Effect of Prazosin vs Placebo

on Chronic Left Ventricular Heart Failure

WILBERT S. ARONOW, M.D., MARK LURIE, M.D., MARVIN TURBOW, M.D., PH.D.,
KENNETH WHITTAKER, PHARM. D., M.D., STEVEN VAN CAMP, M.D.,
AND DAVID HUGHES, M.D.

SUMMARY The effect of the vasodilator prazosin vs placebo on exercise duration until marked dyspnea,
and on left ventricular function measured by echocardiography, was evaluated in a double-blind, randomized
study in 24 patients with chronic left ventricular failure despite digitalis and diuretic therapy. Compared with
the double-blind placebo, prazosin reduced resting systolic and diastolic blood pressure and systolic blood
pressure times heart rate, improved clinical symptoms, decreased cardiothoracic ratio measured by chest
roentgenography, decreased left ventricular and left atrial dimensions, improved ejection fraction and Vcf
measured by echocardiography, and improved treadmill exercise duration. All 12 patients taking prazosin had
> 20% improved treadmill exercise duration; none of 12 receiving placebo improved. In six of 12 patients tak-
ing prazosin, roentgenographic evidence of pulmonary venous congestion disappeared compared with none of
the patients on placebo. These data suggest that prazosin may be effective in treating chronic left ventricular
failure.

