I.V. ISOSORBIDE DINITRATE/Rabinowitz et al.

771

timation of infarct size. Circulation 52: 1, 1975
37. Cairns, JA, Missirlis E, Fallen EL: Myocardial infarction size
from serial CPK: variability of CPK serum entry ratio with size
and model of infarction. Circulation 58: 1143, 1978
38. Slutsky AS: Individualized values for the disappearance rate
parameter (Kd) in the enzymatic estimation of infarct size. A
critique. Circulation 56: 545, 1977
39. Sobel BE, Markham J, Karlsberg RP, Roberts R: The nature
of disappearance of creatine kinase from the circulation and its
influence on enzymatic estimation of infarct size. Circ Res 41:
836, 1977
40. Cairns JA, Klassen GA: Intravenous propranolol therapy for
acute myocardial infarction in man. Hemodynamic and serial
creatine kinase assessment. Chest 79: 277, 1981

Intravenous Isosorbide Dinitrate in Patients

with Refractory Pump Failure and
Acute Myocardial Infarction
BABETH RABINOWITZ, M.D., ISRAEL TAMARI, M.D., ELLA ELAZAR, M.SC.,
AND HENRY N. NEUFELD, M.D.

SUMMARY We studied the hemodynamic effects of isosorbide dinitrate administered by continuous i.v.
infusion to 22 patients with chronic refractory pump failure and 18 with pump failure due to acute myocardial
infarction. In patients with severe pump failure, i.v. ISDN markedly decreased pulmonary capillary wedge
pressure (p < 0.001), moderately increased cardiac output (p < 0.01), and decreased systemic vascular
resistance (SVR) (p < 0.001). There were no deleterious effects on arterial pressure and heart rate. The effects
obtained in acute and chronic left ventricular failure were similar. Patients with initial SVR levels lower than
1500 dyn-sec-cm'1 did not significantly increase their cardiac output (p < 0.005). Cardiac output increased
more than 25% only in patients with initial high SVR levels (> 2000 dyn-sec-cm-1). Positive correlations were
found between high SVR and elevated plasma catecholamines (r = 0.53,p < 0.05) and between the initial SVR
and initial heart rate (r = 0.70, p < 0.01). The i.v. administration of isosorbide dinitrate appears to be an effi-
Downloaded from http://ahajournals.org by on February 24, 2021

cient therapy, particularly in selected patients with ischemic pump failure.

NITRATE PREPARATIONS and other vasodila-
tors have been used to treat acute and chronic pump
failure,1-4 but few data have been reported on the
effects' of i.v. isosorbide dinitrate (ISDN).6-8 Although
nitrates are predominantly venodilators,3' 9-1" de-
creases in systemic vascular resistance and increases
in cardiac output have been reported.7' 12-16 Nitrates
are helpful in many patients with coronary heart dis-
ease, although a sudden, sharp decrease in the dia-
stolic arterial pressure may enhance myocardial ische-
mia, particularly in the acute cases.17 19
The present study was performed to evaluate the
hemodynamic effects of i.v. ISDN in 40 patients with
acute and chronic pump failure. Serial determina-
tions of plasma catecholamines were also made to
assess their relation to hemodynamic status and
response to therapy.
Patients and Methods
Forty patients with pump failure received i.v. ISDN
(Isoket) in a dose range of 2-8 mg/hour, based on the
From the Heart Institute, Chaim Sheba Medical Center, Tel
Hashomer, and Sackler School of Medicine, Tel Aviv University,
Tel Aviv, Israel.
Address for correspondence: Babeth Rabinowitz, M.D., Heart
Institute, Chaim Sheba Medical Center, Tel Hashomer, 52621,
Israel.
Received August 18, 1980; revision accepted July 1, 1981.
Circulation 65, No. 4, 1982.
hemodynamic and clinical response. We studied 22
consecutive patients with chronic refractory pump
failure (group A) and 18 patients with acute myocar-
dial infarction and pump failure (group B) (table 1).
Refractory pump failure (group A) was manifest by
persistent left-heart failure despite intensive conven-
tional therapy. All patients in this category had
dyspnea and hypoxia, wet rales and radiographic
evidence of pulmonary venous congestion and cardio-
megaly. Medical therapy in these patients included
large i.v. doses of potent diuretics, digitalis, and bed
rest in hospital. Other therapy included correction of
electrolyte imbalance and treatment of arrhythmias.
The patients with chronic ischemic failure (group A,
subgroup 1) all had recurrent myocardial infarction.
These patients were not considered for surgery either
because of their poor left ventricular function or be-
cause their coronary anatomy did not appear amen-
able to bypass grafting. Poor left ventricular function
was demonstrated noninvasively or by complete car-
diac catheterization and angiography. The left ventric-
ular ejection fraction was less than 25% with diffuse
hypokinesis and areas of akinesis or dyskinesis on nu-
clear ventriculography and, when measured, left ven-
tricular end-diastolic pressure was greater than 20 mm
Hg. Five of the coronary patients had significant mi-
tral regurgitation and systolic regurgitant murmurs.
Mitral regurgitation was diagnosed in three patients
as 3+ on contrast ventriculography and was docu-
 772 CI RCULATION VOL 65, No 4, APRIL 1982

