Download Contrasting Haemodynamic Effects

Clinical Science and Molecular Medicine (1976) 51, 575s-578s.
Vasodilating drugs :contrasting haemodynamic effects
R . C . T A R A Z I , H. P. D U S T A N , E. L. B R A V O A N D A . P. N I A R C H O S
Research Division, The Cleveland Clinic Foundation, and
The Cleveland Clinic Educational Foundation, Cleveland, Ohio, U.S.A.
Summary
1. We investigated the haemodynamic effects of
intravenously administered hydrallazine, diazoxide
and nitroprusside and orally administered minoxidil
to determine whether vasodilators (such as nitroprusside) which d o not increase cardiac output might
be better treatment for hypertensive complications
associated with, or caused by, myocardial failure
than those that do.
2. Hydrallazine and diazoxide caused increases in
heart rate, cardiac output, cardiopulmonary blood
volume, the ratio of cardiac output to cardiopulmonary volume, and pulmonary artery pressure.
Nitroprusside, although decreasing pressure and
vascular resistance, caused no significant change in
the other functions except for reducing pulmonary
artery pressure. Minoxidil, when given orally, had
the potential for causing pulmonary hypertension.
This seemed explained by increased flow (hyperdynamic type) in some but by congestive cardiac
failure in others; the latter condition was probably
intensified by the marked fluid retention that the
drug can cause.
3. On the basis of these results a classification of
vasodilators was constructed which depends on the
presence or absence of a venodilating effect. Vasodilators which produce no (or little) venodilatation,
increase heart rate, cardiac output, cardiopulmonary
blood volume and pulmonary artery pressure. In this
class are diazoxide, hydrallazine and minoxidil.
Those that cause venodilatation d o not stimulate the
heart nor d o they cause pulmonary hypertension.
Nitroprusside and nitroglycerine are drugs of this
type.
Correspondence: Dr Robert C. Tarazi. Research Division,
Cleveland Clinic, 9500 Euclid Avenue, Cleveland, Ohio
44106,
U.S.A.
4. These results suggest that drugs producing both
venodilatation and arteriolar dilatation may be more
specific therapy for hypertensive complications
associated with cardiac failure than those that cause
only arteriolar dilatation.
Key words: anti-hypertensive drugs, pulmonary
hypertension, vasodilators.
Introduction
Development of vasodilating drugs that are orally
active has long been considered a therapeutic goal in
the continued search for specific treatment of hypertension. Such drugs have been shown also to be
useful in normotensive cardiac failure (Franciosa,
Guiha, Limas, Rodriguera & Cohn, 1972; Cohn,
Mathew, Franciosa & Snow, 1974). However, those
that have been reported on have little effect on
cardiac output and heart rate (Chatterjee, Parmley,
Ganz, Forrester, Walinsky, Crexells & Swan, 1973;
Kovick, Tillisch, Berens, Bramowitz & Shine, 1976),
in contrast to the vasodilators that are traditionally
used for treatment of hypertension (Freis, Rose,
Higgins, Finnerty, Kelley & Partenope, 1953;
Wilson & Okun, 1963). These differences may result
from differences in cardiac function or may represent
variations in pharmacological action. In the latter
case, these variations might determine the usefulness
of a particular vasodilator in the therapy of hypertension, depending on the degree of cardiac impairment present.
We have therefore investigated the haemodynamic
effects of hydrallazine, diazoxide and nitroprusside
given intravenously and minoxidil given orally in
a group of patients with hypertension of various
degrees of severity but without cardiac decompensa-
R. C. Tarazi et al.
576s
tion. The results suggest that vasodilators that produce venodilatation in addition to arteriolar dilatation are better suited for treatment of hypertension
with significant hypertensive heart disease, than are
those that primarily cause arteriolar dilatation.
Methods
Studies were carried out in forty-five hypertensive
patients; thirteen were given diazoxide, and seven
hydrallazine, by single intravenous injection; twelve
were given an infusion of nitroprusside sufficient to
reduce arterial blood pressure by at least 20 mmHg.
