Management of Amiodarone-Related Thyroid Problems

Abstract

Amiodarone is a highly effective and well-established antiarrrhythmic drug. It can be used to treat supraventricular and ventricular tachyarrhythmias and has the added advantage of being well tolerated in patients with impaired left ventricular systolic function with a low incidence of arrhythmic events, such as torsades de pointes. However, owing to its marked lipid affinity, it is highly concentrated in tissues and is linked to a number of adverse effects, including thyroid dysfunction. Amiodarone can lead to both hypothyroidism (amiodarone-induced hypothyroidism) and less commonly hyperthyroidism (amiodarone-induced thyrotoxicosis) and relates to high iodine content within the molecule as well as to several unique intrinsic properties of amiodarone. Dronedarone is a recently approved antiarrhythmic drug. It is structurally very similar to amiodarone, however the iodine moiety, present with amiodarone has been removed and replaced with a methylsulfonamide group to reduce fat solubility and adverse effects. We present an overview of the effects of amiodarone on thyroid function and the treatment options available, as well as a brief insight into dronedarone and its potential as an alternative to amiodarone.

Introduction

Amiodarone is the most commonly used antiarrhythmic drug worldwide [Singh, 2008]. It is effective in the treatment of both supraventricular and ventricular tachyarrhythmias and has the added advantage of being well tolerated in patients with both normal and impaired left ventricular systolic function [Singh, 2008; Vassallo and Trohman, 2007]. It was originally introduced as an antianginal compound in 1962 [Rao et al. 1986] following its discovery in 1961 at the Labaz company in Belgium [Deltour et al. 1962]. Its unique antiarrhythmic and antifibrilla-tory properties were not really appreciated until 1970 [Singh, 1970] and it was only approved for use as an antiarrhythmic in the USA in 1985. Amiodarone affects both myocardial depolarization and repolarization. In addition to its principle class III (potassium channel blockade) antiarrhythmic effects, amiodarone has class I (sodium channel blockade), class II (noncompetitive α- and β-blocking) and class IV (calcium channel activity) related actions [Singh, 2008]. Despite its impressive profile as an antiarrhythmic agent, its use is hampered by a number of potential adverse effects, including thyroid dysfunction [Papiris et al. 2010; Van Erven and Schalij, 2010; Vassallo and Trohman, 2007; Bongard et al. 2006]. Amiodarone can lead to both hypothyroidism and less commonly hyperthyroidism. Dronedarone is a relatively new anti-arrhythmic drug which was designed to replicate the potency of amiodarone without its associated deleterious side effects. In this article we present an overview of amiodarone-related thyroid dysfunction and a brief insight into the use of dronedarone as an amiodarone replacement in certain patient groups.

Pharmacology and pharmacokinetics of amiodarone

Amiodarone is a benzofuran compound (Figure 1) that contains approximately 37% iodine by weight and bears a remarkable structural resemblance to thyroid hormones (Figure 1(a)—(c)) [Rao et al. 1986]. Therefore, a patient taking a standard 200 mg daily dose of amiodarone ingests 75 mg of organic iodine each day. Subsequent deiodination through drug metabolism results in the daily release of approximately 6 mg of free circulating iodine. This equates to a 20–40 times higher than the average daily iodine intake in the USA/UK of 0.15–0.30 mg [Basaria and Cooper, 2005; Martino et al. 2001]. The oral bioavailability of amiodarone is both slow and highly variable and averages 40% [Cohen-Lehman et al. 2010], ranging from 22% to 95% and is higher when taken with food [Siddoway, 2003; Latini et al. 1984].

Figure 1.

Structure of amiodarone and its close relationship to thyroxine (T4), triiodothyronine (T3) and dronedarone. Structure of (a) amiodarone, (b) thyroxine T4, (c) triiodothyronine (T3), (d) dronedarone.

Amiodarone is metabolized in the liver via the cytochrome P450 system and hence increases the concentrations of medications such as statins, calcium channel blocking agents, quinidine, and flecainide. Amiodarone increases digoxin levels, hence concomitant digoxin should either be stopped or the dose reduced (50%) [Van Erven and Schalij, 2010]. Amiodarone inhibits the metabolism of coumadin (e.g. warfarin) derivates, potentiating its anticoagulant effect. This can lead to rapid and unpredictable increases (two to three times treatment levels) in prothrombin times. Hence it is recommended that anticoagulant doses should be reduced by one third to one half with meticulous monitoring of prothrombin times [Van Erven and Schalij, 2010]. Amiodarone is excreted in the faeces with less than 1% of the dose being excreted unchanged in the urine. The only major active metabolite of amiodarone is desethylamiodarone (DEA).