ORAL, SUBLINGUAL AND TOPICAL vasodila-
tors are important in the management of severe heart
failure.' '7 By reducing ventricular preload, nitrates
relieve pulmonary congestion.'`8 By reducing ventric-
ular afterload, oral hydralazine improves cardiac out-
put." 8-12
A combination of nitrates and hydralazine may
relieve dyspnea by reducing pulmonary congestion
and decrease fatigue by increasing cardiac output.5, 8
However, chronic hydralazine therapy may be
associated with induction of the systemic lupus
erythematosus syndrome.
Oral prazosin has balanced vasodilator effects on
the systemic arterial and venous beds.'2-55 Preliminary
data also show that prazosin improves exercise dura-
tion and echocardiographic measurements of ventric-
ular function.15
We performed a double-blind, randomized study to
evaluate the effect of prazosin vs placebo on treadmill
exercise performance and on ventricular function,
measured by echocardiography, in 24 patients with
chronic left-sided congestive heart failure unrespon-
sive to digitalis and diuretic therapy.
Methods
The subjects were 24 men 26-67 years of age with
chronic left-sided congestive heart failure. Fifteen had
documented coronary heart disease. Cardiac
catheterization demonstrated significant multivessel
coronary artery disease and poor left ventricular func-
tion in these subjects. Thirteen of the 15 patients with
From the Cardiology Section, Long Beach Veterans Administra-
tion Hospital, and the University of California College of Medicine.
Irvine, California.
Address for reprints: Wilbert S. Aronow, M.D., Veterans Ad-
ministration Hospital. Long Beach, California 90822.
Received April 27, 1978; revision accepted September 13, 1978.
Circulation 59, No. 2, 1979.
344
Downloaded from http://circ.ahajournals.org/ by guest on April 7, 2016
coronary disease had a documented previous
transmural myocardial infarction; two had rheumatic
heart disease with severe mitral insufficiency and poor
left ventricular function demonstrated by cardiac
catheterization; four had idiopathic congestive cardio-
myopathy with normal coronary angiograms and
diffusely impaired left ventricular contractility; three
had alcoholic congestive cardiomyopathy with normal
coronary angiograms and diffusely impaired left ven-
tricular contractility.
All 24 patients had cardiomegaly on physical ex-
amination and roentgenography, a left ventricular
third heart sound on auscultation, and evidence of
pulmonary venous congestion. Baseline standing
blood pressures ranged from 110-152 mm Hg systolic
and from 70-100 mm Hg diastolic. The patients had
taken digitalis and diuretic therapy for 2-52 months.
Of the 12 patients randomized to prazosin, 11 took
furosemide 40-240 mg daily, and one took hydro-
chlorothiazide 100 mg daily. Of the 12 patients ran-
domized to double-blind placebo, 11 took furosemide
40-160 mg daily and one took hydrochlorothiazide
100 mg daily. Their doses of these drugs were not
changed during the study. Except for potassium
chloride, no other medications were taken by any
patient during this study. All patients signed informed
research consent forms.
After baseline measurements, all 24 patients took
one capsule of single-blind placebo three times daily
for 2 weeks. Then, in a double-blind, randomized
fashion, all patients took prazosin or placebo for 6
more weeks. The prazosin and placebo capsules
looked identical. Twelve patients took one placebo
capsule three times daily for weeks 1 and 2, two
placebo capsules three times daily for weeks 3 and 4,
and three placebo capsules three times daily for weeks
5 and 6 of the double-blind study period. Twelve
patients received one 0.5 mg prazosin capsule three
times daily for week 1, one 1 mg prazosin capsule
three times daily for week 2, two 1 mg prazosin cap-
sules three times daily for weeks 3 and 4, and three 1
 PRAZOSIN IN HEART FAILURE/Aronow et al.
mg prazosin capsules three times daily for weeks 5 and
6 of the double-blind study.
Each patient answered a standard questionnaire
regarding symptoms, had a physical examination in
the baseline period, at the end of 2 weeks of the single-
blind placebo regimen, and at the end of weeks 1, 2, 3,
4, 5 and 6 in the double-blind study period. A standard
12-lead ECG and posteroanterior and lateral chest
roentgenograms were taken in the baseline period, at
the end of 2 weeks of the single-blind placebo period,
and at the end of 3 weeks and 6 weeks of the double-
blind study period.
Echocardiograms were performed in all patients in
the baseline period and approximately 105 minutes
after their morning dose of medication after 2 weeks
of single-blind placebo and after 3 and 6 weeks of
double-blind medication. Echocardiography was per-
formed in all patients with an Ekoline-20 ultrasono-
scope with a 0.5 inch diameter 2.25 MHz transducer
focused at 10 cm with a repetition rate of 1,000 im-
pulses/sec, and an Irex ContinuTrace 101 multi-
channel recorder. All echocardiograms were recorded
with the subjects in a slight left lateral decubitus posi-
tion with 150 elevation of the head. ECGs were
recorded simultaneously with the echocardiograms.
The heart rates were measured from the ECGs.
We selected a transducer position from which we
could consistently and readily obtain distinct left ven-
tricular dimensions. We determined the transducer
position which gave the optimal mitral valve
echogram. The transducer was then angled slightly
laterally and inferiorly to obtain clear left ventricular
septal and posterior wall endocardial echoes. In this
position, the mitral valve echogram was discontinuous
and the posterior leaflet was usually better visualized.
Using the echoes from the mitral valve apparatus as
landmarks, we obtained reproducible left ventricular
dimensions.18-21 Recordings were made at the constant
respiratory phase of held mid-expiration. The left ven-
tricular diameter at end-systole (LVESD) was
measured at the point of least separation of the septal
and posterior wall endocardial echoes. The left ven-
tricular diameter at end-diastole (LVEDD) was
measured at the peak of the R wave in the ECG.
After recognition of the characteristic mitral valve
echoes, the transducer was angled medially,
posteriorly, and superiorly to record the aortic root
and left atrium. Measurements of left atrial dimen-
sion were made at ventricular end-systole between the
external surface of the posterior aortic root and the in-
ternal surface of the left atrial wall.
The ejection time (ET) was calculated as the time
from the beginning of the QRS complex to maximal
anterior systolic motion of the left ventricular
posterior wall minus 50 msec for the preejection
period.'5' 22 The left ventricular volume at end-systole
(LVESV) and at end-diastole (LVEDV) were
calculated by the cube of their respective diameters."9
Stroke volume (SV), cardiac index (CI), ejection frac-
tion (EF), and mean rate of left ventricular circum-
ferential fiber shortening (Vcf) were derived as
follows:
Downloaded from http://circ.ahajournals.org/ by guest on April 7, 2016
345
SV = LVEDV- LVESD
ci heart rate X SV
=
1,000 X body surface area
EF - SV
LVEDV
Vcf= LVEDD- LVESD
ET X LVEDD
All patients performed two practice treadmill tests
within 1 week before the baseline period. After echo-
cardiographic estimation of ventricular function, the
patients performed a multistage, uninterrupted max-
imal treadmill exercise test until the onset of marked
dyspnea in the baseline period, approximately 2 hours
after their morning dose of medication after 2 weeks
of the single-blind placebo study and after 3 and 6
weeks of the double-blind medication study. Marked
dyspnea was the limiting factor in all maximal tread-
mill exercise tests in all patients. The patients exer-
cised at a treadmill speed of 1.7 mph and a treadmill
grade of 0% for the first 3 minutes, a treadmill speed
of 1.7 mph and a treadmill grade of 10% for the next 3
minutes, and then at a treadmill speed of 2.5 mph and
a treadmill grade of 12%.
The patients were monitored with telemetry with
simultaneous leads II and V5 throughout exercise.
Simultaneous leads II and V5 and blood pressures
were recorded with the patients sitting and standing
immediately before exercise, with the patients stand-
ing at the end of each completed stage of exercise and
at the end of exercise, and with the patients in both sit-
ting and standing positions every minute for at least 5
minutes after exercise. Blood pressures were recorded
with a mercury sphygmomanometer. The heart rates
were measured from the electrocardiographic record-
ings. Blood pressures were obtained by the same
technician. The cardiology fellows supervising the ex-
ercise tests did not know the blood pressure readings
to maintain a double-blind exercise test. The chest
roentgenograms and ECGs were blindly coded and in-
terpreted at the end of the study.
The data listed in tables 1-4 were analyzed using
Tukey range tests. Fisher's exact test was used to
analyze the data on improvement of symptoms and
the data on changes in roentgenographic evidence of
pulmonary venous congestion.
Results
Pill counts revealed that all 24 patients had taken
their medication as prescribed. No adverse reactions
to medication occurred during the study.
Table 1 shows the hemodynamic values, exercise
duration, ST-segment depression and mean cardio-
thoracic ratio SD during the baseline period, after 2
weeks of single-blind placebo, and after 3 and 6 weeks
of double-blind placebo. It also shows the causes of
the congestive heart failure in the patients randomized
to double-blind placebo.
Table 2 indicates the same data for the baseline
346 CIRCULATION VOL 59, No 2, FEBRUARY 1979

