International Journal of Cardiology, 37 (1992) l-5 1

0 1992 Elsevier Science Publishers B.V. All rights reserved 0167-5273/92/$05.00

CARD10 01528

Review

Mechanism of cyanotic spells in tetralogy of Fdlot

- the missing link?
Shyam Sunder Kothari
Cardiothoracic Centre, All India Institute of Medical Sciences, New Delhi, India
(Received 18 November 1991: revision accepted 17 May 1992)

Kothari SS. Mechanism of cyanotic spells in tetralogy of Fallot - the missing link? Int J Cardiol
1992;37:1-5.

The mechanism of cyanotic spells in patients with tetralogy of Fallot is not clear. Increases in
infundibular contractility or hyperpnoea have been considered as the key factors, but this explanation
appears inadequate. In this review, arguments are presented against these commonly held views.
Preliminary evidence is synthesised in favour of another more plausible hypothesis that cyanotic spells
may result from mechanoreceptor stimulation from the right ventricle. Increased contractility (due to
catecholamines) and decreased right ventricular size (due to various factors) can trigger a reflex resulting
in hyperventilation, some peripheral vasodilation without bradycardia, and this may initiate a spell. This
mechanism explains most of the precipitating events and many other issues about cyanotic spells more
satisfactorily.

Key words: Tetralogy of Fallot; Cyanosis; Syncope; Mechanoreceptor

Introduction ing, bowel movement or feeding can precipitate

The episodes of paroxysmal dyspnoea with
marked cyanosis (cyanotic spells) in infants and
young children with tetralogy of Fallot are com-
mon and serious [l-4]. The episodes may last
from a few minutes to several hours and the
marked arterial hypoxaemia during the spells may
result in brain damage or death, although usually
these spells are self limiting. The cyanotic spells
frequently occur without warning, although cry-
Correspondence tot Dr. S.S. Kothari, Cardiothoracic
tre, All India Institute of Medical Sciences, Ansari
New Delhi-l 10029, India.
them [l]. The spells increase in frequency during
summer months, and with infections [2]. Cardiac
catheterization [5], and supraventricular tachycar-
dia [6] are also known to precipitate these spells.
In fact, the circumstances precipitating the spells
have been so different and varied that it has been
difficult to envisage a common environmental
factor [2].
It is clear that these episodes result from dras-
tic reduction of pulmonary blood flow, with an
accompanying increase in the right-to-left shunt
and a drop in systemic arterial oxygen saturation.
Cen-
However, the precise mechanism of the spells has
Nagar, not been understood. Increase in infundibular
contractility [7], peripheral vasodilation [8], or
2
hyperventilation [93 have been incriminated as
the key factors.
Increase in infundibular contractility
Most workers now believe that increase in
contractility of right ventricular infundibular mus-
culature results in a decrease in pulmonary blood
flow, and other features of the cyanotic spells
[2-4,7,10]. The clinical findings of reduction in
intensity of murmur across the right ventricular
outflow during the spell appear to favour this
view. In fact, marked reduction in forward blood
flow to the lungs has been angiographically
demonstrated during a spell [lO,ll]. Johnson sug-
gested that [12] increase in the contractility may
result from increase in endogenous nore-
pinephrine (which is possible during most of the
precipitating events listed). The salutary effects
of beta-blockers in the patients with spells also
support this view [ll].
The occurrence of spells, albeit rare in pa-
tients with pulmonary atresia and ventricular sep-
tal defect, casts doubt on the importance of in-
fundibular contractility as the sole mechanism [7].
Furthermore, as discussed below, peripheral va-
sodilation in patients with tetralogy of Fallot may
also reduce the intensity of the murmur and
blood flow to the lungs. Increase in cate-
cholamines has not been demonstrated preceding
the spells. Although an increase in infundibular
obstruction may appear a simple explanation of
the spells, it remains to be clarified why every-day
activities like crying, feeding etc. only occasion-
ally result in spells. By itself, an increase in con-
tractility due to endogenous catecholamines does
not seem to explain completely the dramatic clin-
ical presentation of these patients.
Peripheral vasodilation
Peripheral vasodilation in patients with tetral-
ogy of Fallot would increase shunting of the right
ventricular blood to the aorta through ventricular
septal defect, and decrease pulmonary blood flow.
Previously suggested by Hamilton [8] as an impor-
tant reason for the cyanotic spells, subsequent
workers did not consider it important as there
was no fall in systolic blood pressure during the
spells. The diastolic pressure was not low, nor did
the pulse pressure widen [4,5,7]. Thus, marked
vasodilation does not accompany the spells. How-
ever, peripheral vasodilation with amyl nitrite in
the patients with tetralogy of Fallot has been
shown to result in a marked reduction in the
murmur and increase in cyanosis with relatively
less hypotension [13]. The question of peripheral
vasodilation in cyanotic spells does not seem to
have been adequately investigated.
Hyperventilation
Hyperpnoea increases oxygen demand and
cardiac output [14]. In patients with tetralogy of
Fallot, hyperpnoea results in an increase in arte-
rial desaturation due to an increase in right-to-left
shunt in the face of increased cardiac output but
a relatively fixed pulmonary blood flow [9]. Gun-
theroth et al. suggest that hyperpnoea may be the
important factor initiating and maintaining the