Owing to its benzene ring, amiodarone is highly lipophilic leading to a strong tissue affinity and a very large volume of distribution which is estimated at 60–66 1/kg body weight [Cohen-Lehman et al. 2010; Vassallo and Trohman, 2007]. This leads to a delayed onset of antiarrhythmic action varying from 2 days to as much as 3 weeks following oral therapy because it relies on cumulative dose. Hence when starting amiodarone, a high loading dose is used until arrhythmia suppression or apparent steady state is achieved, which is usually in 2–4 weeks. Furthermore, following therapy cessation its action may persist for a considerable time. Whilst a 50% reduction in plasma concentration may be achieved over 3–10 days, this is followed by a longer terminal half life of 13–142 days until tissue stores are fully depleted [Vassallo and Trohman, 2007; Zipes et al. 1984]. Overall the average half life of amiodarone is 40 days with that of DEA, its principle metabolite, being 57 days. Amiodarone is extensively concentrated in tissues such as the skin, liver, lung, eyes, adipose tissue, muscle (including increased concentration within the myocardium) and thyroid gland, explaining its organ-specific adverse effects (Table 1) [Papiris et al. 2010; Van Erven and Schalij, 2010; Vassallo and Trohman, 2007; Bongard et al. 2006; Shukla et al. 1994].

Table 1.

Amiodarone and its side effects.

System	Major side effects
Endocrine-Thyroid (≤35%)	Abnormal thyroid function tests, hypothyroidism and hyperthyroidism
Cardiovascular (<10%)	Bradycardia (5%), QT prolongation, heart block, and rarely (<1%) torsades de pointes
Respiratory (1–17%)	Lipoid pneumonia (usually asymptomatic), amiodarone toxicity manifesting as pneumonitis (0.1–1.6%) (chronic cough, breathlessness and interstitial infiltrates on chest X-ray), frank pulmonary fibrosis, and very rarely adult respiratory distress syndrome (ARDS)
Skin (≤75%)	Photosensitivity (25–75%), slate grey pigmentation (<10%), alopecia (<10%)
Hepatic (≤30%)	Abnormalities in liver function tests (<30%), hepatitis and cirrhosis (<3%)
Neurological (3–35%)	Tremor, gait problems and cognitive impairment. Sensorimotor polyneuropathy (≤1%) with distal predominance, optic neuritis and neuropathy (≤1%)
Eyes (≤100%)	Corneal microdeposits (100%) — reversible on stopping the medication. Optic neuropathy and optic neuritis

Dosing and clinical efficacy of amiodarone

Given its high volume of distribution, a loading dose of about 10 g is needed over 1–2 weeks. Thereafter a standard daily dose of 200 mg orally is usually sufficient. However, the lowest possible dose that is still clinically effective is recommended in order to reduce the likelihood of serious side effects [Van Erven and Schalij, 2010]. Amiodarone has widespread clinical applicability as an antiarrhythmic. It is probably the most effective agent at maintaining sinus rhythm in patients with both paroxysmal and persistent atrial fibrillation (AF) and with both preserved and impaired left ventricular systolic function [Singh, 2008; Vassallo and Trohman, 2007]. Amiodarone is also highly effective at reducing the burden of AF associated with cardiac surgery [Budeus et al. 2006]. Amiodarone can be used in the acute management of sustained ventricular tachyarrythmias, regardless of hemodynamic stability, and as an adjunct to implantable cardioverter-defibrillator therapy to reduce the burden of shocks [Vassallo and Trohman, 2007].

Normal synthesis of thyroid hormones

Iodine is one of the principle substrates for thyroid hormone synthesis. It is actively transported into the thyroid gland where it is incorporated into the tyrosine residues within the molecule thyroglobulin to form the precursors of both thyroxine (T4) and triiodothyronine (T3), which are enzymatically cleaved and released into the circulation [Cavalieri, 1997]. All of the steps in thyroid hormone biosynthesis, from oxidation and organification of iodide to the secretion of T4 and T3 into the circulation, are stimulated by thyroid-stimulating hormone (TSH) and inhibited by excess iodine. This autoregulation of iodine prevents hyperthyroidism in normal people who are exposed to high iodine load such as that found in radioactive contrast, an effect known as the Wolff—Chaikoff effect. Patients with underlying thyroid disease however fail to benefit from this effect and can develop hyperthyroidism [Stanbury et al. 1988; Braverman et al. 1971].