'Thirteen patients were treated with minoxidil, either
alone or in addition to other oral anti-hypertensive
agents. Most of the patients investigated were either
untreated or had discontinued all forms of treatment
at least a month before the investigation. None
of the patients receiving nitroprusside had other
medication whereas five of the thirteen patients given
diazoxide as well as two of the seven given hydrallazine were on propranolol therapy. The results
during propranolol treatment, however, did not
differ from those obtained in untreated patients so
are included here. The collaboration of each patient
was obtained by carefully explaining the nature of
the study and its importance both individually and
generally.
Haemodynamic measurements were performed in
the morning in the fasting state without premedication. After local anaesthesia, an arterial catheter
was introduced percutaneously and advanced to the
ascending aorta, and a Swan-Ganz venous catheter
was advanced into the pulmonary artery. Cardiac
output (CO) and cardiopulmonary volume (CPV)
were calculated from dye-dilution curves (Indocyanine Green) from classical formulae based on the
Stewart Hamilton technique. The ratio of cardiac
output to cardiopulmonary blood volume (COICPV)
was taken as an index of cardiac performance. The
methods used to determine flow, pressure and heart
rate and to calculate derived indices have been
described in detail (Tarazi, Ibrahim, Dustan &
Ferrario, 1974).
In studies with intravenously administered drugs,
control measurements were obtained in triplicate,
after which seven patients were given 20 mg of
hydrallazine and thirteen were given 300 mg of
diazoxide as single injections. Seven patients received
an infusion of sodium nitroprusside sufficient to
reduce mean arterial pressure from an average of 136
to 114 mmHg. In each instance haemodynamic
measurements were repeated after the reduction of
arterial pressure. For the thirteen patients treated
with minoxidil, control measurements were made
before the drug was given and then repeated days or
weeks after treatment had been started. A total of
thirty-three studies was obtained, one per patient in
six and a median of four per patient in the other
seven. Four patients received minoxidil alone as
initial therapy; in the remaining nine, minoxidil
was added to diuretics or sympathetic-blocking
agents. The doses of minoxidil ranged from 10 to
40 mglday.
Results
The haemodynamic effects of the three intravenously
administered vasodilators are summarized in Table
1. Each reduced systemic arterial pressure and total
TABLE
1. Haemodynamic effects of different vasodilators
P: statistical significance based on paired t-tests for the number (n) of patients shown. Abbreviations: MAP = mean
arterial pressure; CO = cardiac output; TPR = total peripheral resistance; PAP = pulmonary artery pressure;
CPV = cardiopulmonary blood volume.
Hydrallazine ( n = 7)
Heart rate (min-')
MAP (mmHg)
CO (I/min)
TPR (u/m2)
PAP (mmHg)
CPV (ml)
COlCPV
Nitroprusside (n = 12)
Before
After
P
Before
70
120
4.48
47
14.9
818
5.5
90
100
6.34
28
18.8
913
6.9
<0.001
<0405
<0.001
<0.005
<0*02
<0*05
<0-001
78
136
5.10
53
12.7
994
4.9
After
99
114
5.29
42
8.8
975
5.2
P
<om1
Diazoxide ( n = 13)
Before
ns
70
120
4-15
61
14.6
798
ns
5.4
<0*001
ns
<0*001
<0.01
After
91
93
5.71
34
18.5
872
6.7
P
<ow1
<o001
<0401
<0.001
<0*01
<0*005
<0*001
Haemdynamics of vasodilators
peripheral resistance and increased heart rate significantly. With sodium nitroprusside these changes
were not accompanied by consistent modification in
CO,") CPV or the CO/CPV ratio; pulmonary arterial
pressure fell significantly in all thirteen patients. In
contrast, with both hydrallazine and diazoxide, CO,
CPV, COlCPV and pulmonary artery pressure rose
significantly.
