Amiodarone

General adverse effects and reactions

Amiodarone prolongs the QT interval and can therefore cause dysrhythmias; there have also been reports of conduction disturbances. Abnormalities of thyroid function tests can occur without thyroid dysfunction, typically increases in serum T4 and reverse T3 and a reduction in serum T3. However, in up to 6% of patients frank thyroid dysfunction can occur (either hypothyroidism or hyperthyroidism). Several of the adverse effects of amiodarone are attributable to deposition of phospholipids in the tissues. These include its effects on the eyes, nerves, liver, skin, and lungs. Almost all patients develop reversible corneal microdeposits, which can occasionally interfere with vision. There are reports of peripheral neuropathy and other neurological effects. Changes in serum activities of aspartate transaminase and lactate dehydrogenase can occur without other evidence of liver disease, but liver damage can occur in the absence of biochemical evidence. Skin sensitivity to light occurs commonly, possibly due to phototoxicity. There may also be a bluish pigmentation of the skin. Interstitial pneumonitis and alveolitis have been reported and may be fatal. Lung damage due to amiodarone may be partly due to hypersensitivity. Tumor-inducing effects have not been reported.

Studies in the prevention of dysrhythmias

In a study of the use of implantable defibrillators or antidysrhythmic drugs (amiodarone or metoprolol) in 288 patients resuscitated from cardiac arrest, the defibrillator was associated with a slightly lower rate of all-cause mortality than the antidysrhythmic drugs [ ]. However, the small difference was not statistically significant. There was hyperthyroidism in three of those given amiodarone. Drug withdrawal was required in nine of those given amiodarone and 10 of those given metoprolol. There were deaths in five patients fitted with a defibrillator and two patients given amiodarone. There was crossover to the other therapy in 6% in each group, usually because of recurrence of the dysrhythmia. When sudden cardiac death was analysed, the reduction in mortality with defibrillation was much larger (61%). There were no differences in all-cause mortality and sudden death rates between those given amiodarone and those given metoprolol.

Amiodarone and carvedilol have been used in combination in 109 patients with severe heart failure and left ventricular ejection fractions of 0.25 [ ]. They were given amiodarone 1000 mg/week plus carvedilol titrated to a target dose of 50 mg/day. A dual-chamber pacemaker was inserted and programmed in back-up mode at a basal rate of 40. Significantly more patients were in sinus rhythm after 1 year, and in 47 patients who were studied for at least 1 year the resting heart rate fell from 90 to 59. Ventricular extra beats were suppressed from 1 to 0.1/day and the number of bouts of tachycardia over 167 per minute was reduced from 1.2 to 0.3 episodes per patient per 3 months. The left ventricular ejection fraction increased from 0.26 to 0.39 and New York Heart Association Classification improved from 3.2 to 1.8. The probability of sudden death was significantly reduced by amiodarone plus carvedilol compared with 154 patients treated with amiodarone alone and even more so compared with 283 patients who received no treatment at all. However, the study was not randomized, and this vitiates the results. The main adverse effect was symptomatic bradycardia, which occurred in seven patients; two of those developed atrioventricular block and four had sinoatrial block and/or sinus bradycardia; one patient developed slow atrial fibrillation.

A meta-analysis of 13 randomized trials has shown that both total mortality and sudden death or dysrhythmic death was less common over 24 months after randomization to amiodarone than in control subjects [ ].

Studies in atrial fibrillation

Cardiac glycosides such as digoxin are commonly used to treat uncomplicated atrial fibrillation. In those in whom digitalis is not completely effective or in whom symptoms (for example bouts of palpitation) persist despite adequate digitalization, a calcium antagonist, such as verapamil or diltiazem, can be added, or amiodarone used as an alternative.

The use of oral amiodarone in preventing recurrence of atrial fibrillation, for preventing recurrence after cardioversion or for pharmacological cardioversion of atrial fibrillation, has been reviewed [ ]. There is insufficient evidence to support its use as a first-line drug for preventing recurrence of atrial fibrillation or in preventing paroxysmal atrial fibrillation.

In 186 patients randomized equally to amiodarone 200 mg/day, sotalol 160–480 mg/day, or placebo, the incidence of atrial fibrillation after 6 months was higher in those taking placebo compared with amiodarone and sotalol and higher in those taking sotalol compared with amiodarone [ ]. Of the 65 patients who took amiodarone, 15 had significant adverse effects after an average of 16 months. There were eight cases of hypothyroidism, four of hyperthyroidism, two of symptomatic bradycardia, and one of ataxia. There were minor adverse effects in 9% of the patients, including gastrointestinal discomfort, nausea, photosensitivity, and eye problems. These patients had recurrent symptomatic atrial fibrillation. In contrast, only two patients using sotalol developed symptomatic bradycardia and one had severe dizziness.