TABLE 1. Use of Intravenous Infusion of Isosorbide Dinitrate
in 40 Patients
Clinical conditions
Group A Chronic refractory
pump failure
Subgroup Al Coronary artery disease
Subgroup A2 Cardiomyopathy
Subgroup A3 Rheumatic valvular insufficiency
Group B Acute myocardial infarction
with pump failure
mented by giant V waves on the pulmonary capillary
wedge pressure tracing in two other patients not pre-
viously catheterized.20
The four patients with congestive cardiomyopathy
(nonischemic) all had cardiomegaly by x-ray and
echocardiography. A coronary cause was excluded by
lack of suggestive clinical history, ECG or coronary
arteriography that demonstrated normal coronary
vessels.
In three patients, the underlying pathology was
inactive rheumatic valvular heart disease. Two had
severe mitral regurgitation and one severe aortic
regurgitation. They were in refractory pump failure
and received i.v. ISDN before surgery.
All patients with chronic refractory pump failure
(group A) were therefore patients hospitalized because
their clinical condition worsened and was increasingly
refractory to therapy. No acute event occurred before
Downloaded from http://ahajournals.org by on February 24, 2021
During the study, diuretics, narcotics and oral ni-
trates were withheld. The patients taking digitalis con-
n tinued their maintenance dose. Anticoagulants, anti-
biotics and lidocaine were continued or started as
22 necessary.
ISDN in D5W was administered intravenously,
15 starting at a dose of 2 mg/hour. In all patients,
4 baseline hemodynamic measurements were obtained
3 before the infusion at least twice during a minimum
18 period of 30 minutes to ensure stable control levels;
pressures were recorded 15 minutes, 30 minutes and 1
hour after the beginning of therapy. Later, pressures
were measured every 2-3 hours, and cardiac output
was measured at least twice every 24 hours. The dose
was gradually increased to achieve a maximal
decrease in PCWP, while maintaining a systolic AP of
at least 90 mm Hg. The maximal dosc administered
was 8 mg/hour. The dose at which "optimal" values
of PCWP, AP and CO were achieved in individual
patients was maintained for at least 24 hours, and
hemodynamic data were obtained at that dose level.
Measurement of Catecholamines
Total plasma catecholamines were determined
fluorometrically using the method described by Ren-
zini et al.,23 modified by Jiang,24 and used by us pre-
viously.25 The upper level of normal in our laboratory
is 300 ng/l.
Fifteen of the 23 patients with chronic pump failure
(subgroup Al) and seven of 18 patients with acute
myocardial infarction (group B) had daily determina-

the initiation of therapy with i.v. ISDN.
Group B included 18 patients studied during an
acute myocardial infarction, defined by the presence
of at least two of the following criteria: typical history,
typical serial electrocardiographic changes, and char-
acteristic elevations of CK and LDH isoenzymes.21
These patients also had left-heart failure unresponsive
to diuretics and narcotics (table 1).
The average age was 60 years in group A (range
46-77 years), and 57 years in group B (range 41-70
years). Group A included 18 men and four women and
group B 16 men and two women.
Hemodynamic Measurements
A triple-lumen, balloon-tipped Swan-Ganz ther-
modilution catheter was introduced through an ante-
cubital vein and passed into the pulmonary artery.
Right atrial (RA), pulmonary arterial (PA) and pul-
monary capillary wedge (PCWP) pressures were con-
tinuously monitored. Cardiac output (CO) was
measured in triplicate by the thermodilution tech-
nique with a bedside computer (Instrumentation
22
Laboratories). Arterial blood pressure (AP) was mea-
sured either directly by an intraarterial line or from
repeated cuff readings. Mean AP was calculated as D
+ (S - D)/3, where S is the systolic pressure and D
the diastolic pressure (mm Hg). Systemic vascular
resistance (SVR) was calculated as 80 (AP RA)/ - mean AP decreased in patients who had a high initial
CO (dyn-sec-cm-5).
The hemodynamic criteria for entrance into the
study were PCWP greater than 20 mm Hg and
systolic AP greater than 90 mm Hg.
tions of plasma catecholamines (between 8 a.m. and 2
p.m.). In subgroup Al, the blood samples were drawn
2-3 minutes after the hemodynamic measurements.
Statistical Analysis
Changes in the hemodynamic variables were ana-
lyzed by paired t test. The linear regressions of SVR
and heart rate on catecholamines were estimated by
the least-squares method. Estimates of correlation
coefficients and changes during ISDN therapy that
could not be different from zero due to sampling varia-
tion, with probability of 0.05 or larger, were denoted
as significant.
Results
Hemodynamic Profile
Group A
The patients in chronic refractory heart failure
(group A) had significant hemodynamic improvement
during i.v. ISDN. In subgroup Al (the 15 patients
with ischemic pump failure), the following hemody-
namic effects were observed (figs. lA-lE).
The initial mean PCWP varied from 21-42 mm Hg.
PCW decreased markedly in 12 of 15 patients.
Changes in the whole group were highly significant (p
< 0.001). CO increased moderately (p < 0.01). The
AP, but was not affected in the other patients. The
change in heart rate had a similar pattern to that of
AP. The changes of both AP and heart rate in the
whole group showed a slight decrease (p < 0.02). Cal-
 I.V. ISOSORBIDE DINITRATE/Rabinowitz et al. 773

