simultaneous vagal activation and sympathetic inhibition during fainting can be regarded as the neural
mechanism of a vasovagal response which can be activated along at least two different neural pathways (Fig. 1)
[S]. T h e first pathway descends from cortico-hypothalamic centres to medullary cardiovascular centres, hence
emotional events can evoke vasovagal fainting (central
type; Fig. 1) [9]. In susceptible human subjects, severe
emotional stress and severe pain (visceral or deep
somatic) may lead to vasovagal syncope; the bradycardia
and hypotension are usually preceded by moderate tachycardia and hypertension [lo]. The neural centres involved
in humans are not known. In animal experiments two
higher centres have been identified as important in cerebral circulatory control: the ‘defence reaction area’ in the
limbic lobe, and the limbic sympatho-inhibitory center
[ l l ] . Stimulation of the former increases heart rate and
blood pressure and provokes reactions resembling the
human reaction to anxiety and anticipation of muscular
exercise (‘fight o r flight’ response). Stimulation of the
latter causes hypotension and bradycardia [ 111.
A second afferent pathway involved in the vasovagal
response is thought to originate in the heart itself (peripheral type) [l,2, 12-16]. This mechanism was clarified
in animal experiments by ThorCn and co-workers [S, 17,
181. In anaesthetized cats, the slowing of the heart rate,
elicited by rapid haemorrhage or pooling of blood, was
preceded by an increased activity in about 20% of nonmyelinated vagal afferent C-fibres from the left ventricle
[ 181. Increased receptor activity preceded the slowing of
the heart [17]. The main determinants of neuronal firing
in these fibres seem to be wall deformation and the
inotropic state of cardiac muscle [1S, 191. The combination of an increased inotropic stimulus to the heart and a
576
J. J. van Lieshout et al.
CENTRES
AFFERENT
Glossopharyngeal nerv
A
EFFERENT
Carotid
afferent
-W
Blood vessels
Fig. 1. Cardiovascular part of the autonomic nervous
system. Schematic representation of the two different
pathways which may elicit the depressor response: the
baroreceptor- and depressor-reflex arcs and the ‘central
command’ to the vasomotor centres (VMC).(Reproduced
from [106].)
decreased volume of the left ventricle is thought to give
rise to a powerful contraction around an almost empty
heart chamber, leading to deformation and activation of
ventricular mechanoreceptors [8, 17,20-221.
The evidence for this sequence is as follows. Gauer [15,
221 induced hypovolaemic shock in dogs and gave them
adrenaline. From the moment the animals developed circulatory shock a great disparity ensued between aortic
pressure, which fell below 50 mmHg, and intraventricular
pressure, which fluctuated between 120 and 200 mmHg.
Cineangiography showed that the almost emptied ventricle continued to contract isometrically. The subendocardial haemorrhages which develop in severe and
longer-lasting hypovolaemic shock [23] probably result
from the same mechanism: trauma resulting from heavy
mechanical forces, acting on the inner wall of the emptied
left ventricle. These haemorrhages are comparable with
the lesions evoked by exposure to centrifugal force in
hypovolaemic hypotensive animals [ 16,221.
However, not all investigators accept the notion that
the signal which causes vasodilatation comes solely from
cardiac receptors. Morita & Vatner [24] found that the
loss of sympathetic drive to the vessels during haemorrhage in conscious dogs was not prevented by surgical
denervation of the heart. There are also data from animal
experiments indicating that when arterial baroreceptor
areas become shrunken by volume depletion, they may
paradoxically increase their firing rates at low blood pressure levels [25,26]. The contribution of baroreceptors to
the vasovagal response in humans remains, however,
unclear. The recent finding in a cardiac transplant patient
of vasodilator-induced withdrawal of sympathetic nerve
activity with syncope supports the view that ventricular
baroreceptor activation is not the exclusive cause of
vasovagal syncope [27].
The term ‘vasodepressor syncope’ has been used to
describe both vasovagal syncope [4, 28-32] and carotid
sinus syncope [33, 341. In both conditions a depressor
reflex is triggered, which leads to vasodilatation and reflex
bradycardia. However, an important difference is that in
carotid sinus syncope, carotid sinus baroreceptor circulatory control is thought to be abnormal.
Another clinically important depressor reflex originating from the heart itself can be observed in myocardial
ischaemia or infarction [35], coronary reperfusion [36]
and during coronary angiography [37]. It is comparable
with the Von Bezold-Jarisch reflex as produced experimentally in laboratory animals by topical application of
chemical stimulants on the heart or their injection into the
coronary circulation [12, 19, 35, 381. Reflex bradycardia
and hypotension may also be observed in glossopharyngeal neuralgia [39] as an example of reflex syncope involving afferent pathways which do not originate from the
heart or hypothalamic centres [40]. It may be impossible to
discriminate sharply between the central and peripheral
types of fainting [40]. It is conceivable that interactions
between the two types take place; for example, primary
vasodilatation initiated by central factors (e.g. seeing one’s
own blood) may lead to a fall in ventricular filling pressure
which stimulates ventricular mechanoreceptors.
Vasodilatation: active or passive, neural or humoral?
Vasovagal fainting sometimes develops in healthy subjects in a laboratory environment; this permits study of the
circulatory transients (Fig. 2) [28, 31, 41, 421. It has also
been observed in the dental chair [l] and has been
induced deliberately by vasodilating agents [43, 441, passive head-up tilt [45, 461, subatmospheric pressure
applied to the lower part of the body [32, 46-49], bleeding and/or application of venous occluding tourniquets
[13, 14,501 and mental stress [51,52]. From thesediverse
experiments the following data have been obtained.
Roddie [53], Weiss et al. [43] and Barcroft et al. [13, 141
studied the forearm muscle blood flow in fainting humans
and found vasodilatation to be confined to muscle.
Debate concerning the nature of these vascular changes in
the forearm has focused on whether the response is neural
or humoral and, if neural, whether it is active or passive
[40, 541. Data on the mechanisms of vasodilatation are
not easy to interpret: the cardiovascular effects of the
577
The vasovagal response
0
1
2
6
5
(b)
7
11
10
Time (min)
I
Time (min)
Fig. 2. Vasovagal syncope. In two healthy male subjects without a history of previous fainting, the
blood pressure and heart rate responses on standing are normal. Both subjects fainted after a
comparable time in the upright position (indicated by arrows). The subject in ( a ) shows a
predominantly vasodilatory type, whereas the subject in ( b )shows a more cardio-inhibitory type
of vasovagal fainting with a cardiac asystole of 10 s. Note the interruptions in the time scale.
(Reproduced from [106].)
various methods used to induce vasovagal syncope in
healthy subjects [40,51,52] differ considerably. A n active
cholinergic vasodilatory mechanism was suggested to be
causative in vasovagal fainting, instead of the reflex inhibition of tonic sympathetic discharge [9, 13, 14, 51-54].
There is evidence for a sympathetic cholinergic vasodilator mechanism in the cutaneous circulation in humans
[40, 541. Active vasodilatation in skeletal muscle was
shown in the cat [54]. Although data in primates are not
uniform [40, 541, Sanders et al. [55] recently provided
evidence for cholinergically mediated vasodilatation in
humans during isometric exercise by sustained handgrip
[55]. Nevertheless, the overall cardiovascular effect of
handgrip is an elevated arterial pressure caused by a rise
in cardiac output [56]. With respect to the vasovagal
response in humans, the evidence for the operation of
active vasodilator nerves remains limited to two findings:
( 1 ) after cervical sympathectomy, vasodilatation is
absent 'during post-haemorrhagic fainting [14], and (2) at
the time of a faint, blood flow in forearms with intact
vasomotor innervation is higher than after neural blockade [57].These findings document vasodilatation but d o
not provide information on its nature: active or passive
[38, 401. Forearm blood flow increases during mental
stress [51, 581. Barcroft et al. [58] and Blair et al. [51]
showed that vasodilatation during mental stress remains
present, although to a lesser extent, after cholinergic
blockade. T h e finding that adequate intra-arterial cholinergic blockade fails to prevent vasodilatation [58] does
not tally with the supposed involvement of a cholinergic
sympathetic vasodilatory system in the genesis of vasovagal syncope.
T h e evidence for passive vasodilatation in fainting as a
result of decreased sympathetic drive is much stronger.
A n important argument is the finding that in fainting
humans plasma noradrenaline levels are relatively low
despite severe hypotension [28, 31, 451. Eckberg et al.