In 208 patients with atrial fibrillation of various duration, including 50 with chronic atrial fibrillation, randomized to amiodarone or placebo, 80% converted to sinus rhythm after amiodarone compared with 40% of those given placebo [ ]. Amiodarone was given as an intravenous loading dose of 300 mg for 1 hour and 20 mg/kg for 24 hours, followed by 600 mg/day orally for 1 week and 400 mg/day for 3 weeks. Those who converted to sinus rhythm had had atrial fibrillation for a shorter duration and had smaller atria than those who did not convert. The shorter the duration of fibrillation and the smaller the atria the sooner conversion occurred. There was significant hypotension in 12 of the 118 patients who received amiodarone during the first hour of intravenous administration, but in all cases this responded to intravenous fluids alone. There was phlebitis at the site of infusion in 17 patients, and the peripheral catheter was replaced by a central catheter. There were no dysrhythmic effects.

In 40 patients with atrial fibrillation, some with severe heart disease (including cardiogenic shock in eight and pulmonary edema in 12), amiodarone 450 mg was given through a peripheral vein within 1 minute, followed by 10 ml of saline; 21 patients converted to sinus rhythm, 13 within 30 minutes and another 8 within 24 hours [ ]. There were two cases of hypotension, but in those that converted to sinus rhythm there was a slight increase in systolic blood pressure. There were no cases of thrombophlebitis. Efficacy is hard to judge from this study, because it was not placebo-controlled.

In 72 patients with paroxysmal atrial fibrillation randomized to either amiodarone 30 mg/kg or placebo, those who received amiodarone converted to sinus rhythm more often than those given placebo [ ]. The respective conversion rates were about 50% and 20% at 8 hours, and 87% and 35% after 24 hours. The time to conversion in patients who converted did not differ. One patient developed slow atrial fibrillation (35/minute) with a blood pressure of 75/55 mmHg. Three other patients who received amiodarone had diarrhea and one had nausea. In the control group two patients had headache, one had diarrhea, one had nausea, and two had episodes of sinus arrest associated with syncope during conversion to sinus rhythm; the last of these was thought to have sick sinus syndrome.

In a single-blind study 150 patients with acute atrial fibrillation were randomized to intravenous flecainide, propafenone, or amiodarone [ ]. At 12 hours there was conversion to sinus rhythm in 45 of 50 patients given flecainide, 36 of the 50 given propafenone, and 32 of the 50 given amiodarone. Thus, flecainide and propafenone were both more effective than amiodarone. There were no differences between the groups in the incidences of adverse effects; there was one withdrawal in each group, due to cerebral embolism in a patient given amiodarone, heart failure in a patient given propafenone, and atrial flutter in a patient given flecainide. There were no ventricular dysrhythmias during the study.

Amiodarone and magnesium have been compared in a placebo-controlled study to reduce the occurrence of atrial fibrillation in 147 patients after coronary artery bypass graft surgery [ ]. Amiodarone was given as an infusion of 900 mg/day for 3 days and magnesium by infusion of 4 g/day for 3 days. The cumulative occurrences of atrial fibrillation with placebo, amiodarone, and magnesium were 27%, 14%, and 23% respectively. These differences were not significant. Amiodarone delayed the onset of the first episode of dysrhythmia significantly, but the slight benefit was associated with a longer period of invasive monitoring and was not considered worthwhile. Patients who were more likely to develop atrial fibrillation were older and had a plasma magnesium concentration at 24 hours of under 0.95 mmol/l. Patients who were given amiodarone had a slightly higher rate of adverse events, including hypotension, atrioventricular block, and bradycardia; adverse events led to withdrawal in four cases.

Amiodarone, sotalol, and propafenone have been compared for the prevention of atrial fibrillation in 403 patients who had had at least one episode of atrial fibrillation within the previous 6 months; the study was not placebo-controlled [ ]. The rate of recurrence of atrial fibrillation was significantly higher in those given sotalol or propafenone than in those given amiodarone. During the study nine patients given amiodarone died, compared with eight given sotalol or propafenone. Four deaths were thought to be dysrhythmic, three in patients given amiodarone. There were major non-fatal adverse events in 36 of the 201 patients given amiodarone and in 35 of the 202 patients given propafenone or sotalol. These included one case of torsade de pointes in a patient who received propafenone, and congestive heart failure in 11 patients given amiodarone and nine given sotalol or propafenone. There were strokes and intracranial hemorrhages in one patient given amiodarone and nine patients given sotalol or propafenone, of whom most were taking warfarin at the time. In all, 68 of the patients who were given amiodarone and 93 of those given sotalol or propafenone withdrew from the study; 17 of those taking amiodarone withdrew because of lack of efficacy compared with 56 of those taking sotalol or propafenone; 36 of those who took amiodarone withdrew because of adverse events compared with 23 of those who took sotalol or propafenone, and this was almost statistically significant.

