You are on page 1of 4

Effect of Dobutamine on Plasma Potassium

in Congestive Heart Failure Secondary to
Idiopathic or Ischemic Cardiomyopathy
Irvin F. Goldenberg, MD, Maria Teresa Olivari, MD, T. Barry Levine, MD, and Jay N. Cohn, MD
Dobutamine was administered in a dose of 10 f 1
pg/kg/min to 13 patients with severe idiopathic or
ischemic dilated cardiomyopathy. Acute hemody-
namic improvement was noted in all patients. All
patients had a significant decrease in plasma po-
tassium (4.6 f 0.1 to 4.2 f 0.2 mEq/liter, p
<0.0001) at peak infusion. The decrease in potas-
sium persisted for at least 45 minutes after discon-
tinuing the infusion. Three patients had exacerba-
tion of baseline ventricular arrhythmias that re-
solved with infusion discontinuation. Changes in
plasma norepinephrine could not explain the potas-
sium decrease or arrhythmia production, which
also significantly decreased in these patients (771
f 123 to 524 f 73 pg/ml, p <O.Ol). It is concluded
that dobutamine causes a significant decrease in
plasma potassium and that the decrease persists at
least 45 minutes after the infusion is discontinued.
(Am J Cardiol 1989;63:643-646)
From the Minneapolis Heart Institute and the University of Minnesota,
Minneapolis, Minnesota. Manuscript received September 26, 1988;
revised manuscript received and accepted January 3, 1989.
Address for reprints: Irvin F. Goldenberg, MD, Minneapolis Heart
Institute Foundation, 920 East 28th Street, Suite 160, Minneapolis,
Minnesota 55407.
obutamine is a potent, intravenously administered
inotropic agent used for temporary circulatory
support in patients with severe pump failure.’
Recent reports also have advocated its intermittent use
on a long-term basis for treatment of refractory conges-
tive heart failure (CHF).2,3 Dobutamine exerts signifi-
cant /3r stimulatory effects, with only modest stimula-
tion of & receptors.1,4 Other /3 agonists (such as epi-
nephrine, salbutamol) by a postulated ,&-receptor
mechanism have been shown to cause a significant de-
crease in plasma potassium in normal subjects.5-8 These
changes have not been reported in patients with CHF.
However, patients with CHF have been shown to have
down-regulation of ,& and /32 receptors in some or-
ganq9Jo and thus may not have clinically significant po-
tassium decreases when given drugs with only modest 162
effects. Ventricular arrhythmias are often present in pa-
tients with CHF and probably contribute to the high
incidence of sudden death in this syndrome.r1J2 Beta
agonists like dobutamine have been reported to worsen
or precipitate these ventricular arrhythmias. Because
potassium shifts could potentially contribute to arrhyth-
mia production, the present study was designed to de-
termine whether dobutamine, a predominant fir agonist,
could produce significant plasma potassium changes in
patients with CHF.
Thirteen patients with chronic dilated cardiomyopa-
thy were studied before, during and after dobutamine
infusion. Five patients were treated because of acute ex-
acerbations of their chronic CHF and 8 were in chronic,
stable CHF. Of the 13 patients, 6 had idiopathic and 7
ischemic cardiomyopathy. There were 9 men and 4
women, and the mean age was 56 f 5 years. Six were in
New York Heart Association class IV and 7 in class III.
The average left ventricular ejection fraction was 15 f
3%. All patients were taking digoxin and received their
dose on the day of the study. No patient received diu-
retics within 12 hours of the study. All patients under-
went right-sided heart catheterization and arterial line
placement. Measurements were made of mean arterial
pressure, pulmonary artery wedge pressure, right atria1
pressure and cardiac output (thermodilution). Cardiac
index, stroke index and systemic vascular resistance [80
(mean arterial pressure - right atria1 pressure)/cardiac
output] were calculated. Heart rate was monitored con-
tinuously from an electrocardiographic lead. Baseline
hemodynamic measurements were taken every 1.5 min-
utes until 2 determinations varied by <lo%.
