Drug interactions with amiodarone

There are a number of important drug interactions with amiodarone. This agent appears to have
a marked effect on the kinetics of some commonly used cardiovascular drugs, such as warfarin,
digoxin, quinidine, and procainamide, and has dynamic interactions with others, such as the beta
blockers and some calcium antagonists. Bleeding has been reported, apparently caused by a
potentiation of the anticoagulant effect of warfarin by amiodarone. Tofsades de pointes has
been observed when quinidine, propafenone, or mexiletine is given together with amiodarone.
Furthermore, amiodarone may interact with beta-blocking agents and some of the calcium
antagonists to produce symptomatic sinus bradycardia and sinus arrest, especially in a latent or
overt sick sinus syndrome. During surgery, amiodarone may induce hypotension and an
atropine-resistant bradycardia, possibly by interacting with anesthetic agents. A knowledge of
the time of onset, extent, duration, and possible mechanisms of the interactions of amiodarone
with other cardioactive drugs is still incomplete, but further studies are of great therapeutic
importance. (AM HEART J 106:924, 1983.)

Frank I. Marcus, M.D. Tucson, Ark

Amiodarone, like propranolol, has an extremely Table I. Clinical consequencesof drug interactions with
wide therapeutic/toxic ratio during short-term amiodarone
administration. Drug interactions with amiodarone Concomitant drug Toxic interaction
are of consequence, however, because this agent
appears to have a marked effect on the kinetics of Warfarin Bleeding
several important cardiovascular drugs, such as war- Digoxin Sinoatrial and AV node
depression; gastrointestinal
farin, digoxin, quinidine, and procainamide, and has and central nervous system
hemodynamic and electrophysiologic interactions toxicity
with others, such as the beta-blocking drugs and Quinidine Torsades de pointes
calcium antagonists (Table I). This paper will focus Propafenone
on the onset, extent, duration, and mechanism of the Disopyramide
Mexiletine
known interactions. Aprindine Increased side effects; possibly
AMIODARONE AND WARFARIN INTERACTIONS Procainamide cardiovascular effects
Beta blockers Symptomatic bradycardia;
That amiodarone potentiates the anticoagulant sinus arrest
effect of warfarin was first reported in 1979 by Calcium antagonists Sinus arrest; AV blockade
Verapamil
Simpson,’ who surveyed the British experience with Diltiazem
275 patients treated with amiodarone and found an Anesthetic drugs Hypotension; bradycardia
enhanced anticoagulant effect in three patients.
Subsequently other investigators reported seven
patients with bleeding caused by amiodarone poten- starting amiodarone5 but may be delayed for as long
tiation of the anticoagulant effect of warfarin.2-4 as 3 weeks.3 After discontinuation of amiodarone,
Prothrombin time may double after amiodarone the potentiating effect on the activity of warfarin
therapy in patients taking warfarin (Fig. 1). This persists for weeks3v4 or months.6
effect has been observed as early as 3 or 4 days after The mechanism of the potentiation of anticoagu-
lant effects of warfarin has not been established.
Some data suggest that the mechanism of interac-
From the Department of Medicine, Cardiology Section, University of tion may be interference of warfarin metabolism by
Arizona Health Sciences Center.
amiodarone. In one patient evaluated 4 days after
Reprint requests: Frank I. Marcus, M.D., Cardiology Section, University of
Arizona Health Sciences Center, 1501 North Campbell Ave., Tucson, AZ amiodarone was started, plasma concentration of
85724. warfarin increased by 100% .5 Plasma protein