Pathophysiology of amiodarone-related thyroid dysfunction

The majority of the adverse effects of amiodarone on various organs are caused by the deposition of the drug in the parenchyma. However, its effects on the thyroid gland and its metabolism are both unique and quite complex and occur via a number of differing mechanisms. These can be divided into iodine-induced effects (related to the large iodine load of amiodarone) and those due to the intrinsic properties of amiodarone [Stanbury et al. 1988].

Intrinsic amiodarone-related drug effects

Amiodarone inhibits the monodeiodination (5-deiodinase activity) of T4. This leads to a decrease in the generation of T3 from T4, a decrease in the clearance of reverse T3 (rT3) and consequently increased rT3 accumulation [Harjai and Licata, 1996]. Amiodarone can lead to inhibition of T4 and T3 entry into the peripheral tissues. Both amiodarone and its principle metabolite may have direct cytotoxic effects on the thyroid follicular cells, leading to a destructive thyroiditis [Bogazzi et al. 2003; Chiovato et al. 1994]. Furthermore, DEA is a noncompetitive inhibitor of the binding of thyroid hormone (T3) to the β1-thyroid hormone receptor (T3R) [Vanbeeren et al. 1995].

Iodine-related effects of amiodarone on thyroid function

Normal thyroid autoregulation is lost because of the relatively high iodine content in amiodarone. This tends to occur in patients with underlying Hashimoto's disease [Braverman et al. 1971]. In addition there may be iodine-related potentiation of thyroid autoimmunity and unregulated thyroid hormone synthesis in patients with underlying Graves' disease (Jod-Basedow effect).

General thyroid hormonal effects of amiodarone in euthyroid patients

Because of the exposure of increased iodine levels from amiodarone, alterations in thyroid iodine handling occur in order to maintain normal thyroid function. TSH is the first hormone level to change, rising within 48 h and increasing up to an average of 2.7 times the normal levels by the 10th day [Basaria and Cooper, 2005]. There is also an early and measureable rise in serum T4, rT3 and free T4, which peak after 10 weeks of treatment [Melmed et al. 1981]. Conversely, there is a reciprocal and early fall in serum T3 levels over a similar time course. By 3 months a steady state is generally reached with serum levels of total and free T4 and rT3 remaining at the upper end of the normal range or marginally elevated. T3 levels tend to remain at the lower end of the normal range after 3 months, with a fall in TSH to the upper end of the normal range.

Frequency of amiodarone-related thyroid dysfunction

The majority of patients (>70%) on amiodarone will remain euthyroid. However, treatment may lead to either amiodarone-induced hypothyroidism (AIH) or amiodarone-induced thyrotoxicosis (AIT), with AIH more common in iodine-sufficient populations and AIT in iodine-deficient populations. Generally, in the Western world the prevalence of AIH ranges from 5 to 22% whilst that of AIT is somewhat lower, affecting 2.0–9.6% [Batcher et al. 2007; Basaria and Cooper, 2005; Shukla et al. 1994; Trip et al. 1991; Albert et al. 1987; Amico et al. 1984; Martino et al. 1984].

Amiodarone-induced hypothyroidism

Epidemiology and pathology

AIH is more frequent in iodine-sufficient populations [Martino et al. 1987b, 1984]. In the SAFE-Trial subclinical hypothyroidism, defined as a TSH of 4.5–10 mU/l with normal thyroid hormone levels [Nademanee et al. 1986] was detected in 25.8% of patients taking amiodarone, while overt hypothyroidism (TSH > 10 mU/l) occurred in 5% [Batcher et al. 2007]. There does not appear to be a clear association between the daily or cumulative doses of amiodarone and the development of AIH [Bouvy et al. 2002; Trip et al. 1991]. Hashimoto's thyroiditis is the most common risk factor for the development and persistence of AIH and is the likely reason for the female preponderance (female to male ratio 1.5: 1) [Trip et al. 1991]. Although AIH can occur in normal thyroid glands, underlying thyroid abnormalities may be detected in up to 68% of patients [Martino et al. 1987a]. Moreover, amiodarone may accelerate the natural course of Hashimoto's thyroiditis via iodine-induced thyroid cellular damage [Martino et al. 1987b]. The combination of being female and the presence of thyroid peroxidase or thyroglobulin antibodies constitutes a relative risk of 13.5 for the development of AIH [Newman et al. 1998; Trip et al. 1991]. AIH may however spontaneously remit, especially in the absence of autoimmune thyroid disease [Hawthorne et al. 1985; Sanmarti et al. 1984].