150
40 \

1302

I
E 110~
30 -
CL

mL ui

70
20.
u -0-

PRIOR ON "'MAX IMAL

C TO ISDN (e) DOSE OF ISDN (0)
160'
10
140-

I
PRIOR ON'OPTIMAL "
A TO ISDN (-) DOSE OF ISDN (o)
100

u 8
I
Downloaded from http://ahajournals.org by on February 24, 2021

5
I~~~~~~~~~~~~~~~~~~~~~~~~
PR IOR ON "OPTIMAL'
D TO ISDN ( * DOSE OF ISDN (O)

C
4 30001
E
-i
t 3 2600 1
0
2200
2
I UR

IE 1800-
L .
u
u
It
PRIOR ON y 1400-
B TO ISDN (*) ISDN (o)

culated SVR decreased in the whole group (p < 0.01).
SVR decreased markedly in patients who had high ini-
tial values, but did not change in patients with an
initially low SVR. Three patients had low-to-normal
initial SVR (950, 1200 and 1450 dyn-sec-cm 5). All of
the others had increased initial SVR; one patient had
an SVR of 1700 dyn-sec-cm-5 and eight patients had
SVR levels greater than 2000 dyn-sec-cm 5.
Figure 2 shows the individual initial levels of SVR
plotted against the percent change in CO induced by
ISDN in the 15 patients who had chronic ischemic
pump failure. All seven patients who had an increase
in CO greater than 25% had an initial SVR greater
than 2000 dyn-sec-cm-5. However, if the patients are
divided arbitrarily into two groups based on initial
V? 1000u
900 - . -
PRIOR ON " OPTIMAL
E TO ISDN (o) DOSE OF S D N (o)
FIGURE 1. Hemodynamic effects of isosorbide dinitrate
(ISDN) in patients with chronic ischemic pump failure
(group Al). (A) Effects of ISDN on pulmonary capillary
wedge pressure (PCW). The decrease in PCW is highly
significant (p < 0.001). (B) Effects of ISDN on cardiac out-
put (CO). The increase in CO is statistically significant (p <
0.01). (C) Effects of ISDN on arterial pressure (AP). (D)
Effects of ISDN on heart rate (HR). (E) Effects of ISDN on
systemic vascular resistance (SVR). The decrease is
statistically significant (p < 0.001).
 774

<i
<

cp
I.
D
u
+100

+80

460,

+20
40,

-2n JL *v~ w If
0

FIGURE 2. Levels
.

1500

of
0

.
2000
0

IN ITIAL SYSTEMIC VASCULAR RESISTANCE

DYNES /sec/ cm-5
0

0
0

initial systemic vascular resistance

(SVR) and correlation with percent change in cardiac output
induced by isosorbide dinitrate in 15 patients from group
Al. Filled circles indicate three patients with an SVR of less
than 1500 dyn-sec-cm-1 (mean 1200 dyn-sec-cm-5). Circles
indicate 12 patients with SVR greater than 1500 dyn-sec-
cm-' (mean 2370 dyn-sec-cm-5). The difference in percent in-
crease in cardiac output is statistically significant (p <
0.005).
SVR, the subgroup with low-to-normal initial SVR
(mean 1200 dyn-sec-cm-5) has a poorer response to
ISDN in terms of CO increases than did the subgroup
0
0

3000
CIRCULATION
S LOW SVR (<1500)
OH IGH SVR (>1500)

0
.'