[59] showed that the noradrenaline response was already
blunted at a mild to moderate fall in blood pressure
induced by nitroprusside. There is also evidence for a
sudden decrease in cardiac and renal sympathetic nervous activity during vasovagal fainting: Esler et al. [60]
found a severely reduced cardiac and renal noradrenaline
spillover in patients who fainted during cardiac catheterization. T h e disappearance of secondary reflected arterial
pressure waves at fainting also suggested a decrease in
sympathetic tone [28]. Direct evidence for the withdrawal
of vasoconstrictor activity was provided by the demonstration of a sudden cessation of peroneal muscle sympathetic bursts at the onset of a vasovagal faint in healthy
subjects [42,61,62]and the disappearance and reappearance of muscle sympathetic nerve activity coinciding with
578
J. J. van Lieshout et al.
the onset and termination of vasovagal attacks [63]. An
example of a reduction in sympathetic nerve activity
during a nitroprusside-induced vasovagal reaction is
shown in Fig. 3.
There is also evidence for a humoral vasodilating
mechanism in the vasovagal response. Adrenaline has
been suggested as the humoral agent involved: plasma
levels of adrenaline are elevated in fainting humans [45,
48, 641 (see below) and this catecholamine produces /?adrenergic dilatation in both skeletal muscle and splanchnic resistance vessels at concentrations measured in
humans under stress [9, 48, 491. Increased adrenaline
levels would oppose the vasoconstriction needed to maintain blood pressure during hypotension [48]. The finding
of less vasodilatation after adrenalectomy [58] also supports a contributory role for adrenaline in the mechanism
of vasovagal fainting.
Haemodynamics
response
before
and during the vasovagal
The interpretation of the literature on central haemodynamic changes preceding and during vasovagal fainting
is hampered by the lack of beat-to-beat data on changes in
cardiac output and peripheral vascular resistance.
Cardiac output, as measured by single-indicator-dilution
and Fick techniques, either remains unaltered or falls
below control values with a fall in total peripheral resistance in the early phase of spontaneous, post-haemorrhagic or head-up-tilt-induced syncope [ 13,23,29,65]. In
healthy subjects, induction of fainting by lower-body
negative pressure in the sitting position elicits an initial
rise in heart rate and a decline in blood pressure, followed
by a marked drop in heart rate and further fall in blood
pressure in the last minute before the faint [30, 471. The
fall in stroke volume, cardiac output and arterial pressure
usually occurs before the bradycardia [l11. The fall in
peripheral vascular resistance in vasovagal syncope is
apparently not compensated for by a rise in cardiac output. During prolonged lower-body negative pressure
renal vascular resistance does not change, but splanchnic
vasoconstriction increases [66], which is consistent with
the observation of cyanosis and pallor of the colonic
mucosa during syncope in a patient with a colostomy [67].
The behaviour of the veins in human subjects during
fainting has been studied almost exclusively in limbs,
although they almost certainly make no important contribution to cardiovascular control [68]. Epstein et al. [65]
showed that veins retain the ability to constrict during
vasovagal syncope. Nevertheless, in the majority of
studies on vasovagal syncope, usually induced by withdrawal of blood in volunteers, a drop in central venous
pressure was found to precede syncope [ l , 13, 32, 691.
The fall in central venous pressure induced by lowerbody negative pressure was shown to level off in the presyncopal phase. It might even rise again but to values
below control level [32], probably by venoconstriction
without full restoration of central blood volume [32].
Measurements of central venous pressure in vasovagal
fainting remain, however, difficult to interpret. On the one
hand, a reduction in cardiac output at the onset of syncope (see below) would tend to increase central venous
pressure and thereby mask the effects of peripheral venodilatation [70]. On the other hand, some contribution
towards the recorded fall in atrial pressure could be made
by yawning and the increase in depth of breathing which
precede fainting (Fig. 3) [ll].The hypocapnia associated
with pre-syncopal hyperventilation might amplify the
fainting reaction per se by cerebral vasoconstriction and
systemic vasodilatation [ 11,23,38].
The magnitude of blood volume reduction and its distribution between central and peripheral vascular compartments are independent denominators of the
underlying haemodynamics in vasovagal syncope. A large
fraction of the calf volume change during venous occlusion is attributable to filling of the deep venous spaces
[71]. Since their compliance is determined primarily by
the surrounding skeletal muscle [711, these findings indicate that decreased skeletal muscle tone increases orthostatic venous pooling. This is in agreement with the
finding of Mayerson & Burch [72], who showed that
gastrocnemius intramuscular pressures during passive
head-up tilt were lower in fainters. Although postural
changes in plasma volume or capillary filtration rate in
fainting head-up-tilted healthy subjects are not different
from non-fainters [73], a smaller blood volume is found in
fainters [69]. The pre-syncopal chronotropic response
during tilt in subjects who develop vasovagal syncope is,
however, independent of blood volume [74]. Redistribution of blood volume between central and peripheral
vascular compartments may be more important than total
blood volume. This view is supported by the finding that
selective restoration of central blood volume in vasovagal
syncope, by inflation of an anti-gravity suit, results in a
dramatic resolution of symptoms by rapidly elevating
central venous and arterial pressure [29, 751. The beneficial circulatory effects of ‘re-transfusion’ point to the
paramount role of central blood volume in the circulatory
adaptation to orthostatic stress [4,13,29,32,69].
Hypovolaemic hypotension elicits an antagonism between
cardiovascular reflexes
A drop in blood pressure as a result of a fall in venous
return on assumption of the upright posture, or as a result
of loss of blood, may set in motion two opposing reflexes:
the arterial baroreflex and the depressor reflex. The
former is activated by the fall in arterial blood pressure,
and the latter by a decrease in ventricular filling volume
(Fig. 1)[21]. A n antagonism between the control of filling
pressure and volume of the heart, and the control system
of arterial pressure has been suggested [69, 761. When
syncope ensues, the balance may be directed one-way in
that the depressor reflex overrides the baroreflex; the
latter is still functionally intact [21] but switched off [771.
This shift from pre-syncopal tachycardia to bradycardia
and/or arterial hypotension during syncope [ 111has been
referred to as the ‘biphasic response’ (Fig. 2) [22, 78,793.
The sympathetic ‘fight or flight’ response (moderate
tachycardia, hypertension) shifts to a state analogous to
R-R interval
(ms)
ECG
Muscle
sympathetic
activitv
Tidal
volume
Nitroprusside
dose
(pg min-’ kg-’)
149180
Blood
pressure (mmHg)
600
0
The vasovagal response
146165
1.0
100140
2.0
579
78/41
0
Fig. 3. Vasovagal reaction during sodium nitroprusside infusion in a healthy subject. Muscle sympathetic bursts increase
when arterial blood pressure declines, followed by ‘sympathetic silence’ at the verge of vasovagal syncope with a further
fall in arterial blood pressure and an abrupt increase in R-R interval. Note the change in depth of breathing preceding the
faint as reflected by the increase in tidal volume. (Reproduced from [62]with the permission of the publisher.)
the ‘playing dead‘ reaction seen in young animals placed
in a dangerous situation (bradycardia, hypotension) [5,
22,531.
A possible relationship between psychological stress,
vasovagal syncope and sudden death was hypothesized by
Engel [SO]. The concept of sudden death was developed
in which emotional fainting is held potentially lethal.
Although the majority of subjects with sudden cardiac
death have evidence of recent clotting in their coronary
arteries [81] or hypertrophic cardiomyopathy [82],
emotional fainting could interact with pre-existing cardiovascular disease to cause death [35, 53, SO]. Both cardiovascular collapse associated with prolonged cardiac
asystole and syncope or neurally mediated origin may be
elements of the spectrum of the hypotension-bradycardia
syndrome [83]. Milstein et al. [83] recently showed that
patients with suspected or documented cardiac asystole
and normal conventional electrophysiological evaluation
exhibit a susceptibility to tilt-induced, neurally mediated,
hypotension-bradycardia (see below). Maloney et al. [84]
found in an otherwise healthy man who developed vasovagal syncope on head-up tilting with asystole for 73 s no
abnormalities on subsequent electrophysiological testing
[84]. Although these data do not provide conclusive evidence that vasovagal syncope is life threatening, they suggest that the vasovagal reflex may induce life-threatening
cardiac asystole mimicking sudden cardiac death [SO,
83-85].
The teleological significance of the vasovagal response
is problematic. Is the vasovagal response in humans an
appropriate adjustment to hypovolaemic stress or an
inconvenient remnant of the ‘playing dead reaction’ in
animals? It may be disadvantageous in severe haemorrhagic shock to antagonize the arterial baroreceptor reflex
by accentuating hypotension [86]. However, it may
possess cardiac protective properties by rapidly reducing
myocardial oxygen demand when cardiac strain is excessive [20]. When someone faints as a result of a critically
reduced venous return, the induced reflex bradycardia
together with the increase in venous return by the supine
position will allow for a better diastolic filling of the heart
and restoration of arterial blood pressure [87, 881. From
this speculative viewpoint the vasovagal response may be
regarded as the body’s compromise for the antagonism
between the control of systemic arterial pressure and of
volume and pressure of the heart itself.
