Download Effects of digitalis on the normal and the failing heart

51A
JACC Vol. 5, No.5
May 1985:5 1A- 59A
Effects of Digitalis on the Normal and the Failing Heart
EUGENE BRAUNWALD, MD, FACC
Boston , Massachusetts
Although it has been known for more than a century
that digitalis glycosidesexert a powerful beneficialeffect
on patients with heart failure, atrial fibrillation and a
rapid ventricular rate, it was believed for many years
that the drug exerts this clinical effect primarily by slowing the heart rate. It was also thought that the extracardiac vascular actions of digitalis might be responsible
for its therapeutic effect. It has now been established
that cardiac glycosidescause arteriolar and venous constriction in a variety of mammalian species including
human beings, and that this vasoconstriction involves
the coronary vascular bed as well, but it is believed that
these actions are not responsible for any beneficial clinical effect. A variety of investigations on cardiac muscle
in vitro, anesthetized and conscious dogs and anesthetized and conscious human subjects have shown that
cardiac glycosides improve the contractility of failing
mammalian myocardium. It has become clear that digitalis also stimulates the contractility of the nonfailing
heart. The degree of augmentation of contractility induced by cardiac glycosides is related inversely to the
baseline contractile state. Myocardial oxygen consumption, which is increased in the normal heart by the positive inotropic action of glycosides, is actually reduced
or remains constant in the failing heart. Cardiac glycosides increase the contractility of the globally ischemic
William Withering (1) was well aware that digitalis is effective only in certain forms of dropsy , but because edema
had not been clearly identified as a cardinal manifestation
of heart failure two centuries ago, he did not associate this
clinical effect with the cardiac actions of the drug. However,
Withering did recognize that the drug causes an improvement in the pulse, and he stated that digitalis " has a power
over the motion of the heart with a degree yet unobserved
in any other medicine" (1) . Fourteen years later, John Ferriar correctly ascribed a primary effect of digitalis on the
heart and assigned secondary importance to its diuretic actions (2). Subsequently , on the basis of auscultatory studies,
From the Departments of Medicine. Harvard Medical School, Brigham
and Women 's Hospital and Beth Israel Hospital, Boston, Massachusetts.
Address for reprints: Eugene Braunwald , MD, 75 Francis Street, Boston, Massachusetts 02115.
© 1985 by the American College of Cardiology
heart, but their actions in chronic ischemic heart disease
with regional impairment of function are complex. Deterioration of segmental performance occurs in ischemic
and necrotic segments, while improvement of contractility occurs in adjacent normal segments. A modest improvement in left ventricular pump performance occurs
in patients with chronic ischemic heart disease with impairment of global left ventricular function.
There has been considerable debate concerning the
clinical value of digitalis therapy in heart failure. Simple
withdrawal of digitalis resulted in no clinical deterioration in many patients with mild to moderate heart
failure. However, in patients with more severe impairment of cardiac function, aggravation of heart failure
occurred after digitalis withdrawal, and striking clinical
improvement was noted when the drug was readministeredo Thus, cardiac glycosides are relatively weak inotropic agents, have a low therapeutic ratio and are moderately effectivein heart failure secondary to ventricular
volume overload. However, when properly used in patients with severe heart failure and sinus rhythm, particularly in patients with atrial fibrillation, cardiac glycosides improve both cardiac performance and the clinical
state and can be administered effectively for years.
Am Coli CardioI1985;5:51A-59A)
a
Bouillard (3) reported that a rapid and irregular pulse was
improved by digitalis, which he considered to be "opium"
for the heart. By the latter half of the 19th century , it was
well recognized that digitalis enhances ventricular contraction (4). It had also become clear that digitalis glycosides
exert a powerful beneficial clinical effect on patients with
heart failure who had atrial fibrillation and a rapid ventricular
rate, but it was believed that the drug exerted its clinical
effect in such patients primarily by slowing the heart rate.
Actions on the Vascular Bed
Normal subjects. Early in this century it was suggested
that a principal circulatory action of digitalis might be extracardiac, and that the direct effects of the drug on the
vascular bed might be mainly responsible for its therapeutic
action (5). More recently it was established that cardiac
glycosides do exert a distinct direct action on the vascular
0735-1097/85/$3.30
BRAUNWALD
CLINICAL EFFECTS OF DIGITALIS
S'lA
lACC Vol. 5, No. 5
May 1985:5IA-59A
bed; they cause arteriolar and venous constriction in anesthetized dogs (6,7) and human subjects (8) (Fig. l) as well
as in conscious normal humansubjects (9). In the conscious
dog ouabain causes markedelevationof coronary and total
systemic vascular resistance (10). Although cardiac glycosides have a local direct action on vascular smooth muscle, they also act indirectly by causing neurogenically mediated coronary vasoconstriction (2).
Patients with heart failure. In patients with congestive
heart failure, baseline levels of total cardiac output and
forearm blood flow are lower, whereas total and forearm
vascularresistances and venous tone are higher than in normal subjects (9). In such patients, the effect of ouabain on
the vascular bed is directionally opposite to that observed
in normal subjects; that is, it causes reductions of total
systemic forearm vascular resistance and forearm venous
tone (9). It is likelythat these vasodilator actionsare indirect
and result from the reduction of reflex sympathetically mediated arteriolar and venoconstriction that occurs as a consequenceof the elevation of cardiac output induced by digitalis in patients with heart failure.
It now appears that rather than being responsible for the
beneficial effects of these drugs, the systemic and coronary
vascularconstriction producedby rapidly acting glycosides
given intravenously mayactually be responsible for a variety
of complications (11,12) including acute exacerbation of
heart failure. The vasoconstrictor effect of digitalis has a
morerapid onset than its inotropic action(8). Thus, to avoid
systemic and coronary vasoconstriction when rapidly acting
glycosides are indicated, intravenous administration as a
bolus should generally be avoided, particularly in patients
with ischemic heart disease.