Minoxidil had effects similar to those of diazoxide
and hydrallazine but, as treatment continued, these
were modified by fluid retention as well as by
changes in cardiac function (Dustan, Tarazi &
Bravo, 1975). Pulmonary hypertension was encountered frequently but did not occur invariably: thus
pulmonary arterial pressure > 30/15 mmHg was
found in eight of thirteen patients (it occurred in
three of six who had only one haemodynamic
investigation, and in five of seven with repeated
haemodynamic investigations). The haemodynamic
pattern associated with the rise in pulmonary
arterial pressure differed among the seven patients;
in four, pulmonary hypertension was associated with
marked increases in CO and responded to the addition of propranolol therapy. In three patients, however, the rise of pulmonary arterial pressure was
accompanied by a decline in CO from the value
obtained before pulmonary hypertension occurred.
These patients also showed more blood volume
expansion and greater rises in pulmonary wedge
pressure than did the four with sustained increases in
co.
Discussion
The haemodynamic pattern produced by hydrallazine and diazoxide may be best explained by
predominant arteriolar dilatation with little or no
venodilatation; under these conditions arteriolar
dilatation leads to a more rapid transfer of blood to
the venous side, increased venous return, central
location of blood and increased cardiac output. In
this regard our results are in agreement with previous
studies showing that neither diazoxide (Thirlwell &
Zsoter, 1972) nor hydrallazine (Nickerson, 1970) has
an important effect on veins. The pulmonary hypertension we observed is mostly hyperkinetic in
nature, reflecting increased output from the right
ventricle and implying little o r no pulmonary
dilatation. Further, the rise in pulmonary arterial
('1 Abbreviations: CO, cardiac output; CPV. cardiopulmonary volume.
577s
pressure is expected to be more accentuated in case
of diminished left ventricular compliance, such as
occurs in hypertension, or in the presence of a fixed
obstruction (mitral stenosis) to pulmonary venous
return (Aitchison, Cranston & Priest, 1955). N o
consistent changes were found in pulmonary
vascular resistance, but then none of the patients had
a markedly elevated resistance. A reduction in
pulmonary vascular resistance has been reported
only in few patients with initially high resistance
(Just & Stein, 1969), and it is conceivable that a
mild pulmonary vasodilating effect may be amplified
under those conditions.
As indicated earlier, effects of hydrallazine and
diazoxide were not modified by pretreatment with
propranolol. Reasons for this are not apparent but
since a-adrenoreceptors are left unopposed by such
treatment, any reflex sympathetic stimulation that
results from fall in arterial pressure produced by
vasodilatation may cause further decrease in venous
capacitance, a somewhat greater distribution of
blood centrally and tachycardia similar to that
observed with volume expansion (Vatner, Boettcher,
Heyndrick & McRitchie, 1975).
Pulmonary hypertension occurred frequently
during minoxidil therapy. Two haemodynamic
patterns were associated with the rise in pulmonary
arterial pressure and neither suggested a paradoxical
vasoconstrictor effect of minoxidil on the pulmonary
circulation. The first, termed 'congestive pulmonary
hypertension', seemed to reflect diminished cardiac
efficiency, probably because of marked fluid retention and increased preload. It was associated with
marked increase in pulmonary wedge pressure and
expansion of total blood volume. Sequential studies
showed that cardiac output decreased as pulmonary
artery pressure increased; although output was
higher at that time than before minoxidil treatment,
it was lower than the value obtained when pulmonary artery pressure was normal. In contrast, the
'hyperkinetic pulmonary hypertension' was less
pronounced and was associated with lesser rise in
pulmonary wedge pressure and smaller expansion in
total blood volume. It was characterized by marked
increase in cardiac output and was usually responsive
to propranolol therapy. A typical example is given
by a patient in whom pulmonary arterial pressure
rose from 28/15 to 40/20mmHg as cardiac index was
doubled (2-8 to 5.6 1 min-a m-z) and mean arterial
pressure fell only from 192 to 159 mmHg. The
addition of propranolol reduced cardiac index (4
5789
R. C. Tarazi et al.
1 min-' m-2), controlled mean arterial pressure (1 16
circulation in mitral disease. British Heart fournal. 17,
mmHg) and returned pulmonary arterial pressure to
normal (23/13 mmHg). Thus exact haemodynamic
characterization of the pulmonary hypertension
allowed a rational decision in the adjustment of the
therapy.