Neuroendocrine aspects of the vasovagal response
Cardiovascular reflexes, although of decisive importance in short-term circulatory control, are only part of an
elaborate design of servo-control loops [89].In maintaining orthostasis, intact volume control is of vital importance. The neuroendocrine regulation of water and salt
balance is embodied in the long-term control of body
fluids, with the kidney as the main effector organ regulated by catecholamines, the renin-angiotensin-aldosterone system and antidiuretic hormone (vasopressin)
[90]. Since a normal regulatory role has not been established for atrial peptides they will not be discussed.
Catecholamines. The shift from a pre-syncopal moderate tachycardia to bradycardia and/or arterial hypotension (biphasic response) in vasovagal fainting [22,78,79]
is reflected by an initial elevation of the plasma noradrenaline concentration in the minutes preceding syncope followed by a stable, or possibly declining, level
during the vasovagal response [28,31,64,91].In contrast
to the blunted noradrenaline response, an increase in
plasma adrenaline concentration was reported in most
but not all subjects during vasovagal fainting induced by
nitroglycerine [44], head-up tilt [45, 48, 491 and central
hypovolaemia induced by venous tourniquets [92]. This
discrepancy in the behaviour of adrenaline and noradrenaline seems characteristic of the vasovagal response
[28, 31, 45, 641. The role of adrenaline as a vasodilating
hormone in the pathogenesis of vasovagal fainting is discussed above.
Renin. The secretion of renin is partially controlled by
580
J. J. van Lieshout et al.
efferent adrenergic renal nerve fibres innervating juxtaglomerular cell P-adrenoceptors [93]. When humans
change from the supine to the upright posture, plasma
catecholamine, renin and aldosterone concentrations
normally rise within several minutes. Elevation of the
plasma renin levels is subsequently followed by increases
in the plasma angiotensin I1 concentration [93, 941. The
magnitude of the rise in the plasma renin concentration
strongly depends on the effective circulating volume [95].
The decrease in efferent sympathetic nerve traffic during
vasovagal fainting is not restricted to the limb blood
vessels but involves the kidney as well [60, 94, 961. In
some (but not all) studies in humans, plasma renin activity
was reported to fall just before vasovagal fainting induced
by head-up tilt, and to rise again only after the subjects
had returned to the supine position [28,94,97].
Vasopressin. The secretion of vasopressin is influenced
primarily by plasma osmolality, and secondarily by
changes in blood pressure [90, 981 and by nausea [99,
1001. A rise in plasma vasopressin concentration was
found in conscious rabbits during gradual haemorrhage at
the point that blood pressure suddenly fell. Wang et af.
[loll in conscious dogs, prevented the fall in blood pressure and the rise in vasopressin by pharmacological
blockade of the cardiac nerves, thus substantiating the
role of a signal originating in the cardiac ventricles. The
situation in humans, however, is more complex; the vasopressin response to acute blood volume shifts in cardiac
transplant patients is similar to that in normal cardiacinnervated control subjects [ 1021. This implies that in
humans isosmotic vasopressin release is not exclusively
mediated by cardiac receptor reflexes. In normal humans,
slight hypotension does not stimulate vasopressin secretion [46, 1031. The level of vasopressin increases sharply
up to that which has a direct vasoconstrictor action, but
only with a marked fall in blood pressure such as exists
during haemorrhage or vasovagal fainting [45, 64, 991. In
animals, ventricular receptors [8, 1041 send impulses
along afferent C-fibres which activate the vomit centre
and result in a general vagal discharge with reflex relaxation of the stomach [99, 1051. In addition, nausea itself is a
potent direct stimulus of vasopressin secretion by activation of central cholinergic pathways [99, 1001. Since
nausea as a typical premonitory symptom is often associated with vasovagal fainting, a rise in plasma vasopressin
concentration in fainting humans does not permit further
discrimination between central and ventricular reflex
release of vasopressin. Our observations in patients with
pure autonomic failure that very low upright blood pressure is accompanied by a rise in vasopressin, but not by
sensations of nausea [ 1061, support a causal relationship
between hypotensive arterial baroreceptor inhibition and
vasopressin release. Interactions between vasopressin and
the other neural control systems involved in vasovagal
fainting seem relevant, since vasopressin sensitizes cardiac vagal afferent nerves [20] and probably modulates
arterial baroreflex gain [ 1071. The observation of reduced
diuresis in the first hour after vasovagal fainting is
explained by the rise in vasopressin [ l , 4,411.
Pancreatic polypeptide. The hormone pancreatic poly-
peptide controls exocrine pancreas secretion and influences the motility of the biliary ducts [108].The stimuli
for its secretion are hypoglycaemia and the presence of
food in the stomach, mediated by cholinergic fibres in the
vagus nerve [108]. Impaired secretion in diabetic patients
probably indicates early autonomic neuropathy [ 1091.
The drop in blood pressure and heart rate in healthy subjects on syncope provoked by 60" head-up tilt is accompanied by a rapid increase in pancreatic polypeptide
concentration, indicating an increase in tone of the
abdominal part of the vagus nerve [45,86]. It seems likely
that a general vagal discharge underlies both the bradycardia [45] and the increase in pancreatic polypeptide
concentration [ 1081.
Opiates. Opiates may also be involved in the vasovagal
response, since endogenous opiate mechanisms are implicated in the vasodilator response to acute blood loss:
pharmacological doses of the opiate antagonist naloxone
in conscious animals attenuate the fall in blood pressure
that arises when a large volume of blood is rapidly withdrawn [46, 77, 1lo]. The dose of naloxone delivered into
the fourth ventricle that prevents vasoconstriction is
90-900 times less than a corresponding intravenous dose,
indicating that the endogenous opioid mechanisms act in
the central nervous system rather than peripherally [46,
771.
CLINICAL EXPRESSION OF THE VASOVAGAL
RESPONSE
The conimon faint
The relative extent to which blood pressure and heart
rate fall during true vasovagal fainting may vary from
severe hypotension with only a small drop in heart rate
(Fig. 2a) [ 13, 441 to a predominantly cardio-inhibitory
type (sometimes with clear sinus arrest) (Fig. 2b) [6, 31,
64, 84, 86, 92, 11I]. Although vasovagal fainting may
develop when subjects are supine [63, 1111, as a rule it
happens in the upright position. Vasovagal responses are
relatively common in daily life: up to 20% of 'normal' subjects lose consciousness at some time in their lives [112,
1131 with a 5-15% rate in healthy blood donors [79].
Vasovagal syncope provoked by passive head-up tilt has a
comparable incidence of about 15% [ 114-1 161. There
may be a constitutional predisposition: fainting develops
more frequently in slender subjects [5,69, 1171; this might
be related to a larger leg compliance when muscle mass is
less [118]. True vasovagal fainting is rare in the elderly
compared with young subjects [119-1211.
It is not known why patients with orthostatic hypotension owing to autonomic neuropathy or high spinal cord
lesions do not develop premonitory symptoms of vasovagal fainting [50, 1221. Van Lieshout and co-workers
[106, 1231 found in two patients with hypoadrenergic
orthostatic hypotension who had intact vagal cardiac control, that vasovagal bradycardiac responses were surprisingly absent despite a profound fall in arterial pressure on
standing. A possible explanation might be their inability
to release adrenaline, which protects the heart from too
The vasovagal response
vigorous contractions and thereby prevents triggering of
ventricular receptors.
Relatively slow heart rate in hypovolaemic arterial hypotension
An unexpectedly low heart rate was found in patients
in shock during a study of air-raid casualties in World War
I1 [124, 1251, in volunteers bled large amounts experimentally [126] and in patients with acute hypotensive
intraperitoneal bleeding [ 1271. Sander-Jensen et af. [86]
found in patients with serious bleeding from various
causes during the phase of hypovolaemic shock, a mean
heart rate of only 73 beats/min. Resuscitation with fluids
increased both blood pressure and heart rate. Thus a relatively low heart rate may be a clinical sign of severe
hypovolaemic shock. It has even been argued that the
widespread belief that haemorrhage is easily diagnosed by
a rapid pulse and a low blood pressure is based on a historical misconception [ 126, 1271. A relative bradycardia
accompanying an important reduction of venous return
seems a paradox and points to an apparently inappropriate increase in vagal activity [4, 32,861. Patients with leftsided congestive heart failure show the opposite
phenomenon: they tolerate orthostatic stress during passive head-up tilt remarkably well and are reported to be
free of vasovagal syncope [2, 221. The elevated central
blood volume and heightened venous pressure may prevent a fall in upright cardiac filling pressures which results
in improved orthostatic tolerance [ 1281. An alternative
explanation is that decreased /3-adrenoceptor density in
patients with congestive heart failure [129] interferes with
the increased contractility that is thought to increase ventricular receptor firing and provoke vasovagal responses
in healthy subjects.
Factors predisposing to vasovagal syncope
Exercise. Decreased orthostatic tolerance is found in
healthy subjects after exhaustive exercise [ 1301. Blunting
of the vasoconstrictive responses in the exercised muscles
by hyperthermia and local lactic acidaemia are postulated
to underlie the failure in maintaining arterial pressure in
the upright position. Also the fall in arterial pressure
which reduces left ventricular afterload is thought to allow
a more complete emptying of the ventricle, thereby eliciting a vasodepressor reflex [88].