Dobutamine was titrated by starting the infusion at
5.0 pg/kg/min and increasing the infusion rate every 15
minutes (by 2.5 to 5.0 pg/kg/min) until a dose of 15
pg/kg/min was reached or the patient’s heart rate in-
creased by 15 beats/min, the mean arterial pressure in-
creased by 20% or increasing ventricular ectopic activity
was noted. The peak dobutamine infusion rate was
maintained for 15 minutes and then the hemodynamic
measurements were repeated. To avoid any potential ef-
fects of dextrose administration on plasma potassium
levels, all dobutamine infusions were prepared with nor-
mal saline. In all patients the cardiac rhythm was re-
corded op continuous electrocardiographic strips for 60
minutes before the infusion and during the entire infu-
sion period. The continuous electrocardiographic strips
were analyzed and the number of ventricular premature
complexes, couplets and ventricular tachycardia epi-
sodes every minute were determined. To classify the ar-
rhythmia as worsening during the infusion, the patient
had to exhibit a mean increase of at least 5 ventricular
premature complexes/min, or the dt+elopment of ven-
tricular couplets or ventricular tachycardia. The in-
crease in ventricular premature complexes had to be
present for at least 15 minutes.
Central venous blood was obtained for measurement
of plasma potassium at baseline (immediately before
starting infusion), 15 minutes after peak infusion began
and, in the last 7 patients, 45 minutes after the infusion
was discontinued. In 9 patients, venous blood was ob-
3.0Pre Post
flGURE 1. Plasma pdasdm levels before dohtamine infu-
sidn, at peak infusion and 45 minutes after infusion. Thick
so&f line cenneets means f standard enws of all 13 patii
(bmeline to peak infusion). Dashed lime commts means f
standard errers (basehe, peak, postinfusion) in 7 patients
w plasma potatim at all 3 time points. * p <O.OWl peak
basellne in 7 patients.
TABLE I Changes in Systemic Hemodynamics in Thirteen
Patients with Severe Congestive Heart Failure
Baseline Peak p Value
Heart rate (beats/min) 88f6 95f6 0.06
Mean blood pressure (mm Hg) 76f3 76f3 NS
Mean right atrial pressure lOf2 6f 2 <0.05
(mm W
Pulmonary artery wedge pressure 24f3 18f2 0.08
(mm W
Cardiac index (liters/min/m2) 2.1 fO.l 3.3 f0.2 <O.col
Stroke volume index (ml/m*) 25f3 35f3 <O.OOl
Systemic vascular resistance 1,429 f 113 955 f 76 <O.OOl
NS = not s~@ficant.
tained for the measurement of plasma renin activity by
the radioimmunoassay method of Sealey et all3 and
plasma norepinephrine levels by the radioenzymatic as-
say of Passon and Peuler,14 using a CAT-A-KIT device
(Upjohn Company). Using this method, duplicate mea-
surements taken in our laboratory have a coefficient of
variation of 8.4% for norepinephrine.
Statistical analysis was performed using paired t
tests to compare hemodynamic measurements and plas-
ma potassium at the different times stated. Data are
expressed as mean values f standard error.
Dobutamine was titrated to a dose of 10 f 1 pg/kg/
min. The mean duration of infusion before peak hemo-
dynamics and potassium measurements was 60 minutes.
The most common reason for not reaching the target
dose of 15 pg/kg/ min was a heart rate increase >15
beats/min. Hemodynamic changes induced by peak
dobutamine infusion are listed in Table I. Plasma potas-
sium decreased significantly in all patients at peak infu-
sion, from 4.6 f 0.1 to 4.2 f 0.2 mEq/liter, p <O.OOOl
(Figure 1). In 7 patients, plasma potassium was deter-
mined again 45 minutes after the infusion was discon-
tinued. In these 7, the plasma potassium 45 minutes af-
ter discontinuing the infusion was still at a significantly
lower level than at baseline 4.3 f 0.1 vs 4.5 f 0.1 mEq/
liter, p <0.02 (Figure 1). There was no significant cor-
relation between degree of hemodynamic change and
plasma potassium changes. Plasma norepinephrine de-
creased significantly at peak infusion (771 f 123 to 524
f 73 pg/ml, p <O.Ol) (Figure 2). There was no signifi-
cant change in plasma ienin activity (35 f 10 to 32 f 9
ng/ml/hr, difference not significant).