Figure 1. Recording of brachial artery pressure in a 27 year old
woman during repair of an atrial septal defect using cardiopulmonary bypass at a constant systemic perfusion rate. Arterial pressure, which at a constant perfusion rate reflects systemic vascular
resistance, begins to increase 20 seconds after the injection of
acetylstrophanthidin. Seven minutes after injection the pressure
has already fallen, and 12 minutes after injection it is only slightly
higher than the control level. (Reproduced with permission from
Braunwald E, Bloodwell RD, Goldberg LI, Morrow AG. Studies
on digitalis. IV. Observations in man on the effects of digitalis
preparations on the contractility of the non-failing heart and on
total vascular resistance. J Clin Invest 1961;40:52-9.)
Effects on the Failing Heart
Atrial fibrillation versus sinus rhythm. It is now beyond dispute that the digitalis glycosides display efficacy,
both immediately and over the long term, in the control of
heart failure and ventricular rate in patients with atrial fibrillation, Indeed, the development of an increase in atrioventricular block resulting in slowing of an excessive ventricular rate remains a principal indication for the use of
digitalis (13). However, Christian (14) and other investigators suggested early in this century that this drug may
also be effective in the treatment of patients with heart
failure and sinus rhythm. Although cardiac glycosides are
still advocated as the principal inotropic agents for the treatmentof heartfailure (13), and although studiesofthe effects
of their short-term administration to patientshave confirmed
that they improve myocardial performance in patients with
heartfailure and sinusrhythm(15), thereis stillconsiderable
controversy about whether patients with sinus rhythm who
receive maintenance therapyderive clinical benefit (16). To
understand this controversy, it is necessary to reviewseveral
aspects of the action of this drug.
Inotropic action of digitalis. The modem era of digitalis pharmacology began in 1940 when Gold and Cattell
(17) demonstrated that digitalis increases the contractility
of failing isolated cat papillary muscle in a dose-dependent
manner (Fig. '-). Subsequently, it was shown that digitalis
also improves the function of the failing heart-lung preparation of the dog (18). It is now clear that the extent of the
positive inotropic action of the drug is related inversely to
the baseline condition of the myocardium, cardiac muscle
with depressed function showing a more vigorous response
to the glycoside than normally contracting myocardium.
These observations on the direct positive inotropic actions of cardiac glycosides have been extended to human
subjects, and it has been demonstrated (19) that the contractility of papillary muscle removed from patients with
heart failure, as reflected in the time-tension and forcevelocity relations, is enhanced by the addition of strophanthidin(Fig. 3). Hemodynamic studies(2,13,15) haveshown
that when glycosides are administered to patients with heart
failure, they elevate the ventricular function curve, causing
a reduction in ventricular end-diastolic pressureand volume
and an augmentation of cardiac output and stroke work.
.-•
S/I.I"
"'''' HlJ
120
BRACHIAL
ARTtIW
PR£SalRE
.......JtAI;!II . .,
40.
....
.........;r
....
80~
t
ACFTYLSTROPHANTHI»N
I.S"'9
II".
lACC Vol. 5, No.5
BRAUNWALD
CLINICAL EFFECTS OF DIGITALIS
May 1985:51A-59A
11.30
11:25
4:15
2:00
5:21 OUABAIN
rrrrrr mrrnrrmr fffrm nrm ~m
4:40
5:20
5:35
5:55
8:15
6:44
Figure 2. The effect of ouabain on force of contraction of an
isolated papillary muscle from the right ventricle of a cat heart.
(Reproduced from [17) with permission.)
Effects on the Nonfailing Heart
While there is now general agreement regarding the positive inotropic effect of digitalis in the failing heart, there
has been considerable dispute concerning its actions on the
normal heart and the nonfailing though diseased heart. DigFigure 3. The effect of the addition of strophanthidin on the forcevelocity relation of a human left ventricular papillary muscle studied in vitro. The addition of strophanthidin increased both Vmax
and Po (the maximal force of isometric contraction, that is, intercept on the abscissa), while decreasing the time from stimulation
to maximal shortening from 390 to 340 ms. (Reproduced with
permission from Sonnenblick EH, Braunwald E, Morrow AG.
Contractile properties of human heart muscle: studies on myocardial mechanics of surgically excised papillary muscles. J Clin
Invest 1965;44:966-77.)
Slrophonlhidin
6
'U
....~
5
o contra!
• Slrop/lonlhidin
o.5p/cc
E
E
;;4
z
J.S.
OS-/~-/7
z
~4,~
~
FfllICI cl." II
w
RHOl Mil
~ 3
R.,~"
:I:
Ul
15
>-
~
u
2
ooJ
w
>
o
4
6
53A
italis does not increase the cardiac output at rest or during
exercise in subjects without heart failure (20), but its cardiac
stimulating action is reflected in the finding that the same
levels of cardiac output and stroke work are delivered from
a lower ventricular filling pressure or during a shorter period
of time, or both. Several other lines of evidence indicate
that cardiac glycosides do indeed exert a positive inotropic
effect on nonfailing cardiac muscle. The effects of the longterm administration of digitoxin on the contractile state of
nonfailing myocardium were studied in papillary muscles
obtained from normal cats and from cats in which ventricular
hypertrophy had been produced by banding of the pulmonary artery (21). While the muscle's length-tension relations
at rest were not affected by the glycoside, the isometrically
developed force and the maximal rate of force development
were increased in both normal and nonfailing hypertrophied
myocardium.