Aortic and subaortic stenosis. Patients with aortic or
subaortic stenosis may suddenly develop syncope during
or shortly after critical levels of exercise [131]. Recently,
the exertion-syncope of aortic valve stenosis was suggested not to be solely due to flow obstruction but also to
be of reflex origin [22, 132, 1331. During leg exercise
there is a fall in forearm vascular resistance, rather than
the increase that occurs in normal subjects [132]. Reflex
vasodilatation by the sudden large elevation in left ventricular pressure in the face of an obstructed outflow tract
[54, 1331, together with fixed cardiac output, has been
suggested as the most common pathogenic mechanism of
syncope in aortic stenosis [134]. Exertion syncope in
581
aortic stenosis patients is an example of vasodepressor
syncope not primarily elicited by venous pooling.
Cardiovascular drugs. The instantaneous increase in
venous capacitance by drugs, such as organic nitrates and
frusemide, and by alcohol predisposes to vasovagal
responses. Vasodepressor reflex activation due to reduced
cardiac filling sensitizing ventricular receptors seems a
common contributor to postural syncope in patients who
use these drugs (for a review, see [19]).
Diagnosis of vasovagal syncope
The diagnostic work-up of syncope is difficult because
of its intermittent nature and the frequent absence of diagnostic abnormalities [113]. An approach to diagnosis of a
patient presenting with syncope commences with a
thorough history, taken with particular attention to the
following diagnostic clues: age (recent onset of syncope in
the elderly is most commonly due to cardiac disease),premonitory symptoms (profuse sweating, pallor, nausea,
yawning, hyperventilation, pupillary dilation), the conditions provoking syncope (emotion, fear, pain, hot environment, use of medication, prolonged standing, exercise),
and the estimated duration of a prodromal period (usually
longer in vasovagal than cardiac syncope [ 1351).Physical
examination should include recording of blood pressure
and pulse rate in the supine position and after several
minutes upright, looking for hypotension and/or bradycardia, and a neurological examination.
Electrocardiographic abnormalities are found in
5-10°/0 of the patients with syncope [112]. In case of a
normal ECG, prolonged ambulatory electrocardiographic
monitoring seems justified in the light of incidence of
unsuspected rhythm disturbances [ 1361. The main problem with interpreting Holter data is the high incidence of
false-positive results [ 1121. Electrophysiological testing
has been advocated to exclude undetected vagally mediated brady-arrhythmia [40]. In young children [137, 1381
and in trained athletes [134, 1391with a history of cardiac
syncope, resting sinus bradycardia and varying degrees of
sino-atrial and atrio-ventricular block have been considered to indicate increased vagal activity as a cause of
the syncope. Atropine returns these functional conduction abnormalities to normal [137, 140, 1411. The
absence, however, of hypotension during the spells of
bradycardia [ 1381 and the beneficial effects of atropine
[137, 1381 throw doubt on its true vasovagal nature.
Because of the arbitrary definition of excessive vagal tone,
its unclear prognostic importance [ 1361 and the low sensitivity, specificity and predictive accuracy of electrophysiological testing in patients with syncope caused by transient
bradycardia [134, 1421, such testing should be offered
only to patients with structural heart disease and recurrent unexplained syncope [112, 113,134,1431.
Prolonged standing and passive head-up tilt induce
venous pooling to a variable extent. Although the tiltboard was already used in 1945 by Allen et al. [144] to
study orthostatic hypotension, the growing literature on
the use of postural change as a promising diagnostic test
in the assessment of recurrent syncope with reproduction
582
J. J. van Lieshout et al.
of symptoms is of recent onset [83, 84, 113, 134,
145-1481. Vasovagal responses in the cardiovascular
laboratory are, however, not uncommon in healthy subjects who otherwise never faint (Fig. 2) [ l l , 116, 122,
1491. In addition, the pre-test probability of a positive test
result, i.e. syncope during passive head-up tilt, depends
especially on two independent variables: age and nature
of instrumentation (invasive versus non-invasive). Susceptibility to vasovagal syncope during tilt is common in
the young but is reduced in the old [119-1211. In addition, use of intravascular instrumentation increases the
incidence of fainting up to 90% [41, 114, 1501.The diagnostic value of vasovagal fainting as induced by head-up
tilt in a non-invasively instrumented older subject with
unexplained syncope is probably considerable. However,
with invasive instrumentation, the value of a tilt test is still
unknown in the elderly, and is low in the young [150].
Although the addition of intravenous isoprenaline to passive head-up tilt increases the incidence of bradycardia,
hypotension and syncope in patients with unexplained
syncope and negative electrophysiological tests [ 147, 151,
1521, false-positive test results have been reported [ 1511.
Therapeutic approach to the habitual vasovagal fainter
In 1826, Piorry [153] suggested that syncope be treated
by having the patient lie down and by lowering the head
even further and raising the legs, using gravity to beneficial effect. Although patients suffering from true vasovagal syncope in general have a benign prognosis [ 1541,
extreme episodes of reflex asystole can cause major disability [84]. Muscle-tensing manoeuvres of the lower
limbs alone [155] or in combination with abdominal
muscles (anti-G manoeuvre) [47] may be successful in
preventing syncope. In more serious recurrent vasovagal
syncope we propose the following approach. First, the use
of measures to counteract pooling of blood in the legs and
the abdomen such as elastic stockings and abdominal
binders. Secondly, isometric exercise of the muscles of the
legs and abdominal wall may be helpful in improving
orthostatic tolerance [28, 1561. There is also evidence
that moderate endurance training improves orthostatic
tolerance in deconditioned subjects; the resulting increment in plasma volume might contribute to this effect
[ 157- 1591. However, the relation between degree of aerobic fitness and orthostatic tolerance remains controversial: both reduced and normal orthostatic tolerance have
been reported in endurance-trained subjects [ 158, 1601
(for a review see [156]). Orthostatic tolerance in individuals who undergo strength training differs from those
who undergo endurance-type training [ 1561. Muscle mass
seems to be the more important factor contributing to the
prediction of leg compliance independent of aerobic fitness [118, 1581; a larger muscle mass of the lower extremities might restrict orthostatic venous pooling [ 1561.
If true, a training programme should ideally be aimed at
minimizing the blood volume change in the leg veins
under similar pressure changes.
A purely cardio-inhibitory type of vasovagal syncope
(Fig. 1) should respond to either cardiac pacing or anti-
cholinergic drugs. Permanent cardiac pacing, beneficial in
the cardio-inhibitory type of carotid-sinus hypersensitivity, has not been of help in its vasodepressor variant
[33, 341. Permanent pacing has been recommended to
prevent bradycardia in those patients who develop long
periods of asystole during their vasovagal attacks [84]. In
vasovagal fainting, cardiac pacing has, however, never
proved to prevent hypotension [161, 1621. The finding
that atropine neither influences the degree of hypotension
[6, 13, 28-30] nor the time until the onset of vasovagal
syncope during orthostatic stress [163] does not justify
anti-cholinergic drug treatment. Nevertheless, some
patients with recurrent vasovagal syncope who were
treated with transdermal scopolamine remained free of
symptoms for as long as they received the drug [84].
Although its parasympatholytic effects are probably significant [84], administration transdermally leads to low
serum levels in the parasympathomimetic range [113,
1641. The reported efficacy of scopolamine in vasovagal
syncope might be related to modulatjon of cerebral autonomic outflow, since scopolamine, unlike atropine, has a
central depressant effect [113].
Propranolol, in pharmacological doses, was suggested
to decrease the activity of ventricular mechanoreceptors
[165]. Although this was not confirmed in a recent study
[ 1201, cardioselective P,-adrenoceptor blockade was
reported earlier to diminish susceptibility to vasovagal
responses during passive head-up tilt [91, 1521. If true,
B-adrenoceptor blockade might be of help in preventing
vasovagal syncope by modifying neural traffic from and to
the heart, blocking both afferent and efferent pathways
[201.
Normovolaemic patients with debilitating recurrent
vasovagal syncope may improve with measures which
increase blood volume, such as generous salt diets and
mineralocorticoids [84]. We suggest an attempt to
increase the patients’ blood volume by the use of head-up
tilt at night. This was earlier proven to be effective in
patients with orthostatic hypotension due to the postural
tachycardia syndrome [166] and in patients with autonomic circulatory failure [106,166, 1671.