Amiodarone, propafenone, and sotalol have also been compared in the prevention of atrial fibrillation in 214 patients with recurrent symptomatic atrial fibrillation. They were randomized to amiodarone 200 mg/day, propafenone 450 mg/day, or sotalol 320 mg/day. There was recurrence of atrial fibrillation in 25 of the 75 patients who took amiodarone compared with the 51 of 75 who took sotalol and 24 of the 64 who took propafenone. There were adverse effects requiring withdrawal of treatment in 14 patients who took amiodarone, five who took sotalol and one who took propafenone while they were in sinus rhythm. These effects included symptomatic bradycardia in three patients, hyperthyroidism in six, hypothyroidism in four, and ataxia in one patient who took amiodarone. In those taking sotalol the adverse effects were bradycardia in three and severe dizziness in two. In the one patient in whom propafenone was withheld the reason was symptomatic bradycardia. Thus, amiodarone and propafenone were both more effective than sotalol, but amiodarone also caused more adverse effects requiring withdrawal [ ].

In a meta-analysis of five randomized, placebo-controlled trials of amiodarone 200–1200 mg/day for 2–7 days in the treatment of postoperative atrial fibrillation and flutter in 764 patients, the incidence of adverse events with amiodarone was no greater than with placebo [ ].

In a meta-analysis of five randomized, placebo-controlled trials of intravenous amiodarone about 500–2200 mg over 24 hours in the treatment of recent-onset atrial fibrillation in 410 patients, the incidence of adverse events was 27% with amiodarone and 11% with placebo [ ]. Intravenous amiodarone was significantly more effective than placebo in producing cardioversion. The most common adverse effects of intravenous amiodarone were phlebitis, bradycardia, and hypotension; most of these effects were not considered to be dose-limiting.

Of 85 patients with persistent atrial fibrillation after balloon mitral valvotomy given amiodarone (600 mg/day for 2 weeks and 200 mg/day thereafter), 33 converted to sinus rhythm [ ]. Of the other 52 patients, who underwent DC cardioversion at 6 weeks, 41 converted to sinus rhythm. Six patients had adverse effects attributable to amiodarone. Five had mild gastrointestinal symptoms, such as abdominal discomfort and nausea. One developed hypothyroidism after 3 months, which resolved when the dosage of amiodarone was reduced to 100 mg/day.

In 83 patients (27 women, 56 men; mean age 61 years) disopyramide, propafenone, or sotalol were used to prevent recurrence after elective electrical cardioversion for persistent atrial fibrillation [ ]. If there was recurrence cardioversion was repeated and the patient was given one of the other antidysrhythmic drugs. If there was further recurrence, amiodarone was used, a third cardioversion was performed, and, if sinus rhythm was restored, amiodarone 100–200 mg/day was continued. Patients in whom the initial cardioversion was not successful were given amiodarone and underwent repeated cardioversion. The follow-up duration was 12 months. The first electrical cardioversion was effective in 44 (53%) patients, and after 1 year 23 (52%) of them were still in sinus rhythm. None of the patients who underwent a second cardioversion and received a second antidysrhythmic drug stayed in sinus rhythm. Amiodarone as a third antidysrhythmic agent was effective in 10 (48%) patients. After 12 months of antidysrhythmic drug therapy sinus rhythm was maintained in 75% of patients in whom the first cardioversion had been effective, accounting for 40% of all the patients selected for cardioversion. In the 83 patients, sequential antidysrhythmic treatment effectively maintained sinus rhythm in 54 (65%), of whom 31 (57%) took amiodarone. The authors concluded that repeated electrical cardioversion and antidysrhythmic drug therapy enabled maintenance of sinus rhythm in 68% of patients for 1 year, that there was limited efficacy of the first antidysrhythmic drug given after a first effective electrical cardioversion, regardless of the drug used, excluding amiodarone, and that when atrial fibrillation recurred, a second antidysrhythmic drug, other than amiodarone, was completely ineffective. There were very few adverse events in this study. One patient taking amiodarone developed hyperthyroidism and two had symptomatic bradycardia.

Amiodarone 30 mg/kg orally for the first 24 hours plus, if necessary, 15 mg/kg over 24 hours has been compared with propafenone 600 mg in the first 24 hours plus, if necessary, 300 mg in the next 24 hours in 86 patients with recent onset atrial fibrillation [ ]. Conversion to sinus rhythm occurred faster with propafenone (2.4 hours) than amiodarone (6.9 hours). However, by 24 hours and 48 hours the same proportions of patients were in sinus rhythm; one patient given amiodarone had a supraventricular tachycardia and one a non-sustained ventricular tachycardia.

The effects of additional intravenous amiodarone (300 mg in 1 hour followed by 15 mg/kg over 24 hours) have been studied in 45 patients with acute atrial fibrillation who were already taking oral amiodarone for maintenance of sinus rhythm [ ]. In 20 of 23 patients given amiodarone there was conversion to sinus rhythm, compared with 13 of 22 who were given placebo. There were no prodysrhythmic effects and the only adverse effect of intravenous amiodarone was thrombophlebitis in two patients.