cyanotic spells [93. To account for the frequent
occurrence of spells in the morning hours of the
day after a good night’s sleep [3], they propose
that the respiratory drive mechanism may be more
sensitive to changes in blood gases when the
patients wake up in a reasonable metabolic status
after sleep. During such times a sudden decrease
in arterial p0, and pH and an increase in arterial
pC0, induced by a relatively sudden increase in
activity, or a Valsalva-like manoeuvre (such as
crying or bowel movement) may trigger marked
hyperpnoea. The hyperpnoea will increase car-
diac output and decrease pulmonary blood flow
resulting in further right-to-left shunt and greater
arterial hypoxaemia. A vicious cycle may thus be
established resulting in the spell. They also state
that in most of these patients on most days, there
is either an adjustment in sensitivity of respira-
tory drive mechanism or an adjustment in bicar-
bonate, which prevents the occurrence of spells
[93. This elusive sequence of events has never
been proved or disproved. Furthermore, it does
not explain the fact that spells indeed occur at
any time of the day [4]. Moreover, it may be
relevant to note that the sensitivity of the respira-
tory centre to hypoxia is known to be blunted in
patients with cyanotic heart disease [15], and is
not restored even years after corrective surgery
1161.
In the absence of any better alternative the
scheme suggested by Guntheroth et al. has pro-
vided a good working hypothesis up to now. How-
ever, I believe that the following hypothesis may
explain the mechanism of spells and the events
precipitating these spells more satisfactorily.
Hypothesis
The cyanotic spells in patients with tetralogy of
Fallot may result from the stimulation of right
ventricular mechanoreceptors. Increased contrac-
tility (due to endogenous catecholamines), and a
decrease in right ventricular cavity size (such as
can occur with Valsalva-like manoeuvre) may
trigger a reflex response resulting in hyperventila-
tion, some peripheral vasodilation, without brady-
cardia. The hyperpnoea may perpetuate a vicious
cycle, as discussed previously. It is well known
now that a combination of decreased left ventric-
ular volume and raised sympathetic tone may
stimulate left ventricular mechanoreceptors that
may trigger marked peripheral vasodilation and
bradycardia (vasovagal syncope) [17]. The ventric-
ular receptors are a heterogeneous population of
receptor subgroups with varying thresholds of ac-
tivation, and varying reflex effects [18]. Response
to some vagal afferents may not alter heart rate
1181,but may cause hyperventilation and periph-
eral vasodilation. The response to left ventricular
and right ventricular receptor stimulation may
also differ [ 191. Experimental mechanoreceptor
stimulation from the right ventricle in sympathec-
tomised dogs has been reported to cause hyper-
ventilation and hypotension without bradycardia
[20]. Moreover, in some of the head-up tilt table
studies done to provoke left ventricular mechano-
receptors and vasovagal response, a few patients
develop marked hyperventilation without brady-
cardia or hypotension 117,211. Although the au-
thors did not comment on this, it may be specu-
lated that receptors from the right ventricle
elicited such a response. Such studies in patients
with tetralogy of Fallot would be interesting. In-
deed, in 1953 while studying the effects of pos-
ture on saturation in patients with tetralogy of
Fallot, Lurie reported precipitating hyperventila-
tion and syncope (spell) in a patient during a tilt
to 70” [22]. It is recognised that the right ventricle
is only sparsely innervated [181, and more recent
experimental studies indicate no important role
of right ventricle mechanoreceptors in Gardiac or
respiratory control normally [231. However, the
situation in tetralogy of Fallot wherein the
anatomy is altered, may be different from nor-
mals. The hypertrophied ventricular walls, ro-
tated conal septum, or other factors may be im-
portant. How the increase in catecholamines and
decrease in the cavity size in this distinct
pathoanatomy of tetralogy of Fallot predispose
the receptors to trigger responses remains to be
established. It may be relevant that amongst the
patients with tetralogy of Fallot, a slightly differ-
ent anatomy is seen from that in patients who
present with frequent cyanotic spells [24].
The relationship of precipitating events seems
clearer with the proposed mechanisms.
(a> Crying, defaecation, suprauentricular tachy-
cardia and feeding: It is easy to envisage in-
creased contractility due to catecholamines and
decreased cavity size of the right ventricle during
crying, bowel movement (due to Valsalva-like ma-
noeuvre) or supraventricular tachycardia (due to
shortened diastole) [25] that can induce a cyan-
otic spell. Similarly feeding also induces increase
in cardiac output and heart rate and decrease in
cavity size [26].
(b) Hot weather, infections: Peripheral vasodi-
lation during summer, or with infection [27] facili-
tates better emptying of right ventricle into aorta.
This could reduce the cavity size and may in-
crease the susceptibility to cyanotic spells.
Cc) Cardiac catheterization : Direct mechanical
stimulation of the receptors by a cardiac catheter
151may induce a cyanotic spell. Dehydration due
to overnight fasting (which may decrease the right
ventricular size), or peripheral vasodilation in-
duced by premeditation may also be contributory.
Cd) Time of the day: The more frequent occur-
rence of spells during the morning hours may
relate to catecholamine secretion on awakening
or to circadian variations in the endogenous cate-
cholamines secretions [2X].
4