These results indicate that vasodilator agents
have different haemodynamic patterns that may be
important in the choice of therapy in different hypertensiveconditions. Our results do not suggest that the
increase in CPV is the basic mechanism responsible
for the rise in pulmonary arterial resistance; the
latter probably reflects the increased pulmonary
flow. However, changes in CPV are a marker
of the degree of central distribution of blood and of
the systemic effects of venodilatation. A reduction in
pulmonary arterial pressure with nitroprusside
irrespective of changes in cardiac output suggests an
additional direct pulmonary vasodilator effect; it is
interesting that the vasodilators that cause venodilatation Seem also to affect the pulmonary circulation similarly.
CHATTERJEE,
K., PARMLEY.
W.W., GANZ,W., FORRESTER,
J.,
WALINSKY,P., CREXELLS, C. & SWAN, H.J.C. (1973)
Hemodynamic and metabolic responses to vasodilator
therapy in acute myocardial infarction. Circulation, 48,
Acknowledgments
This study was supported in part by grants from the
National Heart, Lung and Blood Institute (HL
6835) and the U. A. Whitaker Fund.
References
AITCHISON,
J.D., CRANSTON,
W.I. & PRIEST,E.A. (1955) The
effects of I-hydrazinophthalazine on the pulmonary
425-430.
1183-1 193.
COHN,J.N., MATHEW,K.J., FRANCIOSA,
J.A. &SNOW,J.A.
(1974) Chronic vasodilator therapy in the management of
cardiogenic shock and intractable left ventricular failure.
Annals of Internal Medicine, 81, 777-780.
DUSTAN,H.P., TARAZI,
R.C. & BRAVO,E.L. (1975) In:
Recent Advances in Hypertension, pp. 247-258. Ed.
Milliez, P. & Safar, M. Boehringer Ingelheim. Reims.
FRANCIOSA,
J.A., GUIHA,N.H., LIMAS,C.J., RODRIGUERA,
E.
& COHN,
J.B. (1972) Improved left ventricular function
during nitroprusside infusion in acute myocardial infarction. Lancet, i, 65C654.
FREIS,
E.D., ROSE,J.C., HIGGINS,T.F., FINNERTY,
F.A., JR.
KELLEY,R.T. & PARTENOPE,E.A. (1953) The hemodynamic effects of hypotensive drugs in man. IV. I Hydrazinophthalazine. Circulation, 8, 199-203.
JUST, V.H. & STEIN, U. (1969) Untersuchungen iiber die
Wirkung von Diazoxid, einer neuen blutdrucksenkenden
Substanz, auf den grossen und den kleinen Kreislauf beim
Hypertoniker. Zeitschrvt fur Kreislaufforschung,58, 925936.
KOVICK,R.B., TILLISCH,
J.H., BERENS,S.C., BRAMOWITZ,
A.D. & SHINE,K.I. (1976) Vasodilator therapy for chronic
left ventricular failure. Circulation, 53, 222-328.
NICKERSON,
M. (1970) In: The Pharmacological Basis of
Therapeutics, p. 729. Ed. Goodman, L.S. & Gilman, A.
Macmillan, New York.
TARAZI,
R.C., IBRAHIM,M.M., DUSTAN.H.P. & FERRARIO,
C.M. (1974) Cardiac factors in hypertension. Circulation
Research, 34 (Suppl. I), 213-221.
THIRLWELL,
M.P. & Z~OTER,
T.T. (1972) The effect of diazoxide on the veins. American Heartfournal, 83,512-517.
VATNER, S.F., BOETTCHER,D.H., HEYNDRICK,G.R. &
MCRITCHIE,R.J. (1975) Reduced baroreflex sensitivity
with volume loading in conscious dogs. Circulation
Research, 37, 236242.
WILSON,W.R. & OKUN,R. (1963) The acute hemodynamic
effects of diazoxide in man. Circulation, 28, 89-93.