In 44 patients who underwent percutaneous balloon mitral commissurotomy for chronic persistent atrial fibrillation, with a procedural success rate of 100% and no immediate morbidity or mortality, amiodarone maintained sinus rhythm in eight patients compared with none in the control group [ ]. The adverse effects of amiodarone included bradycardia in two patients and shortness of breath in one; the last required drug withdrawal. Another patient developed long sinus pauses at 15 months and was treated with a permanent pacemaker without withdrawing amiodarone. Otherwise, there were no serious adverse effects or electrocardiographic abnormalities.

In a double-blind, placebo-controlled trial 665 patients who were taking anticoagulants and had persistent atrial fibrillation were randomized to amiodarone (n = 267), sotalol (n = 261), or placebo (n = 137) for 1.0–4.5 years [ ]. Amiodarone and sotalol were equally effective in producing cardioversion to sinus rhythm (27% and 24% versus placebo 0.8%), but the effect of amiodarone lasted significantly longer (487, 74, and 6 days according to intention to treat, and 809, 209, and 13 days according to treatment received). There were no significant differences in the rates of adverse events, except minor bleeding , which was significantly more common with amiodarone than sotalol or placebo (8.33 versus 6.37 and 6.71 per 100 patient-years). The rates of major bleeding were 2.07, 3.10, and 3.97 per 100 patient-years; of minor strokes 1.19, 0.68, and 0.96; and of major strokes 0.87, 2.03, and 0.95. There were two cases of non-fatal adverse pulmonary effects with amiodarone and one with placebo. There was one case of non-fatal torsade de pointes with sotalol.

In a systematic review of the efficacy and safety of amiodarone for pharmacological cardioversion of recent-onset atrial fibrillation in 21 studies, amiodarone was efficacious in 34–69% with bolus-only regimens, and 55–95% with a bolus followed by an infusion [ ]. The highest 24-hour conversion rates occur with an intravenous regimen of 125 mg/hour until conversion or a maximum of 3 g and an oral regimen of 25–30 mg/kg given as a single loading-dose (over 90% and over 85% respectively). Most conversions occur after 6–8 hours of the start of therapy. Predictors of successful conversion are shorter duration of atrial fibrillation, smaller left atrial size, and higher amiodarone dose. Amiodarone is not superior to other antidysrhythmic drugs but is relatively safe in patients with structural heart disease and in those with depressed left ventricular function. No major prodysrhythmic events, such as sustained ventricular tachycardia, ventricular fibrillation, or torsade de pointes, were reported in these studies. There were minor cardiac effects: first-degree atrioventricular block, self-limited sinus bradycardia, hypotension, and non-sustained ventricular tachycardia. These adverse effects were more common after intravenous administration. Asymptomatic sinus bradycardia was reported in up to 10% of patients and hypotension in up to 18% of patients who received intravenous amiodarone. All the episodes of hypotension were transient and responded to saline volume expansion or inotropic support. Hypotension with intravenous amiodarone is reportedly due to the vehicle. Phlebitis at the amiodarone infusion site occurs up to 16% of patients. Other rare adverse effects were nausea, diarrhea, blurred vision, and allergic reactions. Gastrointestinal adverse effects were predominantly reported after oral administration.

In another systematic review of the studies of the use of amiodarone in the treatment of atrial fibrillation, 21 studies met the eligibility criteria, including 10 of those covered in the systematic review mentioned above [ ]. Bradydysrhythmias and hypotension were the most commonly reported adverse effects. Death rates were reported in 18 studies; there were five deaths among 816 patients given amiodarone and five among 696 in comparison groups.

However, information about adverse events in these randomized trials was inconsistently reported and too scanty to allow proper analysis. This stresses yet again the need for standard methods of reporting adverse events in clinical trials.

Studies in atrial flutter

Antidysrhythmic drugs have been compared with radiofrequency ablation in 61 patients with atrial flutter [ ]. Drug treatment was with at least two drugs, one of which was amiodarone. Of the 30 patients who took drug therapy, 19 needed to come into hospital one or more times, whereas after radiofrequency ablation that happened in only seven of 31 cases. In those who took the antidysrhythmic drugs the mean number of drugs was 3.4 and the range of drugs used was very wide. Quality-of-life and symptoms scores improved significantly in those in whom radiofrequency ablation was used, but not in those who took the antidysrhythmic drugs, apart from the symptom of palpitation, which improved in both groups, but to a greater extent in the non-drug group. Adverse effects were not discussed in this study, but it is clear that it suggests that radiofrequency ablation is to be preferred in these patients.