Canine studies. In investigations on the conscious dog,
it was found that at similar heart rates and left ventricular
end-diastolic pressures, the long-term administration of digoxin increased left ventricular force, maximal rate of rise
of left ventricular pressure (dP/dt) and the velocity of myocardial shortening (22). McRitchie and Vatner (23) also
showed that in normal conscious dogs ouabain increases
arterial pressure as well as indexes of myocardial contractility, such as left ventricular dP/dt divided by pressure
([dP/dt]/P) and the velocity of fiber shortening. The positive
inotropic effect of the drug was also reflected in its ability
to maintain stroke volume and to reduce left ventricular endsystolic dimensions at a constant preload, despite an increase
in arterial pressure. However, in the intact heart of the
conscious dog (for example, in isolated cardiac muscle),
the degree of the augmentation of contractility induced by
cardiac glycosides is related inversely to the baseline contractile state. Thus, in the normal conscious dog, when
myocardial contractility is stimulated by adrenergic activity,
as occurs during exercise, cardiac glycosides exert only a
negligible positive inotropic effect (24). Conversely, the
inotropic effects of glycosides are more prominent in the
failing than in the nonfailing heart; the effects of ouabain
on right ventricular segment length, pressure, velocity of
shortening and (dP/dt)/P were studied in conscious dogs
before and after the induction of chronic right heart failure
(25). Although ouabain exerted an inotropic effect in normal
healthy, conscious dogs, this effect was substantially greater
in the presence of chronic heart failure (Fig. 4). Similarly,
acute depression of myocardial function in the normal dog
induced by sodium pentobarbital or large doses of propranolol enhanced the inotropic effects of ouabain (Fig. 5) (26).
There is also evidence that cardiac glycosides elevate the
performance of the canine heart when the latter is depressed
by halothane (27) or thiopental (28) and reduce the oxygen
debt after exercise in patients with cardiac enlargement but
without heart failure (29).
BRAUNWALD
CLINICAL EFFECTS OF DIGITALIS
54A
lACC Vol. 5, No.5
May 1985:5I A- 59A
OUABAIN
MYOCARDIAL
<>
+40[
+20
,.-
/./
-_2.:':~~/J
..... --....--......-
......... AFTER CHF
-...._- ... ~-6- - -
3mi n
'
41 SeCJ
+
+
40[
20
~~ ::::~~:i'
CO NT ROL
....-_..... _-......
~~~:::~-~~-""._- - , ""
30
60
TIME (MINUTES)
0
90
120
Figure 4. Comparison of the effects of ouabain in the same conscious dog before (triangles) and after (circles) congestive heart
failure (CHF) had been induced. Not only was the peak effect
greater but the positive inotropic action of the drug was more
sustained in the failing heart. Baseline measurements are shown
at the left. VISO = velocity of shortening at the same length
(isolength). (Reproduced with permission of the American Heart
Association, Inc. from VatnerSF, Braunwald E. Effects of chronic
heart failure on the inotropic response of the right ventricle of the
conscious dog to a cardiac glycoside and tachycardia. Circulation
1974;50:728-34.)
Human studies. The effects of digitalis on the force of
contraction of the nonfailing human heart were studied using
a force transducer sewn to the right ventricle of patients
without heart failure undergoing cardiopulmonary bypass
(8). Both acetylstrophanthid in and lanatoside C produced a
Figure 5. The average (± SEM) values for the quotient of dP/dt
and developed left ventricular (LV) pressure, an index of myocardial contractility, before and 30 minutes after ouabain for conscious dogs (left panel), afteranesthesia (middle panel) andconsciousdogs after propranolol (rightpanel); thesmallest augmentation
occurs in the normal conscious dog (left panel). (Reproduced with
permission from Vatner SF, Higgins CB. Patrick T, Franklin D.
Braunwald E. Effects of cardiac depression and of anesthesia on
the myocardial action of a cardiac glycoside. J Clin Invest
1971;50:2585-95.)
70
-=-,
ANESTHETIZED
CONSCIOUS
l
0
CIl
lfl
W
PROPRANOlOL
11
S EM
60
tr
:::>
VI
VI
w
tr 50
a.
-0
.....0..
"0
0
W
a.
9
w
40
>
w
0
~
30
5 min
7m in
10 min
15min
ACETYLSTROPHANTHIDIN
1.5 mv
V ' SO ( "i.. ll. )
o
FORCE
ir---.l BEFORE CHF
dP~d !.- o, ll.)
o
CON TRACTILE
CONTROL
OUABAIN CONTROL
OUABAIN
Figure 6. Right ventricular contractile force recordings immediately before and at intervals after the administration of 1.5 mg of
acetylstrophanthidin in a 27 year old woman with atrial septal
defect (same patient as in Fig. I). Augmentation of contractile
force is evident 3 minutes after injection. (Reproduced with permission from Braunwald E, Bloodwell RD, Goldberg LI, Morrow
AG. Studies on digitalis. IV. Observations in man on the effects
of digitalis preparations on the contractility of thenon-failing heart
and on total vascular resistance. J Clin Invest 1961;40:52-9.)
prompt increase in isometric force development, indicating
that the digitalis glycosides do stimulate the nonfailing heart
of anesthetized human patients (Fig. 6). These studies were
then extended to' conscious human subjects without heart
failure in whom the measurement of maximal dP/dt was
recorded using a high fidelity rnicromanometer catheter;
ouabain increased both right and left ventricular dP/dt (Fig.
7) (30).