924
Volume 106
Number 4, Part 2 Drug interactions with amiodorone 925

;
:
4- l

p<o.o01 p<o.oor

before during
aflliderala tnetmnt
Fig. 1. Prothrombin time in nine patients treated with
amiodarone. (From Martinowitz U, Robinovici J, Gold-
farb D, Many A, Banks H: Interaction between warfarin
sodium and amiodarone. N Engl J Med 304:671, 1981.)
binding of warfarin, measured by equilibrium analy-
sis, was not altered, however, and peripheral antifac-
tor activity was not found.3
The effect of amiodarone on prothrombin activity
appears to be dose independent. Although increases
in prothrombin time vary, it is recommended that
the maintenance dose of warfarin be reduced by one
third to one half when amiodarone is begun (Fig. 2;
Table 11)3,6 and that careful surveillance be main-
tained. Amiodarone may also influence heparin
activity.7
AMIODARONE-DIGOXIN INTERACTION
In 1981 Moysey et al.* observed that when amio-
darone was added to maintenance digoxin therapy
in seven patients, all showed progressive increases in
serum digoxin concentration; the average increase
was 69% (Fig. 3). These observations were subse-
quently confirmed by mostg-12 but not aIlL investiga-
tors. Plasma digoxin increases were evident within
24 hours after amiodarone, 200 mg three times a day,
anioda~a\e treatment
Fig. 2. Weekly dose of warfarin sodium (Coumadin) in
nine patients treated with amiodarone. Dose of warfarin
sodium was reduced by 16 % to 45% after amiodarone
treatment to maintain prothrombin time within therapeu-
tic range. (From Martinowitz U, Robinovici J, Goldfarb D,
Many A, Bank H: Interaction between warfarin sodium
and amiodarone. N Engl J Med 304:671, 1981.)
and rose linearly for 6 or 7 days; then the digoxin
level appeared to reach a plateau.s Of these seven
patients, four developed symptoms compatible with
digoxin toxicity. The toxic effects of digitalis include
sinus arrest, gastrointestinal symptoms, and “neuro-
logic toxicity.““, l2 The fact that none of these
patients developed atrial or ventricular ectopy may
be attributed to the antiarrhythmic effects of amio-
darone. The magnitude of this interaction appears
to be dose related8 and is correlated with the plasma
concentration of amiodaroneg (Fig. 4). When amio-
darone is withdrawn, plasma digoxin concentration
decreases progressively over 2 weeks.
The mechanism of this interaction has not been
described. The time course, with interaction begin-
ning within 24 hours and attainment of steady-state
digoxin level by 7 days (or five half-lives of digoxin),
resembles the digoxin-quinidine interaction. This
suggests that one mechanism of this interaction may
 October, 1983
926 Marcus American Heart Journal

+75
digcain o/--o
.I..----- digain + 2
amiodarone ,,*
1 -0t
+%I-

Digoxin
concentration + 25 -
tv. char1gc)