Four of 13 patients had ventricular arrhythmias at
baseline (i.e., > 1 ventricular premature complex/min or
couplets). Three of these patients had ventricular pre-
mature complexes and 1 had couplets. No patient had
ventricular tachycardia at baseline. The effects of dobu-
tamine on the number of ventricular premature com-
plexes in all 13 patients is shown in Figure 3. Three
patients developed worsening of arrhythmias. One had
an episode of nonsustained ventricular tachycardia and
an increase in the frequency of ventricular premature
complexes, 1 with no baseline arrhythmias developed
frequent couplets and 1 had ventricular tachycardia re-
quiring cardioversion. The latter had a history of recent
cardiac arrest and myocardial infarction and was being
treated with dobutamine because of cardiogenic shock.
Before starting dobutamine, this patient had not had
any significant arrhythmias for 12 hours. Because dobu-
tamine had to be discontinued, an intraaortic balloon
pump was placed to maintain adequate perfusion. The
other 10 patients had no notable change in the frequen-
cy of their arrhythmias. The plasma potassium decrease
in the 3 patients who developed worse arrhythmias
(mean decrease = 0.4 mE!q/liter) was similar to the de-
crease found in the patients without arrhythmias, and it
occurred at a similar dobutamine dose (10 hg/kg/min).
Patients with idiopathic and ischemic dilated cardio-
myopathy are at increased risk for ventricular arrhyth-
mias and sudden death.11J2 In addition to the underly-
ing etiology of their disease (such as ischemia) predis-
posing to arrhythmias, these patients are often given
drugs (e.g., diuretics, digitalis) that facilitate arrhyth-
mia production. l 5
When patients with CHF are treated for severe exa-
cerbations, they are often given additional diuretics and
sometimes an infusion of dobutamine.‘J5 Because cer-
tain patients develop significant ventricular arrhythmias
with dobutamine use, the dose must sometimes be de-
creased or the infusion stopped. Occasionally, because
the infusion must be discontinued, more aggressive
measures are performed, such as intraaortic balloon
placement, as was done in 1 of our patients.
Whereas ,6 agonists may precipitate ventricular irri-
tability by multiple mechanisms, including ischemia, in-
creased automaticity and decreased ventricular fibrilla-
tion threshold, their effect on altering potassium fluxes
has only recently been stressed.16J7 Our study deter-
mines if clinically significant decreases in plasma potas-
sium occur when dobutamine is given to patients with
CHF. We found that patients given dobutamine in dos-
ages producing the expected hemodynamic changes ex-
hibited a significant acute decrease in plasma potassi-
um. This decrease was present shortly after initiation of
the infusion and persisted for at least 45 minutes after
the infusion was discontinued. The decrease was associ-
ated with the development or worsening of ventricular
arrhythmias in 3 patients. A cause and effect relation
FIGURE 2. A, plasma norepinephrine levels at baseline and
peak dobutamine infusion (* p <O.Ol). B, plasma renin
activity at basehe and peak dobutamine infusion.
between the decreased potassium and arrhythmias,
however, was not established as these patients were not
retreated with dobutamine after replacement of potassi-
um. In addition, our criteria for worsening of ventricu-
lar arrhythmias may not have been strict enough, nor
our sampling time long enough to adequately assess the
true incidence of worsening of ventricular arrhythmias.