SUMMARY
The vasovagal response is the development of inappropriate cardiac slowing and arteriolar dilatation. Vasovagal
responses reflect autonomic neural changes: bradycardia
results from sudden augmentation of efferent vagal
activity, and hypotension results from sudden reduction
or cessation of sympathetic activity and relaxation of
arterial resistance vessels. Two different neural pathways
are thought to be involved, one originating in the hypothalamus, the other in the heart. Direct hypothalamic activation of the medullary cardiovascular centres triggered
by emotional stress or pain causes a vasovagal response
(central type). The combination of a reduced central
blood volume secondary to venous pooling or blood loss,
and an increased inotropic state of the heart, may stimulate ventricular mechanoreceptors and provoke vasodila-
The vasovagal response
tation and bradycardia (peripheral type). Cardiovascular
afferents originating from stretch receptors in various
parts of the vascular tree sometimes induce opposite
reflexes when compared with those from ventricular afferents. T h e depressor reflex involved in the peripheral type
of vasovagal response originates i n the heart itself and
overrides normal baroreflex circulatory control; an antagonism between the control of volume and pressure on
the filling side of the heart and the control system of
arterial pressure becomes apparent. Vasovagal responses
are not necessarily abnormal; the neural pathways
involved in the vasovagal response are probably present
in all healthy subjects who individually mainly differ in
susceptibility.
REFERENCES
1. Sharpey-Schafer, E.P. Emergencies in general practice.
Syncope. Br. Med. J. 1956; i, 506-9.
2. Sharpey-Schafer, E.P., Hayter, C.J. & Barlow, E.D.
Mechanism of acute hypotension from fear or nausea. Br.
Med. J. 1958; ii, 878-80.
3. Palmer, J.F. Works of John Hunter 1837. London, vol. 3.
4. Weissler, A.M. & Warren, J.V. Vasodepressor syncope.
Am. Heart J. 1959,40,786-94.
5. Hill, L. The influence of the force of gravity on the circulation of the blood. J. Physiol. (London) 1895; 18,15-53.
6. Lewis, T. Vasovagal syncope and the carotid sinus
mechanism with comments on Gowers’s and Nothnagel’s
syndrome. Br. Med. J. 1932; i, 873-6.
7. Marin Neto, J.A., Gallo, L., Manco, J.C., Rassi, A. &
Amorim, D.S. Postural reflexes in chronic Chagas’s heart
disease. Cardiology 1975; 60,343-57.
8. Thortn, P. Depressor reflex from the heart during severe
haemorrhage. In: Hainsworth, R., McWilliam, P.N. &
Mary, D.A.S.G., eds. Cardiogenic reflexes. Oxford: Oxford
University Press: 1987,389-401.
9. Shepherd, J.T. & Vanhoutte, P.M. The human cardiovascular system. Facts and concepts. New York Raven
Press, 1980.
10. Wahbha, M.M.A.E., Morley, C.A., Al-Shamma, Y.M.H. &
Hainsworth, R. Cardiovascular reflex responses in patients
with unexplained syncope. Clin. Sci. 1989; 77,547-53.
1 1 . Smith, J.J. Circulatory response to the upright posture.
Boca Raton: CRC Press: 1990,30-3.
12. Jarisch, A. Vagovasale Synkope. Z. Kreislaufforsch. 1941;
23.267-79.
13. Bakroft, H., Edholm, O.G., McMichael, J. & SharpeySchafer, E.P. Posthaemorrhagic fainting. Study by cardiac
output and forearm flow. Lancet 1944; i, 489-91.
14. Barcroft, H. & Edholm, O.G. On the vasodilatation in
human skeletal muscle during post-haemorrhagic fainting.
J. Physiol. (London) 1945; 104, 161-75.
15. Gauer, O.H. Evidence in circulatory shock of an isometric
phase of ventricular contraction following ejection. Fed.
Proc. Fed. Am. SOC.Exp. Biol. 1950; 9,47.
16. Gauer, O.H. & Hen?y, J.P. Negative ( - Gz) acceleration in
relation to arterial oxygen saturation, subendocardial
haemorrhage and venous pressure in the forehead. Aerosp. Med. 1964; 35,533-45.
17. Oberg, B. & Thortn, P. Increased activity in left ventricular receptors during haemorrhage or occlusion of caval
veins in the cat - a possible cause of the vaso-vagal reaction. Acta Physiol. Scand. 1972; 85,164-73.
18. Thortn, P., Skarphedinsson, J.O. & Carlsson, S. Sympathetic inhibition from vagal afferents during severe
haemorrhage in rats. Acta Physiol. Scand. 1988; 133
(SUPPI.571), 97-105.
19. Semple, P.F., ThorCn, P. & Lever, A.F. Vasovagal reactions
583
to cardiovascular drugs: the first dose effect. J. Hypertens.
1988; 6,601-6.
20. Abboud, EM. Ventricular syncope: is the heart a sensory
organ? N. Engl. J. Med. 1989; 320,390-2.
21. Blomqvist, C.G. & Stone, H.L. Cardiovascular adjustments
to gravitational stress. In: Shepherd, J.T. & Abboud, EM.,
eds. Handbook of physiology. American Physiological
Society, Washington, D.C.: 1984, 1025-63. (The cardiovascular system, section 2.)
22. Henry, J.P. On the triggering mechanism of vasovagal
syncope. Psychosom. Med. 1984; 46,91-3.
23. Sheehan, H.L. The pathology of obstetric shock. J. Obstet.
Gynecol. 1939; 46,218-31.
24. Morita, H. & Vatner, S.F. Effects of haemorrhage on renal
nerve activity in conscious dogs. Circ. Res. 1985; 57,
788-93.
25. Landgren, S. On the excitation mechanism of the carotid
baroreceptors. Acta Physiol. Scand. 1952; 26,l-34.
26. Angel1 James, J.E. The responses of aortic arch and right
subclavian baroreceptors to changes of non-pulsatile pressure and their modification by hypothermia. J. Physiol.
(London) 1971; 214,201-23.
27. Schemer, U., Vissing, S., Morgan, B.J., Hanson, P. &
Victor, R.G. Vasovagal syncope after infusion of a vasodilator in a heart-transplant recipient. N. Engl. J. Med.
1990; 332,602-4.
28. Goldstein, D.S., Spanarkel, M., Pitterman, A. et al. Circulatory control mechanisms in vasodepressor syncope. Am.
HeartJ. 1982; 104,1071-5.
29. Weissler, A.M., Warren, J.V., Estes, E.H., McIntosh, H.D.
& Leonard, J.J. Vasodepressor syncope. Factors influencing cardiac output. Circulation 1957; 15,875-82.
30. Murray, R.H. & Shropshire, S. Effect of atropine on circulatory responses to lower body negative pressure and
vasodepressor syncope. Aerosp. Med. 1970; 41,717-22.
31. Ziegler, M.G., Echon, C., Wilner, K.D., Specho, P., Lake,
C.R. & McCutchen, J.A. Sympathetic nervous withdrawal
in the vasodepressor (vasovagal) reaction. J. Auton. Nerv.
Syst. 1986; 17,273-8.
32. Murray, R.H., Thompson, L.J., Bowers, J.A. & Albright,
C.D. Hemodynamic effects of graded hypovolemia and
vasodepressor syncope induced by lower body negative
pressure. Am. Heart J. 1968; 76,799-81 1.
33. Almquist, A., Gornick, C., Benson, D.W., Dunnigan, A. &
Benditt, D.G. Carotid sinus hypersensitivity: evaluation of
the vasodepressor component. Circulation 1985; 71,
927-36.
34. Zee-Cheng, C. & Gibbs, H.R. Pure vasodepressor carotid
sinus hypersensitivity.Am. J. Med. 1986; 81, 1095-7.
35. Robertson, D., Hollister, AS., Forman, M.B. & Robertson,
R.M. Reflexes unique to myocardial ischemia and infarction. J. Am. Coll. Cardiol. 1985; 5,99-104B.
36. Wei, J.Y., Markis, J.E., Malagold, M. & Braunwald, E.
Cardiovascular reflexes stimulated by reperfusion of
ischemic myocardium in acute myocardial infarction. Circulation 1983; 67,796-801.
37. Eckberg, D.L., White, C.W., Kioschos, J.M. & Abboud,
EM. Mechanisms mediating bradycardia during coronary
arteriography. J. Clin. Invest. 1974; 54,1455-61.
38. Hainsworth, R. Fainting. In: Bannister, R., ed. Autonomic
failure, a textbook of clinical disorders of the autonomic
nervous system. Oxford: Oxford University Press, 1988:
142-58.
39. Wallin, B.G., Westerberg, C.E. & Sundlof, G. Syncope
induced by glossopharyngeal neuralgia: sympathetic outflow to muscle. Neurology 1984; 34,522-4.
40. Rowell, L.B. Active neurogenic vasodilatation in man. In:
Vanhoutte, P.M. & Leusen, I., eds. Vasodilatation. New
York: Raven Press, 1981: 1-17.
41. Glick, G. & Yu, P.N. Hemodynamic changes during spontaneous vasovagal reactions. Am. J. Med. 1963; 34,
42-51.