TABLE 1. Hemodynamic Values, Exercise Duration, ST-Segment Depression and Cardiothoracic Ratio During Baseline, Single-
Blind Placebo and Double-Blind Placebo Periods (Mean 1 SD)
Single-blind Double-blind placebo
Parameter Baseline placebo 3 Weeks 6 Weeks
Resting heart rate (beats/min) 83.8 9.9 85.1 i7.6 84.1 i10.4 84.1 = 10.4
Resting systolic blood pressure (mm Hg) 124.2 11.8 125.7 9.2 127.3 10.1 126.2 9.9
Resting diastolic blood pressure (mm Hg) 82.3 8.4 82.4 6.5 84.2 7.4 83.7 8.3
Resting systolic pressure times heart rate/100 104.2 16.9 107.1 14.0 107.3 17.0 106.2 16.1
Exercise heart rate (beats/min) 133.1 l 14.0 133.6 - 14.5 133.5 + 13.2 134.2 14.8
Exercise systolic blood pressure (mm Hg) 148.7 - 12.5 149.2 - 13.2 149.8 13.4 149.2 13.7
Exercise diastolic blood pressure (mm Hg) 90.0 - 6.6 90.4 - 5.9 90.5 +7.0 90.3 6.0
Exercise systolic pressure times heart rate/100 197.4 - 22.6 198.8 = 22.8 199.5 20.8 199.6 23.9
Exercise duration (sec) 211.9 - 56.0 221.2 - 59.7 218.7 54.5 224.4 60.5
Maximal ST depression below resting level (mm) 1.42 = 0.47 1.48 - 0.45 1.48 0.44 1.50 0.44
Cardiothoracic ratio 0.58 = 0.04 0.58 - 0.04 0.58 f0.04 0.58 +0.04
Eight patients had coronary heart disease; two had congestive cardiomyopathy; two had rheumatic heart disease with severe
mitral insufficiency and poor left ventricular function.

period, after 2 weeks of single-blind placebo, after 3
weeks of prazosin (6 mg daily), and after 6 weeks of
prazosin (9 mg daily). It also indicates the causes of
the congestive heart failure in the patients randomized
to prazosin and statistical levels of significance.
Figure illustrates the percent change in exercise
duration for each patient after 6 weeks of prazosin
therapy from the average of the baseline and single-
blind placebo periods. All 12 patients randomized to
prazosin had 20% improvement in exercise dura-
tion (range 22-97%).
Figure 2 shows the percent change in exercise dura-
tion for each patient after 6 weeks of double-blind
placebo therapy from the average of the baseline and
single-blind placebo periods. None of the 12 patients
randomized to double-blind placebo had > 20% im-
provement in exercise duration.
Table 3 lists the echocardiographic data during the
baseline period, after 2 weeks of single-blind placebo,
and after 3 and 6 weeks of double-blind placebo. Table
4 indicates the same data for the baseline period, after
2 weeks of single-blind placebo, after 3 weeks of