0
6

5.

control.
3

21

L
L- -
ISD N
10_

10
C
MEAN LEVELS

data from the 18 patients with acute myocardial in-

farction are illustrated in terms of left ventricular
function curves in figure 5.
In group B, mean PCWP decreased from 26.6 mm
Hg to 17.6 mm Hg (p < 0.001), and mean CO in-
creased from 3.7 to 4.2 I/min (p < 0.001). However,
four patients did not improve either clinically or
hemodynamically. All of these patients had initial
low-to-normal SVR levels (< 1500 dyn-sec-cm-5).
Figure 6 shows the mean levels of PCWP and CO in
20
PCW

FIGURE 3. Effects of isosorbide dinitrate (ISDN) on left

ventricular function curves in patients with chronic ischemic
pump failure (group Al, n 15). CO = cardiac output; C =
=
VOL 65, No 4, APRIL 1982

mmHg
30 40

with a high initial SVR (mean 2370 dyn-sec-cm-5) (p < the 18 acute patients (group B) and the 22 chronic pa-
Downloaded from http://ahajournals.org by on February 24, 2021

0.005). A less pronounced correlation (p < 0.02) was
found when the decrease in PCW was compared with
the initial levels of SVR.
Figure 3 illustrates the hemodynamic effects of
ISDN in the patients with chronic refractory pump
failure due to ischemia. Their left ventricular function
tients (group A). There was no significant difference
between the two groups; in both groups, the ventricu-
lar function curves were shifted to the left and slightly
upward.
5

curves shifted markedly to the left and slightly up-

ward.
In group Al, the five patients with ischemic mitral 4I

regurgitation appeared to respond more markedly

than the 10 patients with chronic ischemic failure who E
3
did not have mitral regurgitation (fig. 4). However, the 0
changes in PCW and CO appeared more marked be- I.)

cause the initial hemodynamic profile in these patients 2

showed poorer baseline function compared with the

other subgroups.
The patients with chronic refractory pump failure ir-~~~~~~Z21a T0 30 40

not due to ischemia had similar hemodynamic re- PCW mmHg

sponses. Figure 4 illustrates the average shift in ven- ON 22 ALL GR. A

O

O- N 10 (NO MITR. INS.)

tricular function in all 22 patients and in various sub- 0--O N 5 (WITH MITR. INS .)t GRAI
4 GR A2
groups according to cause. The changes in CO and & N

D-_ N1 3 GR A3
PCWP were statistically significant in the patients
with chronic pump failure (both p < 0.001). The four FIGURE 4. Effects of isosorbide dinitrate (ISDN) on left
patients in whom pump failure was caused by car- ventricular function in all patients with chronic refractory
diomyopathy (group A2) and the three patients with failure (group Al, n 22) and in the different subgroups ac-
=

rheumatic valvular insufficiency (group A3) had a
response similar to that of patients with ischemic
pump failure.
Group B
The patients with acute myocardial infarction and
pump failure benefited markedly from i.v. ISDN. The
cording to etiology; ischemic (group Al), cardiomyopathy
(group A2) and rheumatic (group A3). In group Al, the data
plotted separately for patients with mitral regurgitation
(MITR INS) and patients without evidence of mitral
regurgitation. Closed symbols represent control mean values
before ISDN and open symbols represent mean values dur-
ing the optimal ISDN dose.
 I.V. ISOSORBIDE DINITRATE/Rabinowitz et al. 775
6
TABLE 2. Total Plasma Catecholamines
Control ISDN
Group A (chronic)
E
1 1870 750
_~ 4 2 1670 448
3 265 170
0
u
3 4 168 110
5 1650 560
6 450 275
2
7 170 160
1 MEAN LEVE,S
8 214 210
v

30
9 240 220
20 40
PCW - mmHg 10 500 325
FIGURE 5. Effects of isosorbide dinitrate (ISDN) on left 11 120 130
ventricular function curves in acute myocardial infarction 12 115 250
patients with pump failure (group B); mean and individual 13 280 275
levels before (open symbols) and during ISDN (closed sym- 14 300 250
bols) are shown. The decrease pulmonary capillary wedge 15
pressure (PCW) and the increase in cardiac output (CO) are
1420 140
both statistically significant (p < 0.001). Group B (acute)
1 300 230
Plasma Catecholamines 2 600 500
Twenty-two patients, 15 from group Al and seven 3 710 205
from group B, had serial determination of plasma cat- 4 140 110
echolamines during the daytime (table 2). 5 255
The initial levels of peripheral plasma catechola- 240
mines varied in both groups. In the patients who had 6 900 350
7
Downloaded from http://ahajournals.org by on February 24, 2021