42. Wallin, B.G. & Sundlof, G. Sympathetic outflow to muscles
584
J. J. van Lieshout et al.
during vasovagal syncope. J. Auton. Nerv. Syst. 1982; 6,
287-91.
43. Weiss, S., Wilkins, R.W. & Haynes, F.W. The nature of circulatory collapse induced by sodium nitrite. J. Clin. Invest.
1937; 16.73-84.
44. Robertson, D., Johnson, G.A., Robertson, R.M., Nies,
S.A., Shand, D.G. & Oates, J.A. Comparative assessment
of stimuli that release neuronal and adrenomedullary catecholamines in man. Circulation 1979; 59,637-43.
45. Sander-Jensen, K., Secher, N.H., Astrup, A. et al. Hypotension induced by passive head-up tilt: endocrine and circulatory mechanisms. Am. J. Physiol. 1986; 251 (Regul.
Integr. Comp. Physiol. 22), R742-8.
46. Schadt, J.C. & Ludbrook, J. Hemodynamic and neuro
humoral responses to acute hypovolemia in conscious
mammals. Am. J. Physiol. 1991;260 (Heart Circ. Physiol.
29). H305-18.
47. Newberry, P.D., Hatch, A.W. & MacDonald, J.M. Cardiorespiratory events preceding syncope induced by a
combination of lower body negative pressure and head-up
tilt. Aerosp. Med. 1970; 41,373-8.
48. Rowell, L.B. & Blackmon, J.R. Hypotension induced by
central hypovolemia and hypoxemia. Clin. Physiol. 1989;
9,269-77.
49. Rowell, L.B. & Seals, D.R. Sympathetic activity during
graded central hypovolemia in hypoxemic humans. Am. J.
Physiol. 1990; 259 (Heart Circ. Physiol. 28), H1197-206.
50. Warren, J.V., Brannon, E.S., Stead, E.A. & Merrill, A.J.
The effect of venesection and the pooling of blood in the
extremities on the atrial pressure and cardiac output in
normal subjects with observations on acute circulatory collapse in three instances. J. Clin Invest. 1945; 24,337-44.
51. Blair, D.A., Glover, W.E., Greenfield, A.D.M. & Roddie,
I.C. Excitation of cholinergic vasodilator nerves to human
skeletal muscles during emotional stress. J. Physiol.
(London) 1959; 148,633-47.
52. Greenfield, A.D.M. An emotional faint. Lancet 1951; i,
1302-3.
53. Roddie, I.C. Human responses to emotional stress. Ir. J.
Med. Sci. 1977; 395-417.
54. Shepherd, J.T. Circulation to skeletal muscle. In: Shepherd,
J.T. & Abboud, EM., eds. Handbook of physiology.
American Physiological Society, Washington, D.C.: 1984,
3 19-70. (The cardiovascular system, section 2.)
55. Sanders, J.S., Mark, A.L. & Ferguson, D.W. Evidence for
cholinergically mediated vasodilatation at the beginning of
exercise in humans. Circulation 1989; 79,815-24.
56. Lind, A.R., Taylor, S.H., Humphreys, P.W., Kennelly, B.M.
& Donald, K.W. The circulatory effects of sustained voluntary muscle contraction. Clin. Sci. 1964; 27,229-44.
57. Greenfield, A.D.M. Survey of the evidence for active
neurogenic vasodilatation in man. Fed. Proc. Fed. Am. SOC.
Exp. Biol. 1966; 25, 1607-10.
58. Barcroft, H., Brod, J., He$, Z., Hirsjarvi, E.A. & Kitchin,
A.H. The mechanism of the vasodilatation in the forearm
muscle during stress (mental arithmetic). Clin. Sci. 1960;
19,577-86.
59. Eckberg, D.L., Harkins, S.W., Fritsch, J.M., Musgrave,
G.E. & Gardner, D.F. Baroreflex control of plasma norepinephrine and heart period in healthy subjects and diabetic patients. J. Clin. Invest. 1986; 78,366-74.
60. Esler, M., Jennings, G., Lambert, G., Meredith, I., Horne,
M. & Eisenhofer, G. Overflow of catecholamine neurotransmitters to the circulation: source, fate and functions.
Physiol. Rev. 1990; 70,963-85.
61. Burke, D., Sundlof, G. & Wallin, B.G. Postural effects o n
muscle nerve sympathetic activity in man. J. Physiol.
(London) 1977; 272,399-414.
62. Eckberg, D.L. & Sleight, P. Human baroreflexes in health
and disease. Oxford: Oxford University Press, 1991. (In
press).
63. Yatomi, A., Iguchi, A., Uemura, K., Sakamoto, N., Iwase, S.
& Mano, T. A rare case of recurrent vasodepressive
64.
65.
66.
67.
68.
69.
70.
71.
72.
attacks of 2-hours duration: analysis of the mechanisms by
muscle sympathetic nerve activity recording. Clin. Cardiol.
1989; 12,164-8.
Robinson, B.J. & Johnson, R.H. Why does vasodilatation
occur during syncope? Clin. Sci. 1988; 74,347-50.
Epstein, S.E., Stampfer, M. & Beiser, G.D. Role of the
capacitance and resistance vessels in vasovagal syncope.
Circulation 1968; 37,524-33.
Hirsch, A.T., Levenson, D.J., Cutler, S.S., Dzau, VJ. &
Creager, M.A. Regional vascular responses to prolonged
lower body negative pressure in normal subjects. Am. J.
Physiol. 1989; 257 (Heart Circ. Physiol. 26), H219-55.
Grayson, J. & Swan, H.J.C. Intestinal blood-flow changes
in man during fainting. J. Physiol. (London) 195 1; 112,44.
Hainsworth, R. The importance of vascular capacitance in
cardiovascular control. News Physiol. Sci. 1990; 5,250-4.
Bergenwald, L., Freyschuss, U. & Sjostrand, T. The
mechanism of orthostatic and haemorrhagic fainting.
Scand. J. Clin. Lab. Invest. 1977; 37,209-16.
Rothe, C.F. Venous system: physiology of the capacitance
vessels. In: Shepherd, J.T. & Abboud, EM., eds. Handbook
of
physiology. American
Physiological
Society,
Washington, D.C.: 1984, 397-453. (The cardiovascular
system, section 2.)
Buckey, J.C., Peshock, R.M. & Blomqvist, C.G. Deep
venous contribution to hydrostatic blood volume change in
the human leg. Am. J. Cardiol. 1988; 62,449-53.
Mayerson, H.S. & Burch, G.E. Relationships of tissue
(subcutaneous and intramuscular) and venous pressure to
svncoDe induced in man bv gravitv. Am. J. Phvsiol. 1940:
i z s , i58-69.
Tarazi. R.C.. Melsher. HJ.. Dustan. H.P. & Frohlich. E.D.
Plasma volume changes with upright tilt: studies in hypertension and in syncope. J. Appl. Physiol. 1970; 28, 121-6.
Foud-Tarazi, EM. & Maloney, J.D. Vasovagal syncope:
lack of relationship between baseline blood volume and
presyncopal chronotropic response. Clin. Res. 1989; 37,
880A.
Kravik, S.E., Keil, L.C., Geelen, G. et al. Effect of antigravity suit inflation on cardiovascular, PRA, and PVP
responses in humans. J. Appl. Physiol. 1986; 62,766-74.
Tarazi, R.C. & Fouad, EM. Circulatory dynamics in progressive autonomic failure. In: Bannister, R., ed. Autonomic failure, a textbook of clinical disorders of the
autononic nervous system. Oxford: Oxford University
Press, 1983: 109-14.
Ludbrook, J. Faint heart. Br. Med. J. 1989; 298, 1053-4.
Vingerhoets, A.J.J.M. Biochemical changes in two subjects
succumbing to syncope. Psychosom. Med. 1984; 46,
96-103.
Graham, D.T. Prediction of fainting in blood donors. Circulation 1961; 23,901-6.
Engel, G.L. Psychologic stress, vasodepressor (vasovagal)
syncope, and sudden death. Ann. Intern. Med. 1978; 89,
403-12.
Davies, M.J. & Thomas, A. Thrombosis and acute
coronary-artery lesions in sudden cardiac ischemic death.
N. Engl. J.Med. 1984;310,1137-40.
McKenna, W.J. & Camm, A J . Sudden death in hypertrophic cardiomyopathy. Assessment of patients at high
risk. Circulation 1989; 89, 1489-92.
Milstein, S., Buetikofer, J., Lesser, J. et al. Cardiac asystole:
a manifestation of neurally mediated hypotensionbradycardia. J. Am. COIL Cardiol. 1989; 14, 1626-32.
Maloney, J.D., Jaeger, F.J., Fouad-Tarazi, EM. & Morris,
H.H. Malignant vasovagal syncope: prolonged asystole
provoked by head-up tilt. Cleveland Clin. J. Med. 1988;
55,542-8.