The results of several additional studies on the action of
cardiac glycosides on nonfailing human myocardium are in
accord wiith these findings. Thus. the effects of ouabain on
ventricular force-velocity relations were studied in conscious patients without heart failure who had previously
undergone corrective cardiac operations, at which time
radiopaque markers had been sewn to the external surfaces
of the ventricle (31). The glycoside augmented myocardial
contractility, simultaneously increasing velocity of shortening and intraventricular pressure (Fig. 8). Crawford et at.
(32) examined the effects of digoxin on left ventricular
function in normal subjects using echocardiography. They
observed a positive inotropic action, as reflected in an increase in the left ventricular ejection fraction and an augmentation of the extent and rate of left ventricular shortening
as well as significant improvement in left ventricular performance after acute pressure loading.
The observations just summarized indicate quite clearly
that digitalis glycosides stimulate the contractility of the
nonfailing human heart. However , these drugs do not affect
cardiac output at rest or maximal levels of oxygen uptake,
pulmonary ventilation or heart rate in normal subjects (20) .
Therefore, adjustments of the other determinants of cardiac
output, such as preload, afterload and heart rate, prevent
this increase in contractility from being translated into an
augmentation of cardiac output (33). There is no evidence
that the administration of digitalis is of any clinical value
in patients with sinus rhythm without heart failure and its
administration prophylactically to such patients is not ordinarily indicated (13).
BRAUNWALD
CLINICAL EFFECTS OF DIGITALIS
lACC Vol. 5, No.5
May 1985:5IA-59A
OUABAIN 0 .6 mg.
CONTROL
1005mm Hlj/sec.
dp/dt
Figure 7. Recordings of left ventricular (LV) pressure and rate of change
of pressure (dP/dt) before and30 minutes after ouabain administration in a
patient without heart failure . EKG =
electrocardiogram . (Reproduced with
permission from Mason OT. Braunwald E. Studies on digitalis. IX . Effects of ouabain on the non-failing human heart. J Clin Invest 1963;
""ft.
Hlj/St<
1495 mm. Hglsec.
f
I
'~
100
5
2000
1500 "1ft
....
1000 HljISoc.
LV
o
55A
500
..Ill
HQ
150
100
75
42: 1105-11 .)
25
'---l,_--~_'---"""oIo
0 ----..._""----le_'---~
EKG
Effects in Myocardial Ischemia and
Ischemic Heart Disease
Figure 8. The effects of ouabain on the instantaneous force-velocity relations of the right ventricle (R.V,) of a conscious human
subject with radiopaque markers implanted on the epicardium at
the time of corrective surgery . The velocity of shortening of the
segment of myocardium between themarkers at theisolength (ISL)
point is plotted ontheordinate. Thecorresponding right ventricular
pressure is plotted on the abscissa. Each point represents a single
ventricular contraction, and the crossbars represent ± I SO.
(Reproduced with permission of the American Heart Association .
Inc. from Sonnenblick EH , Williams JF Jr, Glick G, Mason OT.
Braunwald E. Studies on digitalis. XV. Effects of cardiac glycosides on myocardial force-velocity relations in the non-failing
human heart. Circulation 1966;34:532-9.)
•
•
••
•
7h ••
u
:: 5
<,
'"
'0
L:
..
•
c:
--'
'al
-s:
z
z
w
~ 4
o
J:
Vl
u,
o
>-
lU
o
0
~ :3
0
>
»
ci
a
a
~
CONTROL
0
OUABAIN
O.Olmo/Ko
•
a
40
35
R.V
PRESSURE ISL
45
-
mmH9
Myocardial oxygen consumption (MVO z). To understand the action of glycosides in myocardial ischemia and
ischemic heart disease, it is useful, first, to consider their
effects on MVOz. Because the MVOz is determined largely
by intramyocardial tension. myocardial contractility and heart
rate, the net action of digitalis on these three hemodynamic
variables governs its effect on MVOz (34) . In anesthetized
dogs without heart failure or cardiac dilation, digitalis augments myocardial contractility with little effect on intramyocardial tension and heart rate; therefore, it might be
expected to, and indeed it does, elevate MVOz in the nonfailing heart. However, in the presence of heart failure,
digitalis reduces ventricular volume and thereby intramyocardial tension. Heart rate may also be slowed in heart
failure as a consequence of diminished sympathetic drive,
and these two effects tend to balance the increase in MVOz
resulting from augmented myocardial contractility; the net
effect is to produce either no change or a small increase in
MVOz (35). DeMots et al. (36) demonstrated in patients
with chronic ischemic heart disease without heart failure
that ouabain reduces left ventricular end-diastolic pressure
and volume and augments the maximal velocity of estimated
contractile element shortening; it causes no significant change
in MVOz, presumably because of the opposing effects of
the glycoside-induced reduction in tension and augmentation
of contractility.
Acute hypoxia, ischemia and infarction. There is considerable evidence that digitalis glycosides are capable of
stimulating the acutely hypoxic or ischemic heart (37). Thus,
in dogs in which severe systemic hypoxia was induced by
respiring 7 to 10% oxygen, acetylstrophanthidin caused elevations of left ventricular dP/dt and stroke work that were
distinct but smaller than the changes induced by identical
doses of the drug in nonhypoxic animals (38). Similarly,
Sarnoff et al. (39) reported that the intracoronary injection
56A
BRAUNWALD
CLINICAL EFFECTS OF DIGITALIS
of acetylstrophanthidin into a globally ischemic isolated canine heart caused a marked improvement of cardiac contractility. In the conscious dog with acute coronary occlusion, ouabain improves global left ventricular performance
(40), and in the nonfailing dog heart cardiac glycosides
intensify the ischemia produced by coronary occlusion (41),
presumably by augmenting myocardial oxygen requirements.