TABLE 2. Hemodynantic Values, Exercise Duration, ST-Segment Depression and Cardiothoracic Ratio During Baseline, Single-
Blind Placebo and Prazosin Periods (Mean 1 sD)
Single-blind Double-blind prazosin
Parameter Baseline placebo 3 Weeks 6 Weeks
Resting heart rate (beats/min) 90.3 = 11.2 =
89.6 9.5 86.7 8.2i 86.8 = 7.3
Resting systolic blood pressure (mm Hg) 122.8 d 8.8 124.7 d 9.2 113.3 7.9* 108.5 7.8*
Resting diastolic blood pressure (mm Hg) 83.0 7.5 83.0 6.5 75.6 i 6.1* 72.3 6.4*
Resting systolic pressure times heart rate/100 110.8 14.3 111.6 13.7 98.2 lil.t 94.3 10.9*
Exercise heart rate (beats/min) 129.2 - 10.2 131.3 - 9.7 131.1 - 9.1 133.2 - 10.1
Exercise systolic blood pressure (mm Hg) 148.7 - 11.2 149.8 - 13.2 147.7 - 11.6 147.9 - 12.6
Exercise diastolic blood pressure (mm Hg) 88.3 - 8.3 88.8 = 8.5 87.5 - 7.6 86.8 - 7.7
Exercise systolic pressure times heart rate/100 192.4 = 24.6 197.2 - 26.6 193.9 - 23.6 197.3 - 26.0
Exercise duration (sec) 213.3 - 54.9 223.4 - 59.0 291.7 - 71.8t 342.1 - 81.4*
Maximal ST depression below resting level (mm) 1.42 = 0.25 1.37 - 0.25 1.24 0.26 1.33 - 0.27
Cardiothoracic ratio 0.58 i 0.04 0.58 - 0.04 0.57 *0.04 0.56 i 0.04t
*p <0.001 for prazosin at 3 weeks minus an average of its baseline and single-blind placebo periods, compared with double-
blind placebo at 3 weeks minus an average of its baseline and single-blind placebo periods; or for prazosin at 6 weeks minus an
average of its baseline and single-blind placebo periods, compared with double-blind placebo at 6 weeks minus an average of its
baseline and single-blind placebo periods.
tP <0.005 for prazosin at 3 weeks minus an average of its baseline and single-blind placebo periods, compared with double-
blind placebo at 3 weeks minus an average of its baseline and single-blind placebo periods; or for prazosin at 6 weeks minus an
average of its baseline and single-blind placebo periods, compared with double-blind placebo at 6 weeks minus an average of its
baseline and single-blind placebo periods.
tp <0.01 for prazosin at 3 weeks minus an average of its baseline and single-blind placebo periods, compared with double-bliind
placebo at 3 weeks minus anl average of its baseline and single-blind placebo periods.
Seven patients had coronary heart disease; five had congestive cardiomyopathy.

Downloaded from http://circ.ahajournals.org/ by guest on April 7, 2016

 PRAZOSIN IN HEART FAILURE/Aronow et al. 347

+100. z
0
+20-
0
+16 -
+ 90- +1 2- 0

z w
+ 8-
0
C) + 4-
< + 80- w .
0-

0 - 4- 0

w * 70- - 8-
co *
0 .
z
-12-
I
w
+ 60- CD
0
-16-
w _-n
I -
S'
z
0
2 4 6 8 10 12
+ 50-
w PATIENT NUMBER
z 0
FIGURE 2. Percent change in exercise duration for each
M 40- 0
patient after 6 weeks of double-blind placebo therapy from
average of baseline and single-blind placebo periods.
-0
0-
+ 30-

Figure 3 is the chest roentgenogram during the

a
0 baseline period in a patient randomized to prazosin
20-
therapy, and figure 4 is the chest roentgenogram in the
2 4 6 8 same patient after 6 weeks of prazosin therapy.
Baseline characteristics between the patients ran-
PATIENT NUMBER domized to prazosin and to double-blind placebo
FIGURE 1. Percent change in exercise duration for each showed inequalities in four of the 19 parameters listed
patient after 6 weeks of prazosin therapy from average of in tables 1 and 2 EF, (p < 0.01), LVEDD
-

baseline and single-blind placebo periods.
prazosin (6 mg daily), and after 6 weeks of prazosin (9
mg daily). Table 4 also lists statistical levels of
significance.
All 12 patients randomized to prazosin had good or
excellent relief of symptoms; none of the 12 patients
randomized to placebo achieved these results
(p < 0.001). Six of 12 patients randomized to prazosin
had roentgenographic evidence of disappearance of
pulmonary venous congestion, while none of 12
patients randomized to placebo had these results
(p = 0.007).
(p < 0.01), LVESD (p < 0.01), and Vcf (p < 0.01).
However, the inequalities in these four baseline
characteristics did not influence the therapeutic
response to prazosin.
Discussion
Prazosin has been shown to cause a prolonged dilat-
ing action on both the arterial and venous systems.2' 14
As a result, it relieves congestive heart failure by
reducing left ventricular filling pressure and by reduc-
ing impedance with augmentation of cardiac out-
put.'2' 14 The studies by Miller and associates14 and
Mehta and associates12 and this study also show that