McIntosh, H.D. The stabilizing and unstabilizing
influences of neurogenic and vascular activities of the heart
as related to sudden cardiac death. J. Am. Coll. Cardiol.
1985; 5,105-10B.
Sander-Jensen, K., Secher, N.H., Bie, P., Warberg, J. &
.-
73.
74.
75.
76.
77.
78.
79.
80.
81.
82.
83.
84.
85.
86.
I
,
I
,
<
The vasovagal response
Schwartz, T.W. Vagal slowing of the heart during
haemorrhage: observations from 20 consecutive hypotensive patients. Br. Med. J. 1986; 292,3646.
87. Oberg, B. & White, S. The role of vagal cardiac nerves and
arterial baroreceptors in the circulatory adjustments to
haemorrhage in the cat. Acta Physiol. Scand. 1970; 80,
395-403.
88. Bjurstedt, H., Rosenhamer, G., Balldin, U. & Katkov, V.
Orthostatic reactions during recovery from exhaustive
exercise of short duration. Acta Physiol. Scand. 1983;
119,25-31.
89. Guyton, A.C., Cowley, A.W., Norman, R.A., Coleman,
T.G. & Samar, R.E. The arterial baroreceptor reflex, a
pressure buffering system. In: Guyton, A.C., ed. Circulatory physiology 111. Arterial pressure and hypertension.
Philadelphia: W.B. Saunders, 1980,248-65.
90. Bennett, T. & Gardiner, S.M. Involvement of vasopressin
in cardiovascular regulation. Cardiovasc. Res. 1985; 19,
57-68.
91. Goldenbeg, I.F., Almquist, A., Dunbar, D.N., Milstein, S.,
Pritzker, M.R. & Benditt, D.G. Prevention of neurallymediated syncope by selective beta-1 adrenoceptor blockade. Circulation 1987; 76 (Suppl.),IV-133.
92. Sander-Jensen, K., Mehlsen, J., Secher, N.H. et al. Progressive central hypovolaemia in man resulting in a vasovagal
syncope? Haemodynamic and endocrine variables during
venous tourniquets of the thighs. Clin. Physiol. 1987; 7,
231-42.
93. Zanchetti, AS. Neural regulation of renin release. Circulation 1977; 56,691-8.
94. Oparil, S., Vassaux, C., Sanders, C.A. & Haber, E. Role of
renin in acute postural homeostasis. Circulation 1970; 41,
89-95.
95. Tuck, M.L., Dluhy, R.G. & Williams, G.H. Sequential
responses of the renin-angiotensin-aldosterone axis to
acute postural change: effect of dietary sodium. J. Lab.
Clin. Med. 1975; 86,754-63.
96. Morganti, A., Lopez-Ovejero, J.A., Pickering, T.G. &
Laragh, J.H. Role of the sympathetic nervous system in
mediating the renin response to head-up tilt. Am. J.
Cardiol. 1979; 43,600-4.
97. Shvartz, E., Convertino, V.A., Keil, L.C. & Haines, R.F.
Orthostatic fluid-electrolyte and endocrine responses in
fainters and nonfainters. J. Appl. Physiol. 1981; 51,
1404-10.
98. Goldsmith, S.R. Baroreflex control of vasopressin secretion in normal humans. In: Cowley, A.W., Liard, J.F. &
Ausiello, D.A., eds. Vasopressin. Cellular and integrative
functions. New York: Raven Press 1988,389-97.
99. Sander-Jensen, K., Mehlsen, J., Stadeager, C. et al. Increase
in vagal activity during hypotensive lower-body negative
pressure in humans. Am. J. Physiol. 1988; 253 (Regul.
Integr. Comp. Physiol. 24), R149-56.
100. Verbalis, J.G., Richardson, D.W. & Stricker, E.M. Vasopressin release in response to nausea-producing agents
and cholecystokinin in monkeys. Am. J. Physiol. 1987;
252 (Regul.Integr. Comp. Physiol. 23), R749-53.
101. Wang, B.C., Flora-Ginter, G., Leadley, R.J. & Goetz, K.L.
Ventricular receptors stimulate vasopressin release during
hemorrhage. Am. J. Physiol. 1988; 254 (Regul. Integr.
Comp. Physiol. 23), R204-11.
102. Convertino, V.A., Thompson, C.A., Benjamin, B.A. et al.
Haemodynamics and ADH responses to central blood
volume shifts in cardiac-denervated humans. Clin. Physiol.
1990; 10,55-67.
103. Goldsmith, S.R. The effect of moderate hypotension on
vasopressin levels in normal humans. Am. J. Med. Sci.
1989; 298,295-8.
104. Abrahamsson, H. & ThorCn, P. Vomiting and reflex vagal
relaxation of the stomach elicited from heart receptors in
the cat. Acta Physiol. Scand. 1973; 88,433-9.
105. Sleight, P. Cardiac vomiting. Br. Heart J. 1981; 46,5-7.
585
106. Van Lieshout, J.J. Cardiovascular reflexes in orthostatic
disorders. Amsterdam: Rodopi, 1989.
107. Cowley, A.W., Quillen, E.W. & Skelton, M.M. Role of
vasopressin in cardiovascular regulation. Fed. Proc. Fed.
Am. SOC.Exp. Biol. 1983; 42,3170-6.
108. Schwartz, T.W. Pancreatic polypeptide: a hormone under
vagal control. Gastroenterology 1983; 85,1411-25.
109. Krarup, T., Schwartz, T.W., Hilsted, J., Madsbag, S.,
Verlaege, 0. & Sestoft, L. Impaired response of pancreatic
polypeptide to hypoglycaemia: an early sign of autonomic
neuropathy. Br. Med. J. 1979; 2, 1544-6.
110. Ludbrook, J. & Rutter, P.C. Effect of naloxone on haemodynamic responses to acute blood loss in unanaesthetized
rabbits. J. Physiol. (London) 1988; 400, 1-14.
1 1 1 . Verrill, P.J. & Aellig, W.H. Vasovagal faint in the supine
position. Br. Med. J. 1970; iv, 348.
112. Day, S.C. & Eisenberg, J.M. Syncope. In: Griner, P.F.,
Panzer, J.R. & Greenland, P., eds. Clinical diagnosis and
the laboratory. Chicago: Year Book Medical Publishers
Inc., 1980: 107-22.
1 1 3. Abi-Samra, F., Maloney, J.D., Fouad-Tarazi, EM. & Castle,
L.W. The usefulness of head-up tilt testing and hemodynamic investigations in the workup of syncope of
unknown origin.Pace 1988; 11,1202-14.
114. Stevens, P.M. Cardiovascular dynamics during orthostasis
and the influence of intravascular instrumentation. Am. J.
Cardiol. 1966; 17,211-18.
1 1 5. Shvartz, E., Meroz, A., Magazanik, A., Shoenfeld, Y. &
Shapiro, Y. Exercise and heat orthostatism and the effect
of heat acclimation and physical fitness. Aviat. Space
Environ. Med. 1977; 48,836-42.
116. Graybiel, A. & McFarland, R.A. The use of the tilt-table
test in aviation medicine. Aviat. Med. 1941; 11, 194-21 1 .
117. Ogata, H., Iinuma, N., Nagashima, K. & Akabane, T.
Vasovagal reactions in blood donors. Transfusion 1980;
20,679-83.
118. Convertino, V.A., Doerr, D.F., Flores, J.F., Hoffler, G.W. &
Buchanan, P. Leg size and muscle functions associated
with leg compliance. J. Appl. Physiol. 1988; 64, 1017-21.
119. Dambrink, J.H.A., Imholz, B.P.M., Karemaker, J.M. &
Wieling, W. Circulatory adaptation to orthostatic stress in
healthy 10- 14-year-old children investigated in a general
practice. Clin. Sci. 1991; 81,51-8.
120. Lipsitz, L.A., Marks, E.R., Koestner, J., Jonsson, P.V. &
Wei, J.Y. Reduced susceptibility to syncope during
postural tilt in old age. Arch. Intern. Med. 1989; 149,
2709-12.
121. Imholz, B.P.M., Dambrink, J.H.A., Karemaker, J.M. &
Wieling, W. Orthostatic circulatory control in the elderly
evaluated by non-invasive continuous blood pressure
measurement. Clin. Sci. 1990; 79,73-9.
122. Wagner, H.N. Orthostatic hypotension. Bull. Johns
Hopkins Hosp. 1959; 105,322-59.
123. Van Lieshout, J.J., Wieling, W., Wesseling, K.M. & Karemaker, J.M. Pitfalls in the assessment of cardiovascular
reflexes in patients with sympathetic failure but intact vagal
control. Clin. Sci. 1989; 76,523-8.
124. Grant, R.T. & Reeve, E.B. Clinical observations on air-raid
casualties. Br. Med. J. 1941; 2,293-7 and 329-32.
125. McMichael, J. Clinical aspects of shock. J. Am. Med.
ASSOC.