I
0.5 1 1.5 2
AMlODAROtVE t?ICg/t?I
- 25 J
1 2 3 f. 5 6 7 Fig. 4. Plot of plasma amiodarone and plasma digoxin
Days concentrations in 33 patients treated for an average of 52
months. There was a correlation between these two
Fig. Mean percentage change in daily plasma digoxin
3. parameters (r = 0.66). Dose of digoxin was 0.25 mg 0);
concentrations in seven patients treated with digoxin. that of P-methyl digoxin was 0.2 mg (0). Mean daily dose
Amiodarone, 600 mg daily, was added from day 3 (o ). Six of amiodaronewas 154 t- 24 mg in 23 patients treated for
controls were treated with digoxin alone (x). (From Moy- paraxysmal supraventricular tachycardia, and it was
sey JO, Jaggarao NSV, Grundy EN, Chamberlain DA: 238 + 73 mg in 10 patients treated for ventricular ar-
Amiodarone increases plasma digoxin concentrations. Br rhythmias. (From Furlanello F, Inama G, Ferrari M,
Med J 282:272, 1981.) Padrini R, Piovan D, Guarnerio M, Vergara G, Del
Fauvero A, Dal Farno P, Disertori M: Amiodarone and
amiodarone plus digitalis in the treatment of paroxysmal
be a decrease in elimination. Digoxin is excreted supraventricular reciprocating tachyarrhythmias. Phar-
macol Res Commun l&731,1982.)
primarily by the kidneys, by both glomerular filtra-
tion and tubular secretion. Since the glomerular
filtration rate is not decreased by amiodarone, it
may be that amiodarone reduces the elimination of antidysrhythmic drugs, such as quinidine, disopy-
digoxin by renal tubular secretion. Other mecha- ramide, mexiletine, propafenone, aprindine, and
nisms that may contribute to the increase in serum procainamide. Drug-induced torsades de pointes
digoxin levels include a decrease in extrarenal excre- has been reported for combinations of amiodarone
tion and tissue displacement of digoxin by amioda- and quinidine, disopyramide, propafenone, or mexi-
rone. These hypotheses await experimental proof. letine.l* In each case the QT interval was markedly
Our experience with 40 patients treated with prolonged prior to the onset of the torsades de
amiodarone and digoxin is consistent with that pointes (Fig. 5). When amiodarone, 200 mg three
previously reported. Serum digoxin levels usually times a day, was given to a normal volunteer taking
doubled when amiodarone was given. Two patients maintenance doses of quinidine, there was an
developed symptoms of digitalis toxicity associated increase in the plasma quinidine level, as well as an
with serum digoxin levels greater than 2 rig/ml; one additive effect on prolongation of the QT intervalI
had nausea, the other visual disturbances. Digitalis (Fig. 6). This enhanced effect was observed after 2
toxic arrhythmias were not observed. Based on our days of combination therapy. Other investigators
experience and that of others, we recommend that have also documented increased serum concentra-
the dose of digoxin be decreased by half when the tions of quinidine and reduced dosage requirements
two drugs are given together (Table II). with this combination of drugs’” (Table III).
Thus it appears that the marked QT prolongation
AYIODARONE INTERACTION WITH ANTIDYSRHYTHMIC
associated with quinidine and amiodarone therapy
DRUGS
is due in part to a pharmacokinetic interaction and
Some of the most hazardous drug interactions that some patients given these drugs can be expect-
with amiodarone are those which occur with other ed to develop torsades de pointes. Therefore, in
Volume la6
Number 4, Part 2 Drug interactions with amiodarone 927

Table II. Pharmacokinetic drug interactions with amiodarone

Interaction
Recommended
Concomitant drug Onset (days) Magnitude Duration dose correction

Warfarin 3 to 4 Increases prothrombin time by 2wkto4mo 1 l/3

100%
Digoxin 1 Increases serum concentration 2 wk 1 112
by 70%
Quinidine 2 Increases serum concentration ? 1 l/3 (or avoid)
by 33%
Procainamide <7 Increases plasma concentration ? 1 l/3 (or avoid)
by 55 % ; NAPA concentration
by 33%
NAPA = N-acetylprocainamide.

Table III. Serum concentrations and daily requirements of quinidine and procainamide before and during
amiodarone, 800 to 1200 mg/day, for less than a weekI
Mean +_ SD

Antiarrhythmic agent Before A During A % Change p Value*

Serum concentration (pg/ml)

Quinidinet 4.4 +- 1.1 5.8 r 0.9 t 33 <O.OOl
Procainamidel 6.8 i 1.8 10.6 + 4.8 t 55 <O.Ol
N-acetyl procainamide 6.9 + 4.8 9.1 + 4.7 t 33 <0.05
Procainamide plus N-acetyl 13.7 + 6.2 19.7 * 8.7 T 44 <O.Ol
procainamide
Daily dose (mg/day)
Quinidine 2960 zt 1115 1875 ? 794 133 <0.005
Procainamide 3708 r 1391 2958 f 1501 120 0.01
A = amiodarone.
*Paired t test.
tll patients.
t12 patients.