TABLE 3. Echocardiographic Data During Baseline, Single-Blind Placebo and Double-Blind Placebo Periods
(Mean - 1 SD)
Single-blind Double-blinid placebo
Parameter Baseline placebo 3 Weeks 6 Weeks
LVEDD (cm) 5.5 0.9
- 5.5 0.9 5.5 0.9 5.5 - 0.9
LVESD (cm) 4.5 - 0.9 4.6 0.9 4.5 0.9 4.6 - 0.9
LAD (cm) 4.4 0.8 4.3 0.7 4.4 0.8 4.4 0.8
Systolic ejection time (msec) 237.9 29.2 235.0 29.7 237.5 34.5 238.3 32.8
Stroke volume (ml) 75.5 31.6 75.8 34.1 73.3 30.3 72.8 28.2
Cardiac index (1/min/m2) 3.28 1.42 3.26 1.49 3.14 = 1.25 3.13 1.22
Ejection fraction (%) 44.5 9.4 43.7 9.0 44.2 8.9 42.3 8.6
Vcf (circ/sec) 0.76 0.19 0.75 0.17 0.75 - 0.16 0.71 0.15
Abbreviations: LVEDD = left ventricular dimension at end-diastole; LVESD = left ventricular dimen-
sion at end-systole; LAD = left atrial dimension; Vcf = mean rate of left ventricular circumferential fiber
shortening.

Downloaded from http://circ.ahajournals.org/ by guest on April 7, 2016

348 CIRCULATION VOL 59, No 2, FEBRUARY 1979

TAB3LE 4. Echocardiographic Data During Baseline, Single-Blind Placebo and Double-Blind Prazosin Periods
(Mean 1 SD)
Single-blind Double-blind prazosin
Parameter Baseline placebo 3 Weeks 6 Weeks
LVEDD (cm) 6.4 0.5 6.5 0.6 6.4 0.6 6.2 0.61
LVESD (cm) 5.7 0 R5 5.7 - 0.6 5.5 - 0.6 5.4 0.6*
LAD (cm) 4.4 0.6 4.5 0.6 4.3 0.51 4.2 0.6t
Syst olic ejection time (msec) 230.0 33.2 228.8 i 25.1 229.2 26.4 227.1 24.7
Stroke volume (ml) 82.8 i 23.0 85.3 - 21.7 88.4 i 19.3 87$i~ 24.8
Cardiac index (1/min/M2) 3.79 1.30 3.84 i 1.26 3.95 M 1.08 3.90 1.29
Ejection fraction (C/0) 30.8 7.0 31.6 4.9
4 34.4 i 5.1 36.2 6.9t
Vcf (circ/sec) 0.51 i 0.12 0.52 i 0.10 0.58 =fi 0.10 0.62 0.13*
*p <0.001 for prazosin at 6 weeks minius an average of its baseline and single-blind placebo periods,
compared with double-blind placebo at 6 weeks minus an average of its baseltine and single blind placebo
periods.
tP <0.005 for prazosin at 6 weeks minus an average of its baseline and single-blind placebo periods, com-
pared with double-blind placebo at 6 weeks minus an average of its baseline and single-blinid placebo periods.
lp <0.05 for prazosin at 3 weeks minus an average of its baseline and single-blinid placebo periods, com-
pared with double-blind placebo at 3 weeks minus an average of its baseline and single-blind placebo periods;
or for prazosin at 6 weeks rninus an average of its baseline and single-blind placebo periods, compared with
double-blind placebo at 6 weeks minius aII average of its baseline and single-blind placebo periods.
Abbreviatioins: LVEDD left ventricular dimension at end-diastole; LVESI) left ventricular dimen-
=

sion at end-systole; LAD left atrial dimension: Vef = mean rate of left venitricular circumferential fiber
shortening.