However, in the failing heart, digitalis has the apposite
effect (42); by reducing myocardial oxygen requirements.
cardiac glycosides may reduce the severity 'of ischemia induced by coronary occlusion (Fig. 9) .
The routine use of digitalis in the management of patients
with acute myocardial infarction was first suggested by Herrick in 1912 (43) and the drug was initially employed widely
in this condition (44). Subsequently, its use became more
Figure 9. Average effecIs (:!: SEM) of ouabain on mean ST
segment elevation and hemodynamic variables after coronary ligation and with arterial pressure increased in six anesthetized open
chest dogs with experimentally produced heart failure. There is
little effect on mean aortic pressure or heart rate . Ouabain caused
a striking decline of markedly elevated mean left atrial pressure ,
accompanied by a striking reduction of ST segment elevation of
the ischemic myocardium. (Reproduced with permission from Watanabe A, Covell JW. Maroko PR, Braunwald E, Ross J Jr. Effects
of increased arterial pressure and positive inotropic agents on the
severity of myocardial ischemia in the acutely depressed heart .
Am J Cardiol 1972;30:371 -7.)
0....1---'---'-----'--'---
150
rX
'0
u:r
~1100
~W
~iil 50
",III
2'"
f
0....1---'---'-----'--'---
30l
cf 20
SoJ
!CO!
~'"
""c
~!
~W
10-
~A.
O--'----'--L----'---'---
...E
elOO
'"
!c
c
~ 50
c
c
'"
:r
0 --I----'--=-'----=-~=-':-:-:-:--
c
OUABAIN
lACC Vol. 5, No.5
May 1985:5IA-59A
controversial (37) . Thus , Cohn et al. (II) administered cardiac glycosides intravenously to patients with cardiogenic
shock . They reported a short-term increase in systemic vascular resistance that was associated with the development
of pulmonary edema in one patient. Later, there was an
increase in myocardial contractility, as reflected in an increase in the left ventricular maximal dP/dt, usually accompanied by a decrease in the left ventricular end-diastolic
pressure; however, cardiac output was not increased significantly . In patients with acute myocardial infarction and
congestive heart failure without cardiogenic shock, several
investigators (45,46) observed that digitalis increases ejection fraction slightly . Morrison et al. (47) also reported that
in patients with acute myocardial infarction, digoxin causes
a modest but significant elevation of a depressed ejection
fraction (from 0.29 to 0.33).
Chronic ischemic heart disease. Kleiman et al. (48,49)
studied the effects of both intravenous and oral digoxin in
patients with stable coronary artery disease without clinical
evidence of heart failure after coronary artery bypass grafting. Using computer-assisted fluoroscopic analysis of the
motion of markers implanted into the myocardium at operation, they observed that the ejection fraction and the
velocity of circumferential fiber shortening both increased,
while the left ventricular end-diastolic volume and cardiac
output did not change significantly. The improvement in
performance induced by digoxin was greater in normal myocardial segments than in segments that were contracting
abnormally due to ischemia or fibrosis. Indeed, deterioration
of segmental performance occurred in approximately half
of the segments with depressed function, a deterioration that
may be due to the inability of infarcted segments of myocardium to sustain the increased wall tension resulting from
the digitalis-induced augmented contractility of the normal
myocardium. Therefore, in patients with ischemic heart disease, the response of particular segments of the myocardium
to cardiac glycosides appears to depend on the relative quantity of viable nonischemic muscle, on the one hand, and of
ischemic or infarcted muscle in the particular segment, on
the other. It has been suggested that it is the failure of
ischemic and necrotic segments of the myocardium to respond to digitalis which is responsible for the lack of impressive improvement of left ventricular function in patients
with acute myocardial infarction who are treated with the
drug (50).
In patients with chronic ischemic heart disease, the acute
administration of digoxin had a mild beneficial effect on
global left ventricular performance at rest, but did not alter
the onset of exercise-induced angina or affect exerciseinduced deterioration of left ventricular function (51) . Similarly, Firth et al. (52) reported that digoxin did not cause
any significant change in left ventricular function at rest in
patients with ischemic heart disease. Although the glycoside
prevented the exercise-induced increases in left ventricular
BRAUNWALD
CLINICAL EFFECTS OF DIGITALIS
IACC Vol. 5. No.5
May 1985:5IA-59A
end-diastolic and end-systolic volumes in patients with a
normal ejection fraction at rest, it failed to do so in patients
with a depressed ejection fraction. Thus, the positive inotropic effects of long-term digoxin therapy could be demonstrated more readily in patients with chronic ischemic
heart disease whose left ventricular function was well preserved than in those with an extensive infarct. These observations support those of Kleiman et al. (48,49) and suggest that the extent of improvement of left ventricular function
by glycosides in chronic ischemic heart disease without
overt heart failure depends on the relative quantities of well
preserved and extensively damaged myocardium.
In light of the foregoing data, there is little doubt that
cardiac glycosides can improve myocardial performance in
patients with ischemic heart disease. The degree of improvement, however, varies among different patients and
in different portionsof the ventriclein anyone patient. Thus,
glycosides cause no improvement or can even result in systolic expansionof severelyischemic or scarredmyocardium.
They cause only slight augmentation of contractility in normal myocardium that is adrenergically stimulated because
of left ventricular failure as a consequence of extensive
ischemia or infarction. Therefore, digitalis glycosides cause
only modest improvement of global left ventricular function
in heart failure occurring as a consequence of acute myocardial infarction. Indeed, transient acute deterioration of
left ventricular function may occur when a rapidly acting
glycoside is administered intravenously and elevates systemic vascular resistance and ventricular afterload acutely
in patients with heart failure secondary to ischemic heart
disease. Digitalis may cause a modest improvement of left
ventricularpump performance in patients with chronic ischemic heart disease and impaired global left ventricularfunction. However, the salutary effects of glycosides on left
ventricular pump function in such patients are less than in
patients with left ventricular failure secondary to chronic
ventricular volume overload, such as occurs in valvular
regurgitation.