Although p agonist-induced hypokalemia is not a newly
discovered phenomenon in normal subjects, this study
does show that dobutamine, a relatively Pi-selective
drug with modest p2 effects, can produce significant
changes in plasma potassium in patients with heart fail-
ure at clinically used dosages.
Some p agonists have been reported to increase plas-
ma norepinephrine during their infusion.* This increase
in plasma norepinephrine could potentially be responsi-
ble for some of the arrhythmias produced by these
agents. In addition, norepinephrine has weak ,&-stimu-
lating effects’* and can produce potassium decreases.6J9
In the present series, however, plasma norepinephrine
decreased significantly during infusion of dobutamine.
Not all p agonists have been reported to cause the
same degree of plasma potassium decrease.7,8 Brown et
al7 gave both epinephrine and isoproterenol to normal
subjects. They found no change in plasma potassium
with isoproterenol, but a significant decrease in plasma
potassium with epinephrine despite similar increases in
heart rate (25 beats/mm) and pulse pressure with the 2
drugs. These investigators attributed the difference in
response of these 2 catecholamines to the relative p2 se-
lectivity of epinephrine. Other investigators8 have also
shown differences between p agonists in their ability to
cause hypokalemia, with /I2 agonists causing a greater
decrease in plasma potassium than predominantly /3r
agonists. Because the potassium decrease caused by
these agents can be prevented by nonselective p antago-
nism (such as propranolol) but not by selective pr an-
tagonism (such as atenolol), a ,&-receptor mechanism
for the decreased potassium has been suggested.* This
lo T
Baseline Peak
FIGURE 3. Effect of dobutamine on the number of ventricular
premature complexes.
&-receptor stimulation causes an increase in the sodi-
urn-potassium-adenosine triphosphatase activity in skel-
etal muscle with subsequent potassium migration into
the ~~11s.~ The decrease in potassium appears to be
related directly to &receptor stimulation and not to
secondary changes in other hormone systems (e.g., re-
nin, insulin, aldosterone) that regulate potassium
changes.7J6 No change in plasma renin activity was not-
ed during infusion in the present study. In addition, uri-
nary loss of potassium is not a likely mechanism for this
potassium decrease,16v20 but prior use of diuretics ap-
pears to further increase the risk of catecholamine-in-
duced hypokalemia.21
Because patients with dilated cardiomyopathy are
also often treated with diuretics and digoxin, they may
be at considerable risk for arrhythmias if hypokalemia
develops. Because an inverse relation between plasma
potassium levels and malignant ventricular arrhythmias
has been demonstrated in patients with cardiac dis-
ease,22-24 it would seem appropriate to monitor potassi-
um levels closely in patients treated with dobutamine
and to administer supplemental potassium when neces-
Although intermittent dobutamine infusions have
been used to treat patients with chronic congestive car-
diomyopathy,2T3 the safety of this treatment has not
been established. Dobutamine-induced hypokalemia
could be harmful, leading to a worsening of ventricular
arrhythmias and sudden death. A recently performed
multicenter double-blind study comparing intermittent
dobutamine infusions with placebo revealed a disturbing
incidence of death during dobutamine infusions, partic-
ularly if there were high grade ventricular arrhythmias
on baseline Holter recordings. The relation of these
findings to plasma potassium levels is not known,
Acknowledgment: The authors would like to thank
Terri Hanson and Nancy Grunz for secretarial assis-
tance in preparing this manuscript.
1. Leier CV, Unverferth DV. Drugs live years later: dobutamine. Ann Intern Med
2. Roffman DS, Applefeld MM, Grove WR, Talesnick BS, Sutton FJ, Newman
KA, Reed WP. Intermittent dobutamine hydrochloride infusions in outpatients
with chronic congestive heart failure. Chin Pharm 1985;4:195-199.