1944; 124,275-81.
126. Shenkin, H.A., Cheney, R.H., Govons, S.R., Hardy, J.D. &
Fletcher, A.G. On the diagnosis of hemorrhage in man. A
study of volunteers bled large amounts. Am. J. Med. Sci.
1944; 208,421-36.
127. Jansen, R.P.S. Relative bradycardia: a sign of acute intraperitoneal bleeding. Aust. N. Z. J. Obstet. Gynaecol. 1978;
18,206-8.
128. Abelmann, W.H. Alterations in orthostatic tolerance after
myocardial infarction and in congestive heart failure.
Cardiology 1976; 61 (Suppl. l),236-48.
129. Bristow, M.R., Ginsburg, R., Minobe, W. et al. Decreased
catecholamine sensitivity and beta-adrenergic-receptor
J. J. van Lieshout et al.
586
density in failing human hearts. N. Engl. J. Med. 1982;
307,205-1 1.
130. Fleg, J.L. & Lakatta, E.G. Prevalence and significance of
postexercise hypotension in apparently healthy subjects.
Am. J. Cardiol. 1986; 57,1380-4.
131. Braunwald, E., Lambrew, C.T., Rockoff, S.D., Ross, J. &
Morrow, A.G. Idiopathic hypertrophic subaortic stenosis.
Circulation 1964; 30 (Suppl. 4), 1-1 15.
132. Mark, A.L., Kioschos, J.M., Abboud, EM., Heistad, D.D.
& Schmid, P.G. Abnormal vascular responses to exercise
in patients with aortic stenosis. J. Clin. Invest. 1973; 52,
1138-46.
133. Richards, A.M., Nicholls, M.G., Ikram, H., Hamilton, E.J.
& Richards, R.D. Syncope in aortic valvular stenosis.
Lancet 1984; ii, 1 1 13-6.
134. Manolis, AS., Linzer, M., Salem, D. & Estes, N.A.M.
Syncope: current diagnostic evaluation and management.
Ann. Intern. Med. 1990; 112,850-63.
135. Martin, GJ., Adams, S.L., Martin, H.G., Mathews, J., Zull,
D. & Scanlon, P.J. Prospective evaluation of syncope. Ann.
Emerg. Med. 1984; 13,499-504.
136. Noble, R.J. The patient with syncope. J. Am. Med. Assoc.
1977; 237,1372-6.
137. Sapire, D.W., Shah, J.J. & Black, I.F.S. Prolonged atrioventricular conduction in young children and adolescents.
S . Afr. Med. J. 1979; 55,669-73.
138. Sapire, D.W., Casta, A., Safley, W., ORiordan, A.C. &
Balsara, R.K. Vasovagal syncope in children requiring
pacemaker implantation. Am. Heart J. 1983; 106,
1406-1 1.
139. Ector, H., Verlinden, M., Vanden Eynde, E. et al. Bradycardia, ventricular pauses, syncope, and sports. Lancet
1984; ii, 591-4.
140. Mehta, A.V., O'Riordan, A.C., Sanchez, G.R., Balsara,
R.K. & Black, I.F.S. Mobitz Type I second degree atrioventricular block in children: clinical and electrophysiological findings and long-term follow-up. J. Am.
Coll. Cardiol. 1983; 1,613.
141. Schlesinger, Z., Barzilay, J., Stryjer, D. & Almog, C.H.
Life-threatening vagal reaction to emotional stimuli. Isr. J.
Med.Sci. 1977; 13,59-61.
142. Fujimura, O., Yee, R., Klein, G.J., Sharma, A.D. &
Boahene, K.A. The diagnostic sensitivity of electrophysiologic testing in patients with syncope caused by
transient bradycardia. N. Engl. J. Med. 1989; 321,
1703-7.
143. Gulamhusein, S., Naccarelli, G.V., KO, P.T. et al. Value and
limitations of clinical electrophysiological study in assessment of patients with unexplained syncope. Am. J. Med.
1982; 73,700-5.
144. Allen, S.C., Taylor, C.L. & Hall, V.E. A study of orthostatic
insufficiencv bv the tiltboard method. Am. J. Phvsiol.
1945; 143,-11-d20.
145. Kennv. R.A.. Bavliss. J.. Ingram. A. & Sutton. R. Head-uD
tilt: a useful test for investigating unexplained syncope.
Lancet 1986; i, 1352-4.
146. Fitzpatrick, A. & Sutton, R. Tilting towards a diagnosis in
recurrent unexplained syncope. Lancet 1989; i, 658-69.
147. Chen, M.Y., Goldenberg, I.F., Milstein, S . et al. Cardiac
electrophysiologic and hemodynamic correlates of
neurally mediated syncope. Am. J. Cardiol. 1989; 63,
66-72.
148. Strasberg, B., Rechavia, E., Sagie, A. et al. The head-up tilt
table test in patients with syncope of unknown origin. Am.
Heart J. 1989; 118,923-7.
149. Scott, A. Head-up tilt test for unexplained syncope. Lancet
1986; i, 169.
I
,
I
150. Imholz, B.P.M., Wieling, W., Langewouters, G.J. & Van
Montfrants, G.A. Continuous finger arterial pressure:
utility in the cardiovascular laboratory. Clin. Auton. Res.
1991; 1,43-53.
151. Almquist, A., Goldenberg, I.F., Milstein, S. et al. Provocation of bradycardia and hypotension by isoproterenol and
upright posture in patients with unexplained syncope. N.
Engl. J. Med. 1989; 320,346-51.
152. Waxman, M.B., Yao, L., Cameron, D.A., Wald, R.W. &
Roseman, J. Isoproterenol induction of vasodepressortype reaction in vasodepressor-prone persons. Am. J.
Cardiol. 1989; 63,58-65.
153. Piorry, P.A. Recherches sur I'influence de la pesanteur sur
le cours du sang; diagnostic de la syncope et de I'apoplexie;
cause et traitement de la syncope. Arch. GCn. MCd. 1826;
12,527-44.
154. Eagle, K.A., Black, H.R., Cook, E.F. & Goldman, L.
Evaluation of prognostic classifications for patients with
syncope. Am. J. Med. 1985; 79,455-60.
155. Smith, M.L., Hudson, D.L. & Raven, P.B. Effect of muscle
tension on the cardiovascular responses to lower body
negative pressure in man. Med. Sci. Sports Exercise. 1987;
19,436-42.
156. Ebert, T.J. & Barney, J.A. Physical fitness and orthostatic
tolerance. In: Smith, J.J., ed. Circulatory response to the
upright posture. Boca Raton: CRC Press, 1990: 47-63.
157. Frey, M.A.B. Considerations in prescribing preflight
aerobic exercise for astronauts. Aviat. Space Environ.
Med. 1987; 58,1014-23.
158. Convertino, V.A. Aerobic fitness, endurance training, and
orthostatic intolerance. Exercise Sport Sci. Rev. 1987; 15,
223-59.
159. Ten Harkel, A.D.J., Baisch, F., Becj, L. & Karemaker, J.M.
The autonomic nervous system in blood pressure regulation during 10 days 6" head down tilt. The Physiologist
1990; 33, S178-9.
160. Convertino, V.A., Sather, T.M., Goldwater, D.J. & Alford,
W.R. Aerobic fitness does not contribute to prediction of
orthostatic intolerance. Med. Sci. Sports Exercise 1986;
18,551-6.
161. Fitzpatrick, A., Travill, C.M., Vardas, P.E. ct al. Recurrent
symptoms after ventricular pacing in unexplained syncope.
Pace 1990; 13,619-24.
162. Kust, T., Lalonde, G., de Champlain, J. & Shenas, M.
Vasovagal syncope: management with atrioventricular
sequential pacing and beta-blockade. Can. J. Cardiol.
1989; 5,375-8.
163. Sander-Jensen, K., Mehlsen, J., Stadeager, C. et al.
Increased cholinergic activity during gravitation-induced
pre-syncope in man. The Physiologist 1986; S-74.
64. Dibner-Dunlap, M.E., Eckberg, D.L., Magid, N.M. &
Cintron-Trevino, N.M. T h e long-term increase of baseline
and reflexly augmented levels of human vagal-cardiac
nervous activity induced by scopolamine. Circulation
1985; 71,797-804.
65. Ferguson, D.W., Thames, M.D. 6r Mark, A.L. Effects of
propanolol on reflex vascular responses to orthostatic
stress in humans. Circulation 1983; 67, 802-7.
166. MacLean, A.R. & Allen, E.V. Orthostatic hypotension and
orthostatic tachycardia. Treatment with the head-up tilt
bed. J. Am. Med.Assoc. 1940; 115,2162-7.
167. Bannister, R. & Mathias, C. Management of postural
hypotension. In: Bannister, R., ed. Autonomic failure, a
textbook of clinical disorders of the autonomic nervous
system. 2nd ed. Oxford: Oxford University Press, 1988:
569-95.