general, class I drugs should either be avoided with receptors.18 Amiodarone decreases sinus node auto-
amiodarone or used with caution. maticity and prolongs the refractory period of the
It is not known whether serum concentrations of atrioventricular (AV) node. In patients with sinus
propafenone or mexiletine are increased when amio- bradycardia, sick sinus syndrome, or partial AV
darone is given concomitantly, but increases in block, either amiodarone, a beta-blocking drug, or
aprindine17 and procainamide 1evehP (Table III) certain calcium antagonists (e.g., verapamil and
have been found when these drugs were given in diltiazem) can further slow the sinus rate or worsen
conjunction with amiodarone. It would seem reason- AV block. The combination can have additive
able to decrease by half the dose of any class I drug effects? occasionally with adverse consequences. For
given with amiodarone (Table II). At any rate the example, Derrida et al.lg described a patient who had
value of combining class I agents with amiodarone a rapid ventricular response to atrial flutter even
for the control of serious arrhythmias has not been with amiodarone and digitalis therapy. After one
critically evaluated. dose of propranolol given by mouth, the patient had
cardiac arrest. The ECG showed complete absence
AMIODARONE INTERACTIONS WITH
of ventricular activity. Fortunately the patient
BETA-ADRENERGIC-BLOCKING DRUGS AND CALCIUM
responded to isoproterenol infusion. Another
ANTAGONISTS patient with an acute myocardial infarction was
Another type of drug interaction is that between given amiodarone intravenously because of contin-
two classes of drugs that have similar effects. Amio- ued pain. When the pain persisted for 24 hours,
darone has sympathetic blocking activity but is not amiodarone was discontinued, and he received two
a competitive antagonist of the beta-adrenergic doses of oral propranolol. One and one-half hours
 October. 1983
92% Marcus American Heart Journal

QT
msuc

200 I

700
u, 00 00 1W hr .

Fig. 5. QT interval in six patients who developed torsades

de pointes before (0) and after (0) a class1
antiarrhythmic drug wasgiven in addition to amiodarone.QT interval wasgreater than 0.50 secondin all
six patients who developed torsades de pointes. (From Tartini R, Kappenberger L, Steinbrunn W:
Gefahrliche interaktionen Zurischen amiodaron und antiarrhythmika der klasse 1. Schweig Med
Wochenschr 1129585, 1982.)