Clinical Implications
Mulrow et al. (53) recently provided a critical appraisal
of the 16 articles published in English between 1960 and
1982 which specifically address the clinical evaluation of
digitalis therapy in heart failure. In the nine studies in which
the effects of simple withdrawal of digitalis therapy were
studied, no consistent evidence of clinical deterioration was
noted. However, because detailed criteria for patient selection were not provided in these studies, few conclusionscan
be drawn from them. In three other investigations, the addition of digoxin was not found to have any beneficial effects
over therapy with diuretic drugs alone. Arnold et al. (54)
foundthat left ventricularfunctiondeterioratedafter digitalis
withdrawal, whereas Gheorghiade and Beller (55), who
57A
studied patients with chronic ischemic heart disease and
heart failure who were also treated with diuretic drugs and
vasodilators, reported that withdrawal of digoxin had no
adverse clinical or hemodynamic effects.
Murray et al, (56) reported that long-termtreatment with
digitalis in patients with sinus rhythm and symptomatic cardiac failure caused little change in hemodynamic measurements at rest, but did result in lower left ventricular enddiastolic pressures and higher cardiac index and stroke work
during exercise. These changes were observed both after
the short-term administration of ouabain and long-term oral
digoxin therapy. Fleg et al. (57) carried out a placebocontrolled crossover study with digoxin and reported that
the withdrawal of digitalis in patients with compensated
heart failure increased ventricular end-diastolic dimensions
and decreased the velocity of fiber shortening, but did not
cause a clinical exacerbation of heart failure. They concluded that the drug is of little benefit in elderly patients
with stable congestive heart failure, because none manifested deterioration in the 3 months during which they received placebo in place of digoxin. Lee et al. (58) found
that 14 of 25 patients with heart failure showed clinical
improvement while receiving digoxin; improvement was
most striking in patients with severe heart failure and a
protodiastolic gallop. The majority of the patients whose
condition did not improvewith digitalis might not have been
expected to show improvement because they had a normal
left ventricular ejection fraction at rest, were asymptomatic
or had hypertrophic cardiomyopathy.
Other Considerations
Cardiac glycosides are relatively weak inotropic agents
in comparison with catecholamines and the newer inotropic
agents (the bipyridines, amrinone and milrinone) (59). In a
comparison of the inotropic effects of isoproterenol and
ouabain in the normal anesthetized dog (60), isoproterenol
produced a peak increase in contractile force of approximately 160% of control, compared with an increase of only
50% resulting from the administrationof ouabain. However,
the very fact that glycosides are relatively weak inotropic
agents may make them particularly suitable for long-term
administration because it is possible that prolongedvigorous
stimulation of failing myocardium , that is, so-called "flogging" the failing heart, may be deleterious.
Conclusions
The cardiac glycosides have been used continuously in
clinicalmedicinefor two centuries. Duringthe past 25 years,
a great deal has been learned about the cellular mechanism
of action of these drugs: these appear to result from an
inhibition of sarcolemmal Na + -K + adenosine triphosphatase, which causes an augmentation of intracytoplasmic cal-
58A
BRAUNWALD
CLINICAL EFFECTS OF DIGITALIS
cium concentration, which in tum is responsible for the
stimulation of myocardial contractility. The positive inotropic action is seen in normal as well as in failing hearts,
and has been demonstrated in isolated cardiac muscle, heartlung preparations, anesthetized and conscious dogs and human subjects. Although a positive inotropic action is present
in the normal heart, it is not translated into an augmentation
of a normal cardiac output or reduction of a normal left
ventricular filling pressure. However, substantial hemodynamic benefits have been demonstrated in the dilated failing
heart. The augmentation of contractility, as shown by an
increase in both dP/dtJP and velocity of myocardial fiber
shortening, is reflected by an improvement in pump performance, with an elevation of depressed cardiac output and
a reduction of elevated ventricular filling pressure. Secondary vasodilation occurs, presumably as a consequence of
the withdrawal of sympathetic stimulation characteristic of
heart failure.
Digitalis remains a mainstay ofthe therapy for congestive
heart failure. Its effects are most prominent in heart failure
associated with atrial fibrillation and a rapid ventricular rate,
but it is also effective in severe heart failure and cardiac
dilation secondary to volume overload, as well as in patients
with a dilated, failing ventricle secondary to hypertension.
Little improvement is observed in patients with left ventricular failure secondary to ventricular outflow tract obstruction (aortic or pulmonary stenosis) or in patients with
acute left ventricular failure secondary to myocardial ischemia or infarction, or both. When it is administered in
proper doses and for the proper indications, glycoside therapy may be well tolerated by a patient for decades. Therefore, despite its modest inotropic action and therapeutic
ratio, digitalis remains a useful drug in the therapy of many
patients with chronic congestive heart failure.
References
I. Withering W. An Account of the Foxglove, and Some of Its Medical
Uses: With Practical Remarks on Dropsy and Other Diseases. Birmingham, England: M. Swinney, 1785.
2. Hoffman BG, Bigger JT Jr. Digitalis and allied cardiac glycosides.
In: Gilman AG, Goodman LS, Gilman A, eds. The Pharmacologic
Basis of Therapeutics. New York: Macmillan, 729-60.
3. Bouillard J. Traite Clinique des Maladies du Cour. Paris: 1835.
4. Fothergill JM. Digitalis: Its Mode of Action. London: 1871.
5. Dock W, Tainter ML. The circulatory changes after full therapeutic
doses of digitalis, with a critical discussion of views on cardiac output.