prazosin does not significantly change heart rate in blood pressure times heart rate at rest and at the
patients with congestive heart failure. greatest common exercise load, and reduced the size
Our study demonstrated that in patients with of the heart, decreasing the myocardial oxygen de-
chronic left-sided congestive heart failure unrespon- mand. Improved ventricular function with reduced
sive to digitalis and diuretic treatment, prazosin effec-
tively lowered resting systolic and diastolic blood
pressure and product of systolic blood pressure times
heart rate, improved clinical symptoms, decreased the
size of the heart measured by chest roentgenography
and echocardiography, improved roentgenographic
evidence of pulmonary venous congestion, improved
ventricular function measured by echocardiography,
and improved exercise tolerance until marked
dyspnea. Our patients tolerated prazosin without side
effects. These data confirm the preliminary data
reported by Awan and associates.'" The 56% increase
in exercise duration after 6 weeks of prazosin in our
study is similar to the 52% improvement in exercise
duration resulting from prazosin therapy reported by
Awan and associates.'5
Patients with ischemic heart disease may have
regional wall motion abnormalities which may not
be detected by echocardiography. The echocardio-
graphic determination of left ventricular performance
and volumes in patients with ischemic heart disease
may be misleading. However, despite these limita-
tions, directional changes in indices of left ventricular
performance and volumes are probably valid.
The product of systolic blood pressure times heart
rate at the end of maximal exercise was similar after
prazosin and double-blind placebo, indicating that
prazosin probably causes no change in myocardial FIGURE 3. Posteroanterior chest roentgenogram during
oxygen supply. However, compared with double-blind baseline period in patient with cardiomegaly and pulmonary
placebo, prazosin decreased the product of systolic venous congestion.

Downloaded from http://circ.ahajournals.org/ by guest on April 7, 2016

 PRAZOSIN IN HEART FAILURE/Aronow et al.
FiGiURE 4. Posteroanterior chest roentgenogram after 6
weeks of prazosin therapy reveals a reduction in heart size
and disappearance of pulmonary venous congestion in the
patient shown in figure 3.
ventricular dimensions and left ventricular wall ten-
sion and decreased myocardial oxygen demand
relative to myocardial oxygen supply was probably
responsible for the increased exercise tolerance after
prazosin.
There were baseline inequalities in echocardio-
graphic measurements of EF, LVEDD, LVESD, and
Vcf between the prazosin and double-blind placebo
groups. The results of the single-blind placebo inter-
vention in both the prazosin and double-blind placebo
groups support the conclusion that these baseline in-
equalities did not influence the therapeutic response to
prazosin. However, to correct for these baseline im-
balances, we compared prazosin and double-blind
placebo withinr the framework of a one-way analysis of
covariance. This analysis showed that these baseline
inequalities did not alter our conclusions regarding
beneficial effects of prazosin vs double-blind placebo
on clinical symptoms, blood pressure, exercise perfor-
mance, cardiothoracic ratio, pulmonary venous con-
gestion, and left ventricular dimensions and function.
While the combination of hydralazine with long-
acting nitrates may relieve pulmonary congestion and
increase cardiac output,57 oral prazosin has the ad-
vantage of a single drug accomplishing the same
thing.'12 23 Moreover, prazosin does not have the
disadvantage of hydralazine's side effects.
Further data should define the role of vasodilating
drugs in the chronic treatment of left ventricular
Downloaded from http://circ.ahajournals.org/ by guest on April 7, 2016
349
failure. Double-blind hemodynamic and exercise
crossover studies must be performed to compare the
efficacy and side effects of the vasodilator agents.
Acknowledgment
We thank David S. Salsburg, Ph.D., for biostatistical analysis of
the data and Colin R. Taylor, M.B., Pfizer, Inc., for supplying the
prazosin and placebo.
References
I. Franciosa JA, Mikulic E, Cohn JN, Jose E, Fabie A:
Hemodynamic effects of orally administered isosorbide
dinitrate in patients with congestive heart failure. Circulation
50: 1020, 1974
2. Mikulic E, Franciosa JA, Cohn JN: Comparative hemody-
namic effects of chewable isosorbide dinitrate and nitroglycerin
in patients with congestive heart failure. Circulation 52: 477,
1975
3. Gray R, Chatteriee K, Vyden JK, Ganz W, Forrester JS, Swan
HJC: Hemodynamic and metabolic effects of isosorbide
dinitrate in chronic congestive heart failure. Am Heart J 90:
346, 1975
4. Kovick RB, Tillisch JH, Berens SC, Bramowitz AD, Shine KI:
Vasodilator therapy for chronic left ventricular failure. Circula-
tion 53: 322, 1976
5. Chatteriee K, Drew D, Parmley WW, Klausner SC, Polansky
J, Zacherle B: Combination vasodilator therapy for severe
chronic congestive heart failure. Ann Interil Med 85: 467, 1976
6. Taylor WR, Forrester JS, Magnusson P, Takano T, Chatterjee
K, Swan HJC: Heniodynamic effects of nitroglycerin ointment
in congestive heart failure. Am J Cardiol 38: 469, 1976
7. Williams DO, Bommer WJ, Miller RR, Amsterdam EA,
Mason DT: Hemodynamic assessment of oral peripheral
vasodilator therapy in chronic congestive heart failure.
Prolonged effectiveness of isosorbide dinitrate. Am J Cardiol
39: 84, 1977
8. Pierpont GL, Cohn JN, Franciosa JA: Combined oral hydral-
azine-nitrate therapy in left ventricular failure. Chest 73: 8,
1978
9. Chatterjee K, Parmley WW, Massie B, Greenberg B, Werner JI
Klausner S, Normarn A: Oral hydralazine therapy for chronic
refractory heart failure. Circulation 54: 879, 1976
10. Franciosa JA, Pierpont G, Cohn JN: Hemodynamic improve-
ment after oral hydralazine in left ventricular failure. A com-
parison with nitroprusside infusion in 16 patients. Ann Intern
Med 86: 388, 1977
11. Ports T, Chatterjee K, Parmley W, Norman A: Comparative
hemodynamics of oral prazosin and hydralazine in chronic
heart failure (abstr) Am J Cardiol 41: 367, 1978
12. Mehta J, lacona M, Pepine CJ, Conti CR: Comparative
hemodynamic effects of nitroprusside, prazosin and hydrala-
zine in refractory heart failure. (abstr) Am J Cardiol 41: 418,
1978
13. Awan NA, Miller RR, Maxwell K, Mason DT: Effects of
prazosin on forearm resistance and capacitance vessels. Clin
Pharmacol Ther 22: 79, 1977
14. Miller RR, Awan NA, Maxwell K, Mason DT: Sustained
reduction of cardiac impedance and preload in congestive heart
failure with the antihypertensive vasodilator prazosin. N Engl J
Med 297: 303, 1977
l5. Awan NA, Miller RR, DeMaria AN, Maxwell K, Neumann A,
Mason DT: Efficacy of ambulatory systemic vasodilator
therapy with oral prazosin in chronic refractory heart failure.
Concomitant relief of pulmonary congestion and elevation of
pump output demonstrated by improvements in sympto-
matology, exercise tolerance, hemodynamics and echocardiog-
raphy. Circulation 56: 346, 1977
16. Aronow WS, Greenfield RS, Alimadadian H, Danahy DT:
EfTect of the vasodilator trimazosin versus placebo on exercise
performance in chronic left ventricular failure. Am J Cardiol
40: 789, 1977
350 CIRCULATION VOL 59, No 2, FEBRUARY 1979