(e) Cyanotic spells in pulmonary atresia: The References

reflexes originating from the hypertrophied ven-
tricles in the patients with pulmonary atresia and 1 Taussig HB. Tetralogy of Fallot: Especially the care of the
ventricular septal defect can trigger hyperventila- cyanotic infant and child. Pediatrics 1948;1:307-314.
2 Keith JD. Rowe RD, Vlad P. Heart disease in Infancy and
tion, vasodilation and the sequence of the spells.
Childhood. New York: Macmillan 1958;401.
Blood flow to the lungs through collaterals or 3 Nadas AS. Pediatric cardiology. 2nd ed. Philadelphia: W.B.
patent ductus arteriosus would be reduced with Saunders, 1963;353.
peripheral dilation, and hyperventilation would 4 Gasul BM, Arsilla RA, Lev M. Heart disease in children.
increase the metabolic demands. Similar consid- Diagnosis and treatment. London, Philadelphia: Pitman
Medical Publishing Co., 1968;577.
erations may apply to other cyanotic heart dis-
5 Braudo JL, Zion MM. Cyanotic spells and loss of con-
eases where spell occurs. sciousness induced by cardiac catheterization in patients
(f) The effects of beta-blockers and morphine: with Fallot’s tetralogy. Am Heart J 1960;59:10-18.
Beta-blockers are useful in the treatment and 6 Young D, Elbl F. Supraventricular tachycardia as cause of
prevention of cyanotic spells because they de- cyanotic syncopal attacks in tetralogy of Fallot. N Engl J
Med 1971;284:1359-1360.
crease myocardial contractility. Akin to the use of
7 Wood P. Attacks of deeper cyanosis and loss of conscious-
beta-blockers in vasovagal syncope [29], the de- ness (syncope) in Fallot’s tetralogy. Br Heart J 1958;
crease in contractility may also be important in 20:282-286.
preventing mechanoreceptor stimulation and 8 Hamilton WF, Winslow JA, Hamilton WF, Jr. Notes on a
hence the spells. Morphine is very effective in the case of congenital heart disease with cyanotic episodes. J
Clin Invest 1950;29:20.
treatment of cyanotic spells [l], but its mechanism
9 Guntheroth WG, Morgan BC. Mullins GL. Physiologic
of action is not clear. It may be relevant that studies of paroxysmal hyperpnea in cyanotic congenital
endogenous opioids inhibit the sympathetic sys- heart disease. Circulation 1965;31:70-6.
tem, and more recently have been implicated in 10 Anderson RH, McCartney FJ, Shinebourne EA. Tynan M.
pathogenesis of vasodepressor syncope [30]. Pediatric cardiology. Edinburgh, London, New York:
Churchill Livingstone 1987;777.
(g) Age of the patients: The cyanotic spells
11 Honey M, Chamberlain DA, Howard J. The effect of beta
generally start from 2 months of age and mysteri- sympathetic blockade on arterial oxygen saturation in Fal-
ously disappear after about 2 yr. Although en- lot’s tetralogy. Circulation 1964;30:501-510.
tirely speculative, it may be related to growth of 12 Johnson AM. Norepinephrine and cyanotic attacks in Fal-
ventricular cavity and infundibulum such that it lot’s tetralogy. Br Heart J 1961;23:197-202.
13 Vogelpoel L, Schirire V, Nellen M, Swanepoel A. The use
results in decreased susceptibility of receptors to
of amyl nitrite in the differentiation of Fallot’s tetralogy
stimulation. and pulmonary stenosis with intact ventricular septum.
Although the hypothesis explains many facts as Am Heart J 1959;57:803-19.
listed above, it remains to be investigated if the 14 Van Lingen B, Whidborne J. Oximetry in congenital heart