J Clin Invest 1930;8:467-84.
6. Ross J Jr, Waldhausen JA, Braunwald E. Studies on digitalis. I. Direct
effects on peripheral vascular resistance. J Clin Invest 1960;39:930-6.
7. Ross J Jr, Braunwald E, Waldhausen JA. Studies on digitalis. II.
Extracardiac effects on venous return and on the capacity of the peripheral vascular bed. J Clin Invest 1960;39:937-42.
8. Braunwald E, Bloodwell RD, Goldberg LI, Morrow AG. Studies on
digitalis. IV. Observations in man of the effects of digitalis preparations on the contractility of the nonfailing heart and on total vascular
resistance. J Clin Invest 1961;40:52-9.
lACC Vol. 5, No.5
May 1985:51A-59A
9. Mason DT, Braunwald E. Studies on digitalis. X. Effects of ouabain
on forearm vascular resistance and venous tone in normal subjects and
in patients with heart failure. J Clin Invest 1964;43:532-43.
10. Vatner SF, Higgins CB, Franklin D, Braunwald E. Effects of a digitalis
glycoside on coronary and systemic dynamics in conscious dogs. Circ
Res 1971;28:470-9.
I I. Cohn IN, Tristani FE, Khatri 1M. Cardiac and peripheral vascular
effects of digitalis in clinical cardiogenic shock. Am Heart J
1969;78:3I8-30.
12. Baleon R, Hoy J, Sowton E. Haemodynamic effects ofrapid digitalization following acute myocardialinfarction. Br Heart J 1%8;30:373-6.
13. Smith TW, Braunwald E. The management of heart failure. In: Braunwald E, ed. Heart Disease: A Textbook of Cardiovascular Medicine.
2nd ed. Philadelphia: WB Saunders 1984:503-59.
14. Christian MH. Digitalis therapy. Satisfactory effects in cardiac cases
with regular pulse rate. Am J Med Sci 1919;157:593-602.
15. Bloomfield RA, Rapoport B, Milnor JP, Long WK, Mebane JG, Ellis
LB. Effects of cardiac glycosides upon dynamics of circulation in
congestive heart failure. Ouabain. J Clin Invest 1948;27:588.
16. Hamer J. The paradox of the lack of the efficacy of digitalis in congestive heart failure with sinus rhythm. Br J Clin Pharmacol 1979;8:109-13.
17. Gold H, Cattell M. Mechanism of digitalis action in abolishing heart
failure. Arch Intern Med 1940;65:263-78.
18. Farah A, Maresh G. Determination of the therapeutic irregularity, and
lethal doses of cardiac glycosides in the heart-lung preparation of the
dog. J Pharmacol Exp Ther 1948;92:32-42.
19. Sonnenblick EH, Braunwald E, Morrow AG. The contractile properties of human heart muscle: studies on myocardial mechanics of
surgically excised papillary muscles. J Clin Invest 1965;44:966-77.
20. Williams MH Jr, Zohmar LR, Ratner AC. Hemodynamic effects of
cardiac glycosides on normal human subject during rest and exercise.
J Appl Physiol 1958;13:417-21.
21. Williams JF Jr, Potter RD. The effect of chronic digitoxin administration on the contractile state of normal and nonfailing hypertrophied
myocardium. J Clin Invest 1975;56:71-8.
22. Mahler F, Karliner JS, O'Rourke RA. Effects of chronic digoxin
administration on left ventricular performance in the normal conscious
dog. Circulation 1974;50:720-7.
23. McRitchie RJ, Vatner SF. The role of arterial baroreceptors in mediating the cardiovascular response to a cardiac glycoside in conscious
dogs. Circ Res 1976;38:321-6.
24. Horowitz LD, Atkins JM, Saito M. Effect of digitalis on left ventricular function in exercising dogs. Circ Res 1977;41:744-50.
25. Vatner SF, Braunwald E. Effects of chronic heart failure on the inotropic response of the right ventricle of the conscious dog to a cardiac
glycoside and to tachycardia. Circulation 1974;50:728-34.
26. Vatner SF, Higgins CB, Patrick T, Franklin D, Braunwald E. Effects
of cardiac depression and of anesthesia on the myocardial action of a
cardiac glycoside. J Clin Invest 1971;50:2585-95.
27. Goldberg AH, Maling HM, Gaffney TE. The value of prophylactic
digitalization in halothane anesthesia. Anesthesiology 1962;23:207-12.
28. Goldberg AH, Maling HM, Gaffney TE. The effect of digoxin pretreatment on heart contractile force during thiopental infusion in dogs.
Anesthesiology 1961;22:974-6.
29. Kahler RL, Thompson RH, Buskirk ER, Frye RL, Braunwald E.
Studies on digitalis. VI. Reduction of the oxygen debt after exercise
with digoxin in cardiac patients without heart failure. Circulation
1963;27:397-405.
30. Mason DT, Braunwald E. Studies on digitalis. IX. Effects of ouabain
on the nonfailing heart. J Clin Invest 1963;42:1105-1 I.
31. Sonnenblick EH, Williams JF Jr, Glick G, Mason DT, Braunwald E.
Studies on digitalis. XV. Effects of cardiac glycosides on myocardial
force-velocity relations in the nonfailing human heart. Circulation
1966;34:532-9.
l ACC Vol. 5, No. 5
May 1985:5I A-5 9A
BRAUNWALD
CLINICAL EFFECTS OF DIGITALIS
59A
32. Crawford MH, Karliner JS , O 'R ourke RA, Amon KW. Favorable
effects of oral maintenance digoxin therapy on left ventricular performance in normal subjects: echocardiographic study . Am J Cardiol
1976:38:843-7 .