high basic levels of circulating catecholamines, the 214 410

hemodynamic improvement was accompanied by a Values are ng/ 1.
decreased level of plasma catecholamines. Abbreviation: ISDN = isosorbide dinitrate.
Hemodynamics and Circulating Catecholamines compared simultaneous measurements of SVR with
in Chronic Ischemic Pump Failure the levels of circulating catecholamines in patients
Because the patients with a high initial SVR with chronic ischemic failure (group Al) before the
appeared to benefit more from the ISDN therapy, we initiation of ISDN therapy. The initial levels of
catecholamines correlated with the initial SVR values
5
(r = 0.538, p < 0.05) (fig. 7). The initial heart rate and
the levels of plasma catecholamines also correlated in
group Al (r = 0.702, p < 0.01) (fig. 8).
4C
Clinical Results
Each patient who improved hemodynamically also
E
improved clinically. Attempts to discontinue the
ISDN infusion led to the reappearance of pulmonary
congestion and worsening of the hemodynamics.
0
U Resumption of the drug was followed by a marked im-
2 provement. In 10 of 22 patients with chronic pump
failure (group A), we gradually decreased the i.v.
ISDN and replaced it with sublingual and oral ISDN.
L,x. 10 20
These 10 patients left the hospital much improved and
were followed as outpatients. Five patients who did
not improve during the study died in intractable pump
PCW - mmHg failure during the same hospitalization. Five patients
FIGURE 6. Effects of isosorbide dinitrate (ISDN) on mean improved temporarily, but later required combined
levels of pulmonary capillary wedge pressure (PCW) and therapy with dopamine and vasodilators; only two sur-
cardiac output (CO) in patients with acute pump failure vived. One of the latter was a patient with rheumatic
(group B, n 18,; * = control; 0 = during ISDN) and in
=