Studies in ventricular dysrhythmias

The effects of amiodarone in 55 patients with sustained ventricular tachycardia after myocardial infarction have been assessed in a long-term follow-up study [ ]. The patients underwent programmed ventricular stimulation after having been loaded with amiodarone. They were divided into those in whom ventricular tachydysrhythmias could be induced or not, and all were then given amiodarone 200 mg/day. In 11 cases a cardioverter defibrillator was implanted, because the first episode of ventricular tachycardia had been poorly tolerated or had caused hemodynamic instability. A defibrillator was also implanted in five other cases during follow-up, because of recurrence of dysrhythmias. There was a non-significant trend to a difference between the cumulative rates of dysrhythmias during long-term follow-up, with more events in those in whom a dysrhythmia had been inducible after loading. However, mortality rates in the two groups did not differ, and was around 25% at a mean follow-up of 42 months. Survival was significantly higher in patients with a left ventricular ejection fraction over 0.4, and the lower the left ventricular ejection fractions the higher the mortality. Amiodarone was withdrawn in six patients after a mean of 34 months because of neuropathy (n = 1), hypothyroidism (n = 1), prodysrhythmia with incessant ventricular tachycardia (n = 2), and non-specific adverse effects (n = 2). There was no pulmonary toxicity and no cases of torsade de pointes. In two patients there was evidence of hypothyroidism, mild neuropathy, and skin discoloration, but these events did not lead to withdrawal. In two patients the doses of amiodarone was reduced to 100 mg/day because of sinus bradycardia.

In a comparison of amiodarone (n = 23) with sotalol (n = 22) in patients with spontaneous sustained ventricular tachydysrhythmias secondary to myocardial infarction, sotalol was much more effective, 75% of those taking it remaining free of dysrhythmias compared with 38% of those taking amiodarone [ ]. Adverse effects requiring withdrawal occurred in 17% of those taking amiodarone at a median time of 3.5 months. The adverse effects included malaise, rash, headaches, flushing, and dyspnea due to pulmonary fibrosis.

Studies in myocardial infarction and heart failure

There have been reviews of the results of major trials of amiodarone after myocardial infarction [ ] and in chronic heart failure [ ].

• In the Basel Antiarrhythmic Study of Infarct Survival (BASIS) amiodarone significantly reduced all-cause mortality from 13% to 5%, compared with no antidysrhythmic drug therapy [ ].

• In the Polish Arrhythmia Trial (PAT) amiodarone reduced all-cause mortality from 10.7% to 6.9% compared with placebo and cardiac mortality from 10.7% to 6.2% [ ].

• In the Spanish Study of Sudden Death (SSD) amiodarone reduced all-cause mortality from 15.4% to 3.5% compared with metoprolol; however, the mortality in those receiving no antidysrhythmic drugs at all was only 7.7%, and in those the effect of amiodarone was not significant [ ].

• In the European Myocardial Infarction Arrhythmia Trial (EMIAT) amiodarone reduced the risk of dysrhythmic deaths from 8.5% to 4.1% compared with placebo [ ].

• In the Canadian Amiodarone Myocardial Infarction Arrhythmia Trial (CAMIAT) amiodarone reduced dysrhythmic deaths from 6.0% to 3.3% compared with placebo; non-dysrhythmic deaths were not affected [ ].

In a meta-analysis of 10 studies of the use of amiodarone in patients with heart failure, the overall odds ratio for mortality with amiodarone compared with placebo was 0.79 (95% CI = 0.68, 0.92). The corresponding odds ratio for adverse effects was 2.29 (1.97, 2.66) [ ]. The benefit to risk ratio of the use of amiodarone in these patients is not yet clear. The dosage of amiodarone in these studies varied from 50 to 400 mg/day, with an average of around 250 mg/day.

Cardiovascular

The incidence of cardiac dysrhythmias with amiodarone is under 3% [ ], lower than with many other antidysrhythmic drugs, and several randomized controlled trials have failed to show any prodysrhythmic effect [ ].

Other cardiac effects that have been reported include sinus bradycardia, atrioventricular block, infra-His block, asystole, and refractoriness to DC cardioversion [ ].

There is a risk of hypotension and atrioventricular block when amiodarone is given intravenously.

Cardiogenic shock has been reported in 73-year-old woman with a dilated cardiomyopathy who had digitalis and amiodarone toxicity [ ].

Ventricular dysrhythmias

Amiodarone can prolong the QT interval, and this can be associated with torsade de pointes [ ], although this is uncommon. This effect is potentiated by hypokalemia [ ].

Ventricular dysrhythmias due to drugs can be either monomorphic or polymorphic. The class Ia drugs are particularly likely to cause polymorphic dysrhythmias, as is amiodarone (although to a lesser extent). In contrast, the class Ic drugs are more likely to cause monomorphic dysrhythmias [ ].

• A 40-year-old woman developed torsade de pointes within the first 24 hours of intravenous administration of amiodarone 150 mg followed by 35 mg/hour [ ]. The association with amiodarone was confirmed by subsequent rechallenge.

• Three boys with congenital cardiac defects developed polymorphous ventricular tachycardia after having been given intravenous amiodarone; two died [ ].

• An 8-day-old boy was given intravenous amiodarone 5 mg/kg over 60 minutes followed by 10 mg/kg/day for a postoperative junctional ectopic tachycardia after a cardiac operation. He developed ventricular fibrillation 12 hours later, but recovered with defibrillation and internal cardiac massage. His serum amiodarone concentration was 1–2.5 mg/l, within the usual target range.