later he developed marked bradycardia that was
soon followed by ventricular fibrillation, successful-
ly treated with electrical defibrillation.
Amiodarone should be used with caution with
beta-blocking drugs or calcium antagonists, particu-
larly if there is suspicion of underlying dysfunction
of the sinus node, such as bradycardia or sick sinus
syndrome, or if there is partial AV block. It would be
prudent to exercise the same caution with verapamil
and diltiaxem. One area that should be fruitful for
further study of drug interactions with amiodarone
is that with anesthetic agents, because of the obser-
vation that hypotension and atropine-resistant
bradycardia may occur during surgery.20
CONCLUSIONS
The large number of serious drug interactions
with amiodarone make this a potentially hazardous
drug because so many patients with cardiovascular
disease receive multiple drugs. Undoubtedly, inter-
actions of amiodarone with drugs other than those
already observed will be uncovered. Amiodarone
and its major metabolite, desethylamiodarone, con-
centrate in the liver21 and have the potential of
interfering with the hepatic metabolism of drugs. In
addition, the kinetics of the interaction with digoxin
suggest that amiodarone may interfere with renal
clearance of some drugs. Therefore we must be alert
to these many interactions in order to prevent toxic
effects of concomitantly administered cardiovascu-
lar drugs.
I thank Ms. Catherine Abe& for her expert secretarial assis-
tance and Dr. Paul Fenster for his review of the manuscript.
REFERENCES
1. Simpson WT: Amiodarone in cardiac arrhythmias. Interna-
tional Congress Series No. 16, Royal Society of Medicine.
New York, 1979, Grune & Stratton, Inc, p 50.
2. Rees A, Palal JJ, Reid PG, Henderson AH: Dangers of
amiodarone and anticoagulant treatment. Br Med d
282:1757, 1981.
3. Hamer A, Peter T, Mandel WJ, Scheinman MM, Weiss D:
The potentiation of warfarin anticoagulation by amiodarone
and anticoagulant treatment. Circulation 65:1025, 1982.
4. Broekmans AW, Meyboom RHB: Potentiering van het
cumarine-effect door amiodaron (Cordarone). Ned Tijdschr
Geneeskd 1263415, 1982.
5. Serlin MJ, Sibeon RG, Green GJ: Dangers of amiodarone and
anticoagulant treatment. Br Med J 283:57, 1981.
6. Martinowitz U. Rabinovici J. Goldfarb D. Manv A. Bank H:
Interaction between warfarin sodium and amiodarone. N
Engl J Med 304:671, 1981.
7. Tolstopyatov BI: Study of interrelations of etmozin, Corda-
rone, and phenycaberan with heparin. Farmakol Toksikol
44:586, 1981.
8. Moysey JO, Jaggarao NSV, Grundy EN, Chamberlain DA:
Volume 106
Number 4. Part 2
Fig. 6. Changesin plasmaquinidine levels, QT interval,
and heart rate after administration of amiodarone, 600
mglday, to healthy subject receiving quinidine sulfate, 1.2
gm daily for 6 days. Plasma quinidine levels increased
from baseline of 4.4 and 6.6 pmol/L to 9.5 pmol/L after
combination therapy. Quinidine alone caused a 20%
increase in QT interval, whereas combination therapy
resulted in a 40% increase(From Tartini R, Kappenberg-
er L, Steinbrunn W, Meyer UA: Dangerousinteractions
between amiodarone and quinidine. Lancet 1:1327,
1982.)
Amiodarone increases plasma digoxin concentrations. Br
Med J 282:272, 1981.
9. Furlanello F, Inama G, Ferrari M, Padrini R, Piovan D,
Guarnerio M, Vergara G, Del Fauvero A, Dal Forno P,
Disertori M: Amiodarone and amiodarone plus digitalis in the
treatment of paroxysmal supraventricular reciprocating
tachyarrhythmias. Pharmacol Res Commun 14:731, 1982.
10. Furlanello F, Inama G, Ferrari M, Padrini R, Guarnerio M,
Vergara G, Dal Forno P: Digoxin-amiodarone: A further
example of digoxin-antiarrhythmic agents interaction on
long-term treatment. G Ital Cardiol 11:1725, 1981.
11. Oetgen WJ, Sohel SM, Tri TB, Heydron WH, Davia JE,
Rakita L: Amiodarone-digoxin interaction: Clinical and
experimental observations (abstr). Circulation 66(suppl
11):382, 1982.
12. Nademanee K, Kannan R, Hendrickson JA, Burnam M, Kary
I, Singh B: Amiodarone-digoxin interaction during treatment
of resistant cardiac arrhythmias (abstr). Am J Cardiol
49:1026, 1982.
Drug interactions with amiodarone 929
13. Achilli A, Giacci M, Capezzuto A, DeLuca F, Guerra R, Serra
N: Quinidine, digoxin and amiodarone-digoxin interactions:
Effects on serum and digoxin levels. G Ital Cardiol 11:918,
1981.
14. Tartini R, Kappenberger L, Steinbrunn W: Gefahrliche
interaktionen zwischen amiodaron und antiarrhythmika der
klasse 1. Schweiz Med Wochenschr 112:1585, 1982.
15. Tartini R, Kappenberger L, Steinbrunn W, Meyer UA:
Dangerous interactions between amiodarone and quinidine.
Lancet 1:1327, 1982.
16. Saal AK, Werner JA, Gross BW, Gorham JR, Graham EI,
Sears GK, Green HL: Interaction of amiodarone with quini-
dine and nrocainamide (abstr). Circulation 66(suppl 11):224,
1982. _
17. Southworth W, Friday KJ, Ruffy R: Possible amiodarone-
aprindine interaction (Letter to the Editor). AM HEART J
104~323, 1982.
18. Charlier R: Cardiac actions in the dog of a new antagonist of
adrenergic excitation which does not produce competitive
blockade of adrenoceptors. Br J Pharmacol 39:674, 1970.
19. Derrida JP, Ollagner J, Benami R, Haiat R, Chiche P:
Amiodarone and propranolol, a dangerous association. Nouv
Presse Med 8:1429, 1979.
20. Gallagher JD, Lieberman RW, Meranze J, Spielman SR,
Ellison N: Amiodarone-induced complications during coro-
nary artery surgery. Anesthesiology 55:186, 1981.
21. Adams PC, Nicholson MR, Storey GCA, Holt DW, Campbell
RWF: Amiodarone tissue distribution: Relation to adverse
effects (abstr). Br Heart J 49:297, 1983.
DISCUSSION
Dr. Akhtar: When did the casesof torsades de pointes
occur?
Dr. Singh: The one reported recently in Lancet
occurred early.
Dr. Zipes: You stressthat amiodaronehas more inter-
actions than other drugs, making its potentially danger-
ous. I wonder if that stressis needed. Other drugs show
interactions and affect the metabolism, excretion, phar-
macokinetics,or pharmacodynamicsof other drugs. Amio-
darone is no different. We must be careful when adding or
subtracting any drug in patients.
Dr. Marcus: Your point is well taken. On the other
hand, I know of no drug with so many reported interac-
tions, and I don’t believe that we’ve even scratched the
surface.
Dr. Zipes: Maybe we have to scratch the surface of
someof the others. We combined digoxin and quinidine
for a long time before we appreciated their interaction.
Dr. Scheinman: Is the effect of amiodaronecausedby
protein binding?
Dr. Marcus: I don’t know whether there is a common
mechanism.My hypothesis is that amiodaronemay affect
both hepatic metabolism and renal excretion of drugs.
Therefore it produces an unusually large number of
interactions.
Dr. Sobol: We presented data at the American Heart
Association meeting in November; although our data were
similar in terms of the mean increment in serum digoxin
level, we found increases in all patients receiving the
combination. There was at least one patient who had a
quadrupling of his serum digoxin level, and others who