17. Franciosa JA, Cohn JN: Hemodynamic effects of trimazosin in
patients with left ventricular failure. Clin Pharmacol Ther 23:
11, 1978
18. Feigenbaum H, Popp RL, Wolfe SB, Troy BL, Pombo JF,
Haine CL, Dodge HT: Ultrasound measurements of left ven-
tricle: a correlative study with angiocardiography. Arch Intern
Med 129: 461, 1972
19. Popp RL, Harrison DC: Ultrasonic cardiac echography for
determining stroke volume and valvular regurgitation. Circula-
tion 41: 493, 1970
20. DeMaria AN, Vismara LA, Auditore K, Amsterdam EA, Zelis
R, Mason DT: Effects of nitroglycerin on left ventricular
cavitary size and cardiac performance determined by ultra-
sound in man. Am J Med 57: 754, 1974
21. Prakash R, Aronow WS, Warren M, Laverty W, Gottschalk
LA: Effects of marihuana and placebo marihuana smoking on
hemodynamics in coronary disease. Clin Pharmacol Ther 18:
90, 1975
22. Cooper R, O'Rourke JS, Karliner RA, Peterson KL, Leopold
GR: Comparison of ultrasound and cineangiographic measure-
ments of the rate of circumferential fiber shortening in man.
Circulation 46: 914, 1972
23. Braunwald E: Vasodilator therapy - a physiologic approach to
the treatment of heart failure. N Engl J Med 297: 331, 1977

Deep Venous Thrombosis of the Upper Extremity:

A Reappraisal
STEPHEN M. PRESCOTT, M.D. AND GERASIM TIKOFF, M.D.