mitral and aortic regurgitation who later underwent

those with chronic pump failure (group A, n = 15; * con- =
surgery. The remaining two patients, who had rheu-
trol; = during ISDN).
0
matic valvular insufficiency, improved during i.v.
 776
., 3200
I 1-
31
u 3000
0
z S
0
0
*
S
V40 5(05 C
A
z 2000
0 creases in CO in patients with severe pump failure
.
u
00I 0
I provement in chronic patients, some patients did not
I
300 500
PLASMA CATECP4OLAMINES ng/L
FIGURE 7. Correlation between initial systemic vascular
resistance (SVR) and simultaneous circulating catechola-
mines (CA) in 15 patients with chronic ischemic pump
failure (group AI) (r = 0.53, p < 0.05).
therapy and were sent to surgery while receiving i.v.
ISDN. In 14 of 18 patients with acute myocardial in-
farction and pump failure (group B), we gradually dis-
continued i.v. ISDN after 2-5 days and replaced it
with oral ISDN or with diuretics alone. These pa-
tients left the hospital without any signs or symptoms
of heart failure. The four patients who did not improve
during the ISDN infusion subsequently received com-
bined therapy with i.v. hydralazine and dopamine;
Downloaded from http://ahajournals.org by on February 24, 2021
three died in cardiogenic shock and one survived and
was discharged from hospital on hydralazine, diuretics
and digoxin.
Discussion
There have been few reports on the use of i.v.
ISDN,6-8 and most of the data are from patients with
acute infarction or unstable angina.6 6, 8 We studied
the hemodynamic effects of ISDN administered by i.v.
infusion to 40 patients with severe pump failure and
obtained the following data.
Intravenous ISDN not only markedly reduced fill-
ing pressures, but also tended to increase CO mod-
erately (figs. IA and lB and 3-6). There were no sharp
decreases in arterial pressure (fig. 1C), which could
lead to further enhancement of ischemia. In severe
pump failure due to acute myocardial infarction, the
hemodynamic effects were similar to those due to
chronic severe pump failure (figs. 5 and 6). Intra-
venous ISDN was useful in the presence of valvular in-
sufficiency, including mitral regurgitation due to
ischemia (fig. 4).
Vasodilators are generally classified as venodila-
tors acting on preload, arteriolardilators acting on
afterload, and mixed agents acting on both sides of the
peripheral vascular tree; nitrates are in the first group
of agents, which act only on capacitance vessels.3' 9-11, 26
However, several investigators agree that patients
in severe distress who have a low initial CO and mark-
edly increased filling pressure might respond differ-
ently.'2' 118,16, 2-28 Our data and clinical impression
confirm these conclusions.
CIRCULATION VOL 65, No 4, APRIL 1982
0
Goldberg and colleagues29 pointed out that patients
with an initially high SVR responded to ISDN by
0 shifting the left ventricular function curve upward and
to the left, an effect similar to that of nitroprusside.
However, the patients studied by these authors had
pump failure due to valvular insufficiency. Our data
showed that i.v. ISDN also produced moderate in-
without valvular insufficiency.
Although the general hemodynamic effects of i.v.
ISDN demonstrated a statistically significant im-
benefit (figs. 3 and 5) and, in a few, PCWP did not
N = l5
decrease.
.M-
000 IAL-
1500
Armstrong et al.30 31 observed that some patients
with heart failure may respond to nitroglycerin only
with a reduction in right atrial pressure without sig-
nificant change in left ventricular filling pressures. Ac-
cording to their data, the lack of hemodynamic
response to nitrate preparations in certain patients
cannot be predicted from control hemodynamic vari-
ables, but can be anticipated by previous i.v. nitro-
glycerin infusion. However, Baligadoo et al.32 claim
that the initial value of the SVR can predict the effect
of the nitrate on CO. Our data agree with theirs.
The patients in whom CO increased significantly in
response to i.v. ISDN were those who had a high ini-
tial SVR (fig. 2). Patients who had low-to-normal ini-
tial SVR levels appeared to be poor responders to
therapy, in the acute patients as well as in the chronic
group. It appears, therefore, that initial SVR levels
might predict the subsequent response to therapy.
However, to prove the role of SVR as a predictor, a
larger homogenous population with pump failure
should be studied.
There appears to be an advantage of i.v. ISDN over
nonparenteral nitrates. The use of a continuous infu-
sion permits a homogenous plasma level and avoids
0
.
z
I
.
N = 15
300 500 1000 500
PLASMA CATECHOLAMINES
2000
ng/L
-
FIGURE 8. Correlation between initial heart rate (HR) and
simultaneous plasma catecholamines (CA) in 15 patients
with chronic ischemic pump failure (group Al) (r = 0.70, p
< 0.01).
 I.V. ISOSORBIDE DINITRATE/Rabinowitz et al.
sharp fluctuations by a sublingual or even oral
preparation.933 34 Moreover, it seems logical to start
i.v. therapy with the same agent intended for later
nonparenteral administration in chronic patients.
In patients with ischemic heart disease, i.v. ad-
ministration under continuous hemodynamic moni-
toring may be preferable. Although we do not have
data to directly compare the use of i.v. ISDN with i.v.
nitroglycerin, Bussman et al."5 showed that i.v. nitro-
glycerin tends to produce sharp decreases in AP. This
reduction in AP might enhance preexisting ischemia.
Further studies are needed to evaluate this potential
difference between nitrate preparations, because only
limited data are available.6 A randomized comparison
of i.v. ISDN and i.v. nitroglycerin would be of in-
terest.
The catecholamine measurements provided in-
teresting correlations with the hemodynamic data.
Armstrong and colleagues30 took into consideration
variations in the levels of circulating catecholamines
or of angiotensin that might alter the response of the
peripheral circulation in failure. Our limited data
showed a relationship between the initially high SVR
and levels of circulating catecholamines in ischemic
pump failure (fig. 7). Another interesting finding is the
correlation between the initial heart rates and simul-
taneous levels of catecholamines in patients with
chronic, refractory ischemic failure (fig. 8). A possible
mechanism is that the initial increases in both heart
rate and SVR are due to the release of large amounts
Downloaded from http://ahajournals.org by on February 24, 2021
of catecholamines; the increase in heart rate is a
response to the low CO and is mediated through
catecholamines. It would be of interest to study hyper-
kinetic acute myocardial infarction patients,36' 7 a
subgroup that includes patients without pump failure
who have an exaggerated hypertensive-tachycardic
response. Accumulation of additional metabolic data
in larger groups of chronic and acute patients with
different hemodynamic profiles is necessary to
evaluate the significance of these measurements and
the value of circulating catecholamines levels as
predictors of therapeutic response.
Our data demonstrate favorable effects of i.v.
ISDN in patients with severe failure manifested by in-
creased filling pressures and elevated SVR. The use of
i.v. ISDN might be preferable to the use of other i.v.
vasodilators, particularly in patients with ischemic
pump failure.
Acknowledgment
The authors are indebted to Uri Goldbourt for the statistical
analysis, the residents and nurses of the intensive coronary care unit
who conscientiously supervised these patients, and to R. Klein,
M.Sc., for technical help in the biochemical work. Isosorbide dini-
trate (Isoket) was supplied by Pharma-Schwarz GmbH, Monheim,
West Germany.
References
1. Chatterjee K, Swan HJC, Kaushik VS, Jobin G, Magnusson P,
Forrester JS: Effects of vasodilator therapy for severe pump
failure in acute myocardial infarction on short term and late
prognosis. Circulation 53: 797, 1976
2. Kovick RB, Tillisch JH, Berens SC, Bramowitz AD, Shine KI:
Vasodilator therapy for chronic left ventricular failure. Circula-
777
tion 53: 322, 1976
3. Mikulic E, Franciosa JA, Cohn JN: Congestive hemodynamic
effects of chewable isosorbide dinitrate and nitroglycerin in pa-
tients with congestive heart failure. Circulation 52: 477, 1975
4. Franciosa JA, Pier Point G, Cohn JN: Hemodynamic improve-
ment after hydralazine in left ventricular failure: a comparison
with nitroprusside infusion in 16 patients. Ann Intern Med 86:
388, 1977
5. Luther M, Roken U: Die Wirksammiet von Isosorbiddinitrat
intravenos bei Angina pectoris und frischen Myokardinfarct.
Herz/Kreislauf 8: 654, 1976
6. Milstrey HR, Kaiser H, Kruezer H: Wirkungsvergleich von
nitroglycerin i.v. und ISDN i.v. beim frischen Myokardinfarct.
In Nitrate II, Wirkung auf Herz und Kreislauf, edited by
Rudolph W, Schrey A. Munchen, Urban und Schwarzenberg,
1980, p 299
7. Rabinowitz B, Tamari I, Neufeld HN: Use of intravenous
isosorbide dinitrate (Isoket) in severely ill cardiac patients with
heart failure. Herz 3: 206, 1978
8. Distante A, Maseri A, Severi S, Biagni A, Chierchia S:
Management of vasospastic angina at rest with continuous infu-
sion of isosorbide dinitrate. A double crossover study in a cor-
onary care unit. Am J Cardiol 44: 533, 1979
9. Massie B, Chatterjee K, Werner J, Greenberg B, Hart R,
Parmley WW: Hemodynamic advantage of combined ad-
ministration of hydralazine orally and nitrate nonparenterally
in the vasodilator therapy of chronic heart failure. Am J Car-
diol 40: 794, 1978
10. Chatterjee K, Parmley WW: The role of vasodilator therapy in
heart failure. Prog Cardiovasc Dis 19: 301, 1977
11. Williams DO, Amsterdam EA, Mason DT: Hemodynamic
effect of nitroglycerin in acute myocardial infarction. Decrease
in ventricular preload at the expense of cardiac output. Circula-
tion 51: 421, 1975
12. Bussman WD, Kaltenbach M: Sublingual nitroglycerin in the
treatment of left ventricular failure and pulmonary edema. Eur
J Cardiol 4: 327, 1976
13. Gold HK, Leinbach RC, Sanders CA: Use of sublingual nitro-
glycerin in congestive heart failure following acute myocardial
infarction. Circulation 46: 839, 1972
14. Stephens J, Camm J, Spurrell R: Improvement in exercise
hemodynamics by isosorbide dinitrate in patients with severe
congestive cardiac failure secondary to ischemic heart disease.
Br Heart J 40: 832, 1978
15. Franke N, Schmuker P, van Ackern K, Peter K: Treatment of
post-operative left heart failure. (abstr) Intensive Care Med 6:
73, 1980
16. Franciosa JA, Blank RC, Cohn JN: Nitrate effects on cardiac
output and left ventricular outflow resistance in chronic conges-
tive heart failure. Am J Med 64: 207, 1978
17. Flaherty JT, Come PC, Baird MG, Rouleau J, Taylor DR,
Weisfeldt ML, Greene HL, Becker LC, Pitt B: Effects of intra-
venous nitroglycerin on left ventricular function and ST seg-
ment changes in acute myocardial infarction. Br Heart J 38:
612, 1976
18. Miller RR, Awan NA, DeMaria AN, Amsterdam EA, Mason
DT: Importance of maintaining systolic blood pressure during
nitroglycerin administration for reducing ischemic injury in pa-
tients with coronary disease: effects on coronary blood flow
myocardial energetics and left ventricular function. Am J Car-
diol 40: 504, 1977
19. Packer M, Meller J: Oral vasodilator therapy for chronic heart
failure: a plea for caution. Am J Cardiol 42: 686, 1978
20. Shillingford JP: The estimation of severity of mitral regurgita-
tion. Prog Cardiovasc Dis 5: 248, 1962
21. Scheinman M, Benham B: Clinical and anatomic implications
of intraventricular conduction blocks in acute myocardial in-
farction. Circulation 46: 753, 1972
22. Forrester JS, Ganz W, Diamond G, McHugh T, Chonette DW,
Swan HJC: Thermodilution cardiac output determination with
a single flow-directed catheter. Am Heart J 83: 306, 1972
23. Renzini V, Brunori CA, Valori C: A sensitive and specific
fluorimetric method for determination of noradrenalin in
human plasma. Clin Chim Acta 30: 587, 1970
24. Jiang NS, Stoffer SS, Piler GM, Wadel 0, Sheps SG:
Laboratory and clinical observations with a two-column cate-
 778 CIRCULATION VOL 65, No 4, APRIL 1982