930 Marcus
had threefold increases.Therefore more marked eleva-
tions of serum digoxin concentrations can occur.
Dr. Singh: I think that’s worth emphasizing.
Responsesvary greatly. I don’t think the problem is as
great during maintenance therapy as it is during the
loading phases,when digoxin concentrations go up very
steeply. Our unpublished data indicate that the increases
are much smaller during maintenance at 200 and 400
mglday than when we use 1500to 1600 mg daily loading
doses.
Dr. Prystowsky: Does a state of relative hypothy-
roidism affect renal clearanceof digoxin?
Dr. Marcus: Yes, Dr. Doherty’s work seemsto indicate
that hypothyroidism changesthe renal excretion of digox-
in.
Dr. Prystowsky: Do you think that some of the
effects you sawcould be related to a relative hypothyroid
state that might be induced in somepatients with amio-
darone?
Dr. Marcus: No. The effects we saware reproducible,
occur very early, and are not related to hypothyroidism.
Dr. McGovern: A number of the so-calledinteractions
between type 1 drugs and amiodarone have been due to
prolonged QT interals and the development of torsadesde
pointes, or polymorphic ventricular tachycardia, for days
to weeks after starting the combination. Similar QT
prolongation has been causedby amiodarone alone at the
samedose.
You emphasize that a beta blocker combined with
amiodarone may cause bradycardia and sinus arrest. I
know of 11 published casesof sinus arrest. Only one of
October, 1983
American Heart Journal
thesepatients wasreported to be receiving a beta blocker,
but someof the others may have been. Most patients were
receiving digoxin and amiodarone.Digoxin levels were not
known in most cases.
I agreethat we must observepatients carefully when we
combinebeta blockersand amiodarone,but what do we do
for our many patients with angina? Do we stop beta
blockers?Amiodarone alone can causeboth exacerbation
of arrhythmias or sinus arrest and worsening of heart
block.
Dr. Marcus: Certainly, amiodarone may cause tor-
sadesde pointes. Determining whether theseother drugs
can be implicated is still in a preliminary stage; there is
just one report. Prolongation of the QT interval may be
the mechanismof the interaction.
With regard to the use of beta blockers, we need more
information as to which patients would be likely to
develop problems. There have beenseveral reports in the
French literature of high-grade sinusarrest or even, in one
patient, complete absenceof any rhythm. The interactions
of amiodarone, beta blockers, someof the calcium block-
ers, and digoxin together would tend to facilitate depres-
sion of the sinus node, as well as the AV node.
Dr. Singht It is evident that the complex pharmacolog-
ic properties of amiodaroneprovide a significant potential
for not only pharmacodynamic but also pharmacokinetic
interactions with numerouscardioactive compounds.The
severity of such interactions may be determined by the
underlying clinical features, but a great deal of work
remains to be done in this area, especially regarding the
basic mechanismsthat might be involved.