3. Berger M, McSherry CK. Outpatient dobutamine infusion using a totally
implantable infusion pump for refractory congestive heart failure. Chest 1985;
4. Williams RS, Bishop T. Sele&ivity of dobutamine for adrenergic receptor
subtypes. In vitro analysis by radioligand binding. J Clin Znvest 1981,67:1703-
5. Clausen T, Flatman JA. &-adrenoceptors mediate the stimulating effect of
adrenaline on active electrogenic Na-K transport in rat soleus muscle. Br J
6. Clausen T, Flatman JA. The effect of catecliolamines on Na-K transport and
membrane potential in rat soleus muscle. J Physiol 1977;270:383-414.
7. Brown MJ, Brown DC, Murphy MB. Hypokalemia from beta-2 receptor
stimulation by circulating epinephrine. N Engl J Med 1983;309:1414-1419.
8. Vincent HH, Boomsma F, Man in’t Veld AJ, Derkx FHM, Wenting GJ,
Schalekamp MADH. Effect of selective and nonselective B-agonist on plasma
potassium and norepinephrine. J Cardiowx Pharmacol 1984,6:107-l 14.
9. Bristow MR, Ginsburg R, Minobe W, Cubicciotti RS, Sageman WS, Lurie K,
Billingham ME, Harrison DC, Stinson EB. Decreased catecholamine sensitivity
and @-adrenergic-receptor density in failing human hearts. N EngZ J Med
1982;307:205-21 I.
10. Colucci WS, Alexander RW, Williams GH, Rude RE, Holman BL, Konstam
MA, Wynne J, Mudge GH Jr, Braunwald E. Decreased lymphocyte beta-adre-
nergic receptor density in patients with heart failure and tolerance to the beta-
adrenergic agonist pirbuterol. N Engl J Med 1981;305:185-1%
11. Smith WM. Epidemiology of congestive heart failure. Am J Cordial
12. Cohn JN, Levine TB, Olivari MT, Garberg V, Lura B, Francis GS, Simon
AB, Rector T. Plasma norepinephrine as a guide to prognosis in patients with
chronic congestive heart failure. N Engl J Med 1984;311:819-823.
13. Sealey JE, Gerten-Banes J, Laragh JH. The renin system: variations in man
measured by radioimmunoassay or bioassay. Kidney Int 1972;1:240-253.
14. Passon PG, Peuler JD. A simplified radiometric assay for norepinephrine and
epinephrine. Anal Biochem 1973;51:618-631.
15. Smith TW, Braunwald E. The management of heart failure. Id: Braunwald
E, ed. Heart Disease: A Textbook of Cardiovascular Medicine. Second cd. Phila-
delphia WB Saunders, 1984:503-559.
16. Rosa RM, Silva P, Young JB, Landsberg L, Brown RS, Rowe JW, Epstein
FH. Adrenergic modulation of extrarenal Potassium disposal. N Engl J Med
17. Laulei DP. A symposium: potassium, catecholamines and beta-blockade.
Introduction. Am J Cardiol 1985:56:1&2D.
18. Lands AM, Arnold A, McAuliff JP, LuDuena FP, Brown Jun TG. Differen-
tiation of receptor systems activated by sympathomimetic amines. Nature
19. Todd EP, Vick RL. Kalemotropic effect of epinephrine: analysis with adre-
nergic agonist and antagonist. Am J Physioi 1971;220:1964-1969.
20. Smith SK, Thompson D. The effect of intravenous salbutamdl upon plasma
and urinary potassium during premature labour. Br J Obstet GynaecolZ977;84:
21. Struthers AD, Whitesmith R, Reid JL. Prior thiazide diuretic treatment
increases adrenaline-induced hypokalemia. Lmcet 1983;1:1358-1361.
22. Hulting J. In-hospital ventricular fibrillation and its relation to serum potassi-
um. Acta Med Stand 1981,647:109-116.
23. Nordrehaug JE. Malignant arrhythmias in relation to serum potassium in
acute myocardial infarction. Am J Cordtol 1985;56;2OD-23D.
24. Solomon RJ, Cole AG. Importance of potassium in patients with acute
myocardial infarction. Acta Med Scand l981,647(suppl 647):87-93.