cholamine assay. Mayo Clin Proc 48: 47, 1973
25. Rabinowitz B, Elazar E, Orenstein A, Neufeld HN: Plasma
catecholamines and cyclic AMP levels in experimental atrial
tachycardia and after coronary occlusion. In Neural
Mechanisms in Cardiac Arrhythmias, edited by Schwartz PJ,
Brown AM, Malliani A, Zanchetti A. New York, Raven Press,
1978, p 301
26. Mason DT: Afterload reduction and cardiac performance.
Physiologic basis of systemic vasodilators as a new approach in
treatment of congestive heart failure. Am J Med 65: 106, 1978
27. Parmley WW, Chatterjee K: Vasodilator Therapy, Current
Problems in Cardiology Series, Chicago, Year Book Medical
Publishers, 1978, p 75
28. Williams DO, Bommer WJ, Miller RR, Amsterdam EA,
Mason DT: Assessment of oral peripheral vasodilator therapy
in chronic congestive heart failure: prolonged effectiveness of
isosorbide dinitrate. Am J Cardiol 39: 84, 1977
29. Goldberg S, Mann T, Grossman W: Nitrate therapy of heart
failure in valvular heart disease. Importance of resting level of
peripheral vascular resistance in determining cardiac output
response. Am J Med 65: 161, 1978
30. Armstrong PW, Armstrong JA, Marks GS: Pharmacokinetic-
hemodynamic studies of nitroglycerin ointment in congestive
heart failure. Am J Cardiol 46: 670, 1980
31. Armstrong PW, Armstrong JA, Marks GS: Pharmacokinetic-
hemodynamic studies of intravenous nitroglycerin in conges-
tive cardiac failure. Circulation 62: 160, 1980
32. Baligadoo S, Ingrano HC, Derruda JP, Savier CH, Chiche P:
Etude des effets hemodynamiques de la trinitrine intraveineuse.
Coeur Med Int 17: 231, 1978
33. Needleman P, Lang S, Johnson EM Jr: Organic nitrates: rela-
tionship between biotransformation and rational angina pec-
toris therapy. J Pharmacol Exp Ther 181: 489, 1972
34. Taylor T, Chasseand LF, Doyle E: Pharmacokinetics of isosor-
bide dinitrate after intravenous infusion in human subjects.
Biophys Drug Disp 1: 149, 1980
35. Bussman WD, Lohner J, Kaltenbach M: Orally administered
isosorbide dinitrate in patients with and without left ventricular
failure due to acute myocardial infarction. Am J Cardiol 39: 91,
1977
36. Amsterdam EA, Awan NA, DeMaria AN, Mason DT: Vaso-
dilators in myocardial infarction: rationale and current status.
Drugs 16: 506, 1978
37. Forrester JS, Diamond G, Chatterjee K, Swan HJC: Medical
therapy of acute myocardial infarction by application of hemo-
dynamic subsets. N Engl J Med 295: 1356, 1976