48. Kleiman JH, Ingels NB , Daughters G, Stinson EB, Alderman EL,
Goldman RH. Left ventricular dynamics during long-term digoxin
treatment in patients with stable coronary artery disease. Am J Cardiol
1978:4 1:937-42.
33. Braunwald E. On the difference between the heart's output and its
contractile state. Circulat ion 1971 :43 :171-4.
49. Kleiman JH, Alderman EL, Goldman RH. Effects of digitalis on
normal and abnormal left ventricular segmental dynamics. Am J Cardiol 1979;43:1001-8.
34 . Braunwald E. Control of myocardial oxygen consumption: physiologic
and clinical considerations. Am J Cardiol 197 1;27:416-32.
35. Cove ll JW , Braunwald E, Ross J Jr , Sonnenblick EH. Studies on
digitalis. XVI. Effects on myocardial oxygen consumption. J Clin
Invest 1966;45: 1535-42.
36 . DeMots H, Rahimtoola SH , Kremkau EL, Bennett W, Mahler D.
Effects of ouabain on myocardial oxygen supply and demand in patients with chronic coronary artery disease. J Clin Invest 1976:58:312- 9.
37. Karliner JS , Braunwald E. Present state of digitalis treatment of acute
myocardial infarction. Circulati on 1972 :45:891-902.
38. Donlon JV, Yu PN. Cardiovascular response of hypoxic myocardium
to acetyl strophanthidin. Am Heart J 1969:78:238-44.
39 . Sarnoff SJ, Gilmore JP, Wallace AG , Skinner NS Jr, Mitchell JH,
Daggett WM . Effect of acetyl strophanthidin therapy on cardiac dynamics. oxyge n consumpti on and efficiency in the isolated heart with
and without hypoxia. Am J Med 1964;37:3-11.
50. Hodges M, Friesinger GC, Riggins RCK, Dagenais GR . Effect of
intravenously administered digoxin on mild left ventricular failure in
acute myocardial infarction in man. Am J Cardiol 1972;29:749- 55 .
51. Verani MS, Zacca NM, Miller RR, Luchi RJ, Chahine RA. Effects
of digoxin on left ventricular function in coronary artery disease patients. Int J Cardiol 1982:2:221- 3 1.
52. Firth BG, Dehmer OJ, Corbett JR , Lewis SE, Parkey RW, Willerson
JT. Effect of chronic oral digoxin therapy on ventricular function at
rest and peak exercise in patients with ischemic heart disease. Assessment with equilibrium gated blood pool imaging. Am J Cardiol
1980;46:481-9.
53. Mulrow CD, Feussner JR, Velez R. Reevaluation of digitalis efficacy .
New light on an old leaf. Ann Intern Med 1984;101: 113-7.
54 . Arnold SB, Byrd RC, Meister W. Long-term digitalis therapy improves left ventricular function in heart failure. N Engl J Med
1980;303:1443-8.
40 . Vatner SF, Baig H. Comparison of the effects of ouabain and isoproterenol on ischemic myocardium of conscious dogs. Circulation
1978;58:654-62.
55. Gheorghiade M, Beller GA . Effects of discontinuin g maintenance
digoxin therapy in patients with ischemic heart disease and congestive
heart failure in sinus rhythm. Am J Cardiol 1983;51:1243-50.
4 1. Maroko PR, Kjekshus JK , Sobel BE, et at. Factors influencing infarct
size following experimental coronary artery occlusion. Circulation
1971 ;43:67- 82.
56 . Murray RG. Tweddel AC, Martin W , Pearson 0 , Hutton I, Lawrie
TDV. Evaluation of digitalis in cardiac failure . Br Med J 1982:
284:1526-8 .
42. Watanabe T , Covell JW . Maroko PR, Braunwald E, Ross J Jr. Effects
of increased arterial pressure and positive inotropic agents on the
severity of myocardial ischem ia in the acutely depressed heart. Am J
Cardiol 1972:30:371-7 .
57. Fleg JL, Gottlieb SH . Lakatta EG. Is digoxin really important in
treatment of compensated heart failure? A placebo-controlled crossover study in patients with sinus rhythm . Am J Med 1983;73:244-50.
43 . Herrick JB. Clinical features of sudden obstruction of the coronary
arteries. JAMA 1912;49:2015- 7.
44 . Hamman L. The symptoms of coronary occlusio n. Johns Hopkins
Med J 1926 ;38:273-5.
45 . Bostrom PA, Andersson J , Johansson BW, et at. Haemodynami c
effects of prenaterol and cardiac glycosi des in patients with recent
myocardial infarction. Eur J Clin Invest 1984;14:175-80.
46. Marchionni N, Vannucci A, Pini R, et at. Hemodynamic effects of
digoxin in acute myocardial infarction. Eur Heart J 1980;1:319-26.
47 . Morrison J. Coromilas J, Robbins M, et at. Digitalis and myocardial
infarction in man . Circulation 1980;62:8-1 6.
58. Lee DC, Johnson RA, Bingham JB. Heart failure in outpatients: a
randomized trial of digoxin versus placebo. N Engl J Med 1982;
306:699- 705.
59 . Braunwald E, Sonnenbl ick EH. Chakrin LW, Schwartz RP Jr, eds.
Milrinone: Investigation of New Inotropic Ther apy for Congestive
Heart Failure. New York: Raven , 1984 .
60 . Beiser GO . Epstein SE . Goldstein RE , Stampfer M. Braunwald E.
Comparison of the peak inotropic effects of a catecholamine and a
digitalis glycoside in the intact canine heart. Circulation 1970;42:805-13.