• A 3-month-old boy underwent a cardiac operation and 6 hours later developed a junctional ectopic tachycardia. He was given amiodarone as a continuous intravenous infusion of 10 mg/kg/day for 3 hours and developed ventricular fibrillation, from which he was not resuscitated. The serum amiodarone concentration was 0.3 mg/l.

• A 3-month-old boy developed a postoperative junctional ectopic tachycardia 48 hours after operation and was given a continuous intravenous infusion of amiodarone 10 mg/kg/day. After 2 hours he developed ventricular fibrillation and was not resuscitated. His serum amiodarone concentration was in the target range.

It is not clear that the dysrhythmias in these cases were due to amiodarone, particularly since the doses had been very low and the serum concentrations no higher than the usual target range; QT intervals were not reported.

• A 79-year-old woman took amiodarone 4800 mg over 6 days and developed a polymorphous ventricular tachycardia; the associated precipitating factors were a prolonged QT interval and hypokalemia [ ].

• A 71-year-old Japanese man with bouts of sustained monomorphic ventricular tachycardia, in whom non-sustained polymorphic ventricular tachycardia was induced by rapid pacing during electrophysiological studies, was given amiodarone and developed three different types of sustained monomorphic ventricular tachycardia, with slightly different cycle lengths, induced and terminated by rapid pacing [ ].

The authors proposed that amiodarone had modulated the threshold of induction and/or termination of ventricular tachycardia.

• In an 84-year-old woman torsade de pointes occurred after oral amiodarone therapy for 4 days in the presence of multiple exacerbating factors, including hypokalemia and digoxin toxicity [ ]. Transient prolongation of the QT interval during bladder irrigation prompted the episode. When amiodarone was withdrawn, bladder irrigation did not induce torsade de pointes, despite hypokalemia and hypomagnesemia.

• A 69-year-old woman with a history of coronary heart disease, myocardial infarction, and paroxysmal atrial fibrillation had an occipital stroke [ ]. She was given amiodarone 600 mg/day, beta-acetyldigoxin 0.1 mg/day, and bisoprolol 1.25 mg/day, and developed significant QT interval prolongation (maximum 700 ms; QT _c 614 ms) and repetitive short-lasting torsade de pointes, which terminated spontaneously. Her serum electrolytes were normal and plasma concentrations of digoxin (1.8 ng/ml) and amiodarone (1.9 μg/ml) were within the usual target ranges.

In the first case the authors speculated that increased vagal tone during bladder irrigation was responsible for QT interval prolongation associated with bradycardia in the presence of amiodarone. In the second case the authors suggested that the dysrhythmia was due to the triple combination of amiodarone with a beta-blocker and digitalis in a patient with atrial fibrillation and structural heart disease; again it is possible that bradycardia played a part.

Of five patients with torsade de pointes due to amiodarone, three had hypokalemia and those with negative T waves were at greater risk of ventricular fibrillation than those with positive T waves [ ]. Of six patients with torsade de pointes taking chronic amiodarone, five were women, three were taking drugs that inhibit CYP3A4 (loratadine or trazodone), three had hypokalemia, and four had reduced left ventricular function [ ].

Of 189 patients, five had torsade de pointes and all five had prolonged QT intervals [ ]. Two of the five, all women, also had raised blood glucose concentrations, and the authors suggested that hyperglycemia is a risk factor for torsade de pointes. However, the number of cases reported in this series was too small to justify such a conclusion.

It has been suggested that women are more likely to develop torsade de pointes than men in response to antidysrhythmic drugs [ ], and this has been confirmed in the case of amiodarone in a study of 189 patients given intravenous amiodarone [ ]. This is also reminiscent of the finding that prolongation of the QT interval due to quinidine is greater in women than in men at equivalent serum concentrations [ ].

T wave alternans is an occasional presentation, in association with ventricular dysrhythmias [ ].

• A 65-year-old man with atrial fibrillation was given intravenous amiodarone 450 mg over 30 minutes followed by 900 mg over 24 hours [ ]. He reverted to sinus rhythm, but the electrocardiogram showed giant T wave alternans with a variable QT interval (0.52–0.84 seconds). He had a short bout of torsade de pointes and was given magnesium. Two days later the electrocardiogram was normal.

• In a 62-year-old man with dilated cardiomyopathy and an implantable cardioverter defibrillator for ventricular tachycardia, microvolt T wave alternans differed when amiodarone was added [ ]. The onset heart rate with T wave alternans was lower and the alternans voltage higher with amiodarone than without it.

The effects of amiodarone appeared to be related to exacerbations of ventricular tachycardia and an increased defibrillation threshold.

Atrial dysrhythmias

Amiodarone has been reported to cause atrial flutter in 10 patients who had been given it for paroxysmal atrial fibrillation [ ]. In nine of those the atrial flutter was successfully treated by catheter ablation. However, during a mean follow-up period of 8 months after ablation, atrial fibrillation occurred in two patients who had continued to take amiodarone; this was a lower rate of recurrence than in patients in whom atrial flutter was not associated with amiodarone. The authors therefore suggested that in patients with atrial flutter secondary to amiodarone given for atrial fibrillation, catheter ablation allows continuation of amiodarone therapy.