right ventricle in patients with tetralogy of Fallot disease with special reference to the effects of voluntary
hyperventilation. Circulation 1952;6:740-748.
has a higher receptor density, and that such re-
15 Edelman NH, Cherniack NS, Lahiri S, Brando L, Fishman
flexes do indeed occur. The cardiac mechanore- AP. The ventilatory response to hypoxia in cyanotic con-
flexes are involved in the genesis of many clinical genital heart disease. Clin Res 1968;16:369.
conditions including vasovagal syncope, syncope 16 Sorenson SC, Severinghaus JW. Respiratory insensitivity
in patients with aortic stenosis, hypotension dur- to acute hypoxia persisting after correction of tetralogy of
Fallot. J Appl Physiol 1968;25:221-223.
ing exercise, bradycardia with vasodilators etc.
17 Epstein SE, Stampfer M, Beiser GD. Role of capacitance
[31]. Cyanotic spells in tetralogy of Fallot may be and resistance vessels in vasovagal syncope. Circulation
added to this list in the future. 1968;37:524-533.
18 Thoren P. Role of cardiac vagal C fibres in cardiovascular
control. Rev Physiol Biochem Pharmacol 1979;86:1-94.
Acknowledgement 19 Ross J Jr, Frahm CJ, Braunwald E. Influence of intracar-
diac baroreceptors on venons return, systemic vascular
volume and peripheral resistance. J Clin Invest 1961:
I thank Dr. Dinesh Verma, FRCS (Ophth.) for 40:563-72.
critical review of the manuscript. 20 Kostreva DR. Hopp FA, Zupkerku EJ. Kampine JP. Ap-
 nea. tachypnea and hypotension elicited by cardiac vagal 26 Ehrlich W, Tosheff J. Caldini P, Abbey H, Brady JV.
afferens. J Appl Physiol 1979;47:312-318. Adaptation of cardiac output, coronary flow and other
21 Strasberg B, Recharia E. Sagie A et al. The head up tilt circulatory functions in dogs to exercise and to eating.
table test in patients with syncope of unknown origin. Am Johns Hopkins Med J 1972;130:216-234.
Heart J 1989;118:923-927. 27 Thauer R. Circulatory adjustments to climatic require-
22 Lurie PR. Postural effects in tetralogy of Fallot. Am J ments. In: Hamilton WF. Dow P, eds. Handbook of physi-
Med 1953;15:297-306. ology. Section 2. Circulation volume III. Washington, DC:
23 Crisp AJ. Hainsworth R. Tutt SM. The absence of cardio- American Physiological Society 1965;1941-1947.
vascular and respiratory responses to changes in right 28 Turton MB, Deagn T. Circadian variations in plasma
ventricular pressure in anaesthetised dogs. J Physiol catecholamines. cortisol and immunoreactive insulin con-
(London) 1988;407:1-13. centrations in supine subjects. Clin Chim Acta 1974;55:389.
24 Shinebourne EA, Anderson RH, Bowyer JJ. Variation in 29 Goldenberg IF, Almquist A. Dunbar DN. Milstein S.
clinical presentation of tetralogy of Fallot in infancy. An- Pritzker MR, Benditt DC. Prevention of neurally medi-
giographic and pathogenetic implications. Br Heart J ated syncope by selective Bl adrenoreceptor blockade
1975;37:946-955. (abstract). Circulation 1987;76 (suppl IV):l33.
25 Sonnenblick EH, Braunwald E, William JF Jr, Glick G. 30 Perna GP, Ficola U, Salvatori MP et al. Increase of
Effects of exercise on myocardial force velocity relation in plasma beta endorphins in vasodepressor syncope. Am J
intact unanesthetised man. Relative roles of changes in Cardiol 1990;65:929-930.
heart rate, sympathetic activity and ventricular dimen- 31 Abboud FM. Ventricular syncope. Is the heart a sensory
sions J Clin Invest 1965:44;2051-2062. organ? N Engl J Med 1989:320:390-392.