Clinical features

The symptoms of AIH are identical to that of primary hypothyroidism and include lethargy, weakness, intolerance to cold, mental sluggishness, constipation, menorrhagia and dry skin. Hypothyroidism may develop as soon as 2 weeks or as late as 39 months after the initiation of amiodarone treatment [Nademanee, 1989].

Diagnosis and investigations

Although fluctuations in TSH are common after starting amiodarone, the diagnosis of hypothyroidism is usually straightforward and is confirmed by the finding of a persistently raised TSH concentration (> 10 mU/l), in combination with a low—normal or low FT4 [Wiersinga and Trip, 1986]. Low T3 or FT3 concentrations are a less reliable indicator of hypothyroidism as they may occur in euthyroid patients during amiodarone treatment since amiodarone reduces peripheral T4 to T3 conversion [Newman et al. 1998].

Management

Although many patients without pre-existing thyroid disease will become euthyroid within 2–4 months of stopping amiodarone treatment, it is not clinically appropriate to withdraw such a critical drug if it has been of benefit to the underlying arrhythmia. Permanent hypothyroidism requiring T4 replacement is more common in patients with thyroid antibodies [Newman et al. 1998; Martino et al. 1987b]. In such patients if TSH is raised, treatment with T4 should be started without delay [Siddoway, 2003]. The aim of treatment should be to normalize TSH and relieve symptoms. It is important to appreciate that because of the action of amiodarone on thyroid hormone metabolism, higher doses of T4 may be required than for primary hypothyroidism [Albert et al. 1987].

If stopping amiodarone is considered appropriate (possibly because of a lack of beneficial effect on the underlying arrhythmia), then spontaneous remission of hypothyroidism often occurs within 3–4 months in patients without underlying Hashimoto's disease, while those with underlying thyroid disease usually require T4 [Weetman, 1997; Martino et al. 1987b].

Amiodarone-induced thyrotoxicosis

Epidemiology

Unlike AIH, which tends to occur in iodine-sufficient areas, AIT has been more frequently identified in areas of iodine deficiency. Also, in contrast to AIH, which is more preponderant in females (probably because underlying Hashimoto's disease is more common in females), AIT has a higher male to female incidence of 3: 1 [Cohen-Lehman et al. 2010]. AIT usually develops after months of amiodarone treatment, although it can develop in the first few weeks of treatment [Newnham et al. 1988]. Furthermore, because of the prolonged half life of amiodarone, AIT can develop several months after amiodarone withdrawal [Martino et al. 1987a]. Two types of AIT have been described (Table 2). Excess iodine-induced thyroid hormone synthesis is known as type I AIT, whereas destruction of thyroid follicles resulting in a thyroiditis with excess release of T3 and T4 is known as type II AIT.

Table 2.

Pathogenesis and clinical features of amiodarone-induced thyrotoxicosis.

	Type I AIT	Type II AIT
Underlying thyroid abnormality	Yes	No
Pathogenetic mechanism	Excessive hormone synthesis due to iodine excess	Excessive release of preformed hormones due to thyroid destruction
Goitre	Multinodular or diffuse goitre normally present	Occasionally small, diffuse, firm, sometimes tender
Thyroidal radioiodine uptake	Normal/raised (can also be low due to diluting effects of excess iodine)	Low/absent
Serum interluekin-6	Normal/slightly raised	Profoundly raised
Thyroid ultrasound	Nodular, hypoechoic, increased volume	Normal
Colour flow Doppler sonography	High vascularity	Absent vascularity

AIT, amiodarone-induced thyrotoxicosis.

Type I amiodarone-induced thyrotoxicosis

Type I AIT usually occurs in patients with underlying thyroid pathology, such as latent Graves' disease or nodular goitre. In these patients, the sudden iodide load, associated with amiodarone treatment, accelerates thyroid hormone synthesis sufficiently to induce thyrotoxicosis because of increased thyroid hormone production in autonomous thyroid tissue (Jod—Basedow phenomenon) [Newman et al. 1998].

Type II amiodarone-induced thyrotoxicosis

Type II AIT occurs in normal thyroid glands and is the result of a direct toxic effect of amiodarone itself leading to a subacute and destructive thyroiditis. Consequently, there is continuous spillage of thyroid hormones into the blood stream [Bogazzi et al. 1997; Chiovato et al. 1994]. Microscopic histopathological examination of the thyroid gland confirms an inflammatory response as evidenced by infiltration by histiocytes