Amiodarone can sometimes cause atrial flutter, even though it is also used to treat it [ ]. There has been a report of seven cases (six men and one woman, aged 34–75 years) of 1:1 atrial flutter with oral amiodarone [ ]. Four of them had underlying cardiac disease; none had hyperthyroidism. The initial dysrhythmia was 2:1 atrial flutter (n = 4), 1:1 atrial flutter (n = 2), or atrial fibrillation (n = 1). One patient was taking amiodarone 200 mg/day and one was taking 400 mg/day plus carvedilol. The other five all received loading doses of 9200 (sd 2400) mg over 10 (sd 4) days. There was an adrenergic trigger factor (exertion, fever, esophageal stimulation, or a beta-adrenoceptor agonist aerosol) in five patients. One required emergency cardioversion.

In another case there was prolongation of the flutter cycle and infra-Hissian block [ ].

Of 136 patients with atrial fibrillation treated with either amiodarone (n = 96) or propafenone (n = 40), 15 developed subsequent persistent atrial flutter, nine of those taking amiodarone and six of those taking propafenone [ ]. In all cases radiofrequency ablation was effective. It is not clear to what extent these cases of atrial flutter were due to the drugs, although the frequency of atrial flutter in previous studies with propafenone has been similar. Atrial enlargement was significantly related to the occurrence of persistent atrial flutter in these patients.

Bradycardia

Bradycardia has been reported to occur in about 5% of patients taking amiodarone [ ].

Of 2559 patients admitted to an intensive cardiac care unit over 3 years, 64 with major cardiac iatrogenic problems were reviewed [ ]. Of those, 58 had dysrhythmias, mainly bradydysrhythmias, secondary to amiodarone, beta-blockers, calcium channel blockers, electrolyte imbalance, or a combination of those. Amiodarone was implicated in 19 cases, compared with 44 cases attributed to beta-blockers and 28 to calcium channel blockers. Of the 56 patients with sinus bradycardia, 10 were taking a combination of amiodarone and a beta-blocker, six were taking amiodarone alone, and three were taking amiodarone plus a calcium channel blocker.

Amiodarone is superior to placebo for cardioversion of recent onset atrial fibrillation, and even though the onset of conversion is delayed compared with class Ic drugs, efficacy is similar at 24 hours [ ]. However, among 8770 patients aged over 65 years with a new diagnosis of atrial fibrillation who had had a previous myocardial infarction there were 477 cases of bradydysrhythmias requiring a permanent pacemaker and they were matched 1:4 to 1908 controls; the use of amiodarone was associated with an increased risk of pacemaker insertion (OR = 2.14; 95% CI = 1.30, 3.54). This effect was modified by sex, with a greater risk in women (OR = 3.86; 95% CI = 1.70, 8.75) than in men (OR = 1.52; 95% CI = 0.80, 2.89).

During 409 trials of antidysrhythmic drugs to maintain sinus rhythm in patients with previous atrial fibrillation or atrial flutter amiodarone was used in 212 patients (52%), type 1C drugs in 127 (31%), sotalol in 37 (9.0%), and a type 1A drug in 33 (8.1%) [ ]. There were adverse events in 17 patients: three died, three had bradycardia that required permanent pacemaker implantation, and 11 had bradycardia requiring a reduction in drug dosage. Most of the events were due to bradycardia in patients who received amiodarone. There was a significant association between amiodarone-associated bradycardia and female sex. The only event that occurred during the first 48 hours was an episode of bradycardia in a patient who received amiodarone and was managed as an out-patient.

Heart block

• In a 66-year-old woman taking amiodarone 1200 mg/week there was marked prolongation of the QT interval, to 680 ms; the succeeding P waves fell within the refractory period of the preceding beat and were unable to institute conduction [ ]. This resulted in 2:1 atrioventricular block. Amiodarone was withdrawn and the QT interval normalized with a time-course consistent with the long half-life of amiodarone. A subsequent rechallenge with intravenous amiodarone caused further prolongation of the QT interval.

The authors hypothesized that this patient had a silent mutation in one of the genes coding for the two major potassium channel proteins (IKr or IKs) that are involved in the mode of action of amiodarone. However, they did not present any genetic studies to support this hypothesis.

Pacemaker requirements

Of 8770 patients aged 65 years or over with a new diagnosis of atrial fibrillation, 477 had bradydysrhythmias requiring a permanent pacemaker and were matched with 1908 controls [ ]. The use of amiodarone was associated with an increased risk of pacemaker insertion (OR = 2.14; 95% CI = 1.30, 3.54). Women had a greater risk than men (OR = 3.86 versus 1.52).

In a retrospective study of 82 patients with an implanted pacemaker cardioverter defibrillator, those who were also taking amiodarone (for 24 consecutive months without interruption) had a significant three-fold increase in episodes of defibrillation compared with those who did not take amiodarone [ ]. This is an unexpected finding, for which the authors had no explanation. However, the finding was vitiated by the retrospective nature of the study.