Pathophysiology of Chest Pain in Patients

with Cardiomyopathies and Normal Coronary Arteries
ANDRt PASTERNAC, M.D., JACQUES NOBLE, M.D., YVES STREULENS, M.D.,
ROBERT ELIE, M.D., PH.D., CLAUDIA HENSCHKE, M.D., PH.D.,
Downloaded from http://ahajournals.org by on February 24, 2021

AND MARTIAL G. BOURASSA, M.D.

SUMMARY To clarify the pathogenesis of chest pain in patients with cardiomyopathies, we compared
coronary blood flow and other indicators of ischemia at rest and during pacing-induced tachycardia in nine pa-
tients with cardiomyopathy (four hypertrophic and five congestive) and in five control subjects. Coronary blood
flow was reduced at rest and during pacing in cardiomyopathy patients compared with controls. In patients
with hypertrophic cardiomyopathy, pacing induced chest pain in all, increased ST-segment depression in three
patients and increased coronary venous lactate concentration. With pacing, two of five patients with congestive
cardiomyopathy had chest discomfort and three had increased ST-segment depression, but coronary venous
lactate concentration did not change significantly. In both groups of cardiomyopathies, the ratio of the systolic
and diastolic pressure-time indexes tended to decrease more than in controls during pacing. Thus, myocardial
perfusion is decreased in patients with cardiomyopathy, both at rest and during pacing. The changes detected
during pacing point to subendocardial ischemia as the likely mechanism for angina in hypertrophic and
possibly also in congestive cardiomyopathy.

CHEST PAIN is a frequent symptom in patients with
cardiomyopathy: 39-72% of patients with idiopathic
hypertrophic cardiomyopathy experience anginal
pain,"' 2 and as many as 52% of patients with conges-
tive cardiomyopathy report vague chest pain,- despite
the presence of normal or even dilated coronary
arteries.4
From the Department of Medicine, Montreal Heart Institute and
University of Montreal Medical School, and Louis-H. Lafontaine
Hospital, Montreal, Quebec, Canada, and the Department of
Radiology, Harvard Medical School and Brigham and Women's
Hospital, Boston, Massachusetts.
Supported in part by the J.L. Levesque Foundation.
Address for correspondence: Andre Pasternac, M.D., Montreal
Heart Institute, 5000 east, Belanger Street, Montreal, Quebec, HIT
1C8, Canada.
Received May 18, 1978; revision accepted July 2, 1981.
Circulation 65, No. 4, 1982.
However, although myocardial blood flow per unit
mass has already been studied in such patients,5-8 the
pathophysiology of chest pain and its relationship to
the usual markers of ischemia are not clear. There-
fore, we measured myocardial perfusion at rest and
during pacing in patients with hypertrophic and con-
gestive cardiomyopathy and compared clinical and
biochemical indicators of ischemia both at rest and
during pacing in hypertrophic and congestive cardio-
myopathies and in a control group.
Methods
Patient Selection
We reviewed the echocardiographic and hemo-
dynamic data of all patients in whom cardiomyop-
athies were diagnosed during 1 year at the Montreal