SUMMARY Deep venous thrombosis (DVT) of the upper extremity is an unusual thrombotic event (1-2%
of all DVT) which can be conveniently divided into two categories, traumatic (including "stress") and spon-
taneous. The spontaneous form is not reported as often in the literature, but occurs more commonly than the
traumatic form. There is an increased left-sided predominance in spontaneous DVT compared with the
traumatic form, where a right-sided predominance exists. Possible anatomical and physiological explanations
are offered for the left-sided predominance in spontaneous DVT of the upper extremity. The thrombogenesis of
DVT of the upper extremity is compared with DVT of the lower extremity. An analysis of responses to
therapy and considerations for other therapeutic approaches are offered.

DEEP VENOUS THROMBOSIS (DVT) of the
lower extremity is widely recognized as a leading
cause of morbidity and mortality in adult patients.' 2
DVT of the upper extremity is much less common, oc-
curring with an incidence estimated at 1-2% of that of
DVT of the lower extremity.3' I We were interested in
this disparity and the possibility that study of DVT of
the upper extremity might provide some insights into
the broader problem of venous thromboembolism. We
were struck by significant differences between our
series of patients and patients reported previously,5-7
and therefore reviewed our experience with DVT of
the upper extremity and the available literature. Our
analysis suggests that:
1) There is no agreement regarding the best ter-
minology for the subgroups of this disorder. We use a
scheme (table 1) similar to that of Coon and Willis,4
which divides thrombosis into traumatic and spon-
taneous (nontraumatic). Stress, or effort, thromboses
are a subset of the traumatic category,4 as their
pathogenesis and clinical importance seem closely
similar to those of a thrombosis due to clavicular frac-
From the Medical Service, Salt Lake Veterans Administration
Hospital and the Department of Internal Medicine, University of
Utah College of Medicine.
Address for reprints: Stephen M. Prescott, M.D., Medical Ser-
vice (111), V.A. Hospital, 500 Foothill Drive, Salt Lake City, Utah,
84148.
Received April 5, 1978; revision accepted September 26, 1978.
Circulation 59, No. 2, 1979.
ture. In contrast, we feel that other "secondary"
forms of thromboses (i.e., congestive heart failure,
malignancy) have different characteristics (and
perhaps pathogenesis) from those resulting from
clavicular fracture or "stress."
2) The relative incidence of the two groups is
different from that frequently suggested by the
literature, as selection bias in favor of stress throm-
boses exists in a number of review articles. Preferen-
tial lateralization occurs toward the left side in the
spontaneous thromboses in contrast to the usual right-
sided predominance in "stress" thromboses. Plausible
anatomical and physiological reasons explain this
preferential localization.
3) Differences and similarities exist between throm-
bogenesis in the deep veins of the lower and upper ex-
tremities.
4) Prompt administration of anticoagulants is the
preferred therapy, and anatomical localization of
DVT of the upper extremity may be useful in predict-
ing the late outcome of therapy.
Materials and Methods
We studied 12 patients with 14 clinical episodes of
DVT of the upper extremity. Most of them were
referred to us by the physicians at hospitals affiliated
with the University of Utah, but were otherwise un-
selected. Bias can be inherent in any referral popula-
tion, but we feel that our patients are at least as
representative of the disorder as patients reported

Downloaded from http://circ.ahajournals.org/ by guest on April 7, 2016

 Effect of prazosin vs placebo on chronic left ventricular heart failure.
W S Aronow, M Lurie, M Turbow, K Whittaker, S Van Camp and D Hughes

Circulation. 1979;59:344-350
doi: 10.1161/01.CIR.59.2.344
Circulation is published by the American Heart Association, 7272 Greenville Avenue, Dallas, TX 75231
Copyright 1979 American Heart Association, Inc. All rights reserved.
Print ISSN: 0009-7322. Online ISSN: 1524-4539

The online version of this article, along with updated information and services, is located on
the World Wide Web at:
http://circ.ahajournals.org/content/59/2/344

Permissions: Requests for permissions to reproduce figures, tables, or portions of articles originally
published in Circulation can be obtained via RightsLink, a service of the Copyright Clearance Center, not the
Editorial Office. Once the online version of the published article for which permission is being requested is
located, click Request Permissions in the middle column of the Web page under Services. Further
information about this process is available in the Permissions and Rights Question and Answer document.

Reprints: Information about reprints can be found online at:

http://www.lww.com/reprints

Subscriptions: Information about subscribing to Circulation is online at:

http://circ.ahajournals.org//subscriptions/

Downloaded from http://circ.ahajournals.org/ by guest on April 7, 2016