Hypotension

Intravenous amiodarone can cause hypotension in anesthetized patients undergoing cardiac surgery. In a prospective double-blind study, 30 patients undergoing coronary artery bypass graft surgery were randomly assigned to receive intravenous amiodarone or placebo [ ]. At 6 minutes, amiodarone reduced mean arterial pressure by 14 mmHg and placebo reduced it by 4 mmHg. The changes in mean arterial pressure and systolic and diastolic blood pressures between groups were statistically different for the first 15 minutes after drug administration. Hypotension required intervention in three of 15 patients given amiodarone and none of the 15 given placebo. The mean heart rate was 12/minute less after amiodarone, but pulmonary artery pressure, central venous pressure, mixed venous oxygen saturation, and fractional left ventricular area change were not different between the groups. The authors concluded that the hypotension that amiodarone caused during the first 15 minutes after administration was not accompanied by altered left ventricular function, suggesting that selective arterial vasodilatation was the primary cause.

Hypotension has been attributed to solvents in the intravenous formulation of amiodarone, and this hypothesis was supported by the observation that in four trials an aqueous formulation caused hypotension in only three of 278 patients and only during bouts of ventricular tachycardia [ ]. In addition, six patients had cardiac dysrhythmias or heart block, two had erythema and/or pain at the site of injection, and two had thrombophlebitis.

Respiratory

There have been reviews of the lung complications of amiodarone toxicity [ ] and of its mechanisms [ ].

Nervous system

The most common forms of neurological damage attributed to amiodarone are tremor, peripheral neuropathy, and ataxia [ ]. Other effects that have been reported include delirium [ ], Parkinsonian tremor [ ], and pseudotumor cerebri [ ]. Acute myolysis has been described at high dose [ ]. The peripheral neuropathy is probably due to intracellular lipidosis [ ].

• Periodic ataxia has been attributed to amiodarone in a 67-year-old man taking amiodarone 200 mg/day [ ]. The ataxia responded to acetazolamide and eventually to withdrawal of amiodarone. It recurred with rechallenge.

• An 84-year-old woman with hypertrophic obstructive cardiomyopathy and paroxysmal atrial fibrillation developed a progressively debilitating ataxia, which abated over 4 months after withdrawal of amiodarone [ ]. Despite the long half-life of amiodarone, her symptoms began to improve after several days, and she was walking without assistance within 1 week.

Amiodarone-induced neuromyopathy has been studied in three patients by a review of their records, electromyography, and histopathology of muscle and nerve [ ]. Two patients had a slightly asymmetric, mixed, but primarily demyelinating sensorimotor polyneuropathy and the third had an acute neuropathy resembling Guillain–Barré syndrome. Creatine kinase activity did not correlate with clinical or electromyographic evidence of myopathy. In the peripheral nerves there was demyelination, some axon loss, and a variable number of characteristic lysosomal inclusions. Muscle specimens from two patients showed evidence of a vacuolar myopathy. After withdrawal of amiodarone, two patients improved and one died with a cardiac dysrhythmia.

Benign orgasmic headache has been associated with amiodarone [ ].

• A 52-year-old man, who had taken amiodarone 800 mg/day for 7 months, developed acute, severe, throbbing headaches precipitated by coitus and occasionally other forms of exertion. An MRI scan of the brain was normal. When the dose of amiodarone was reduced to 200 mg/day the headaches diminished in frequency and severity. When the dose was increased again to 400 mg/day they increased in frequency and severity. The amiodarone was withdrawn and the headaches resolved.

Amiodarone was originally developed as a vasodilator, and that may have been the cause of headaches in this case.

Sensory systems

The adverse effects of amiodarone on the eyes have been reviewed [ ]. The most common effect is corneal microdeposits. In some cases chronic blepharitis and conjunctivitis have been reported [ ], but the relation of these to amiodarone is not clear.

Psychological, psychiatric

Delirium has rarely been reported with amiodarone [ ].

• A 54-year-old man with no previous psychiatric history took amiodarone 400 mg bd [ ]. After a few days he became depressed and paranoid, suffered from insomnia, and had rambling speech. The dosage of amiodarone was reduced to 200 mg bd and he improved. However, 3 days later he became confused, with tangential thinking, labile effect, and a macular rash on the limbs. His serum sodium was reduced at 127 mmol/l and his blood urea nitrogen was raised. A CT scan of the head was normal. Amiodarone was withdrawn and 4 days later he was alert and oriented. About a week later he started taking amiodarone again and within 4 days became increasingly agitated, confused, and paranoid. He once more recovered after withdrawal of amiodarone.

• Depression has been attributed to amiodarone in a 65-year-old woman who was taking amiodarone (dosage not stated) [ ]. Because the mode of presentation was atypical in onset, course, duration, and its response to antidepressant drugs, amiodarone was withdrawn, and she improved rapidly. There was no evidence of thyroid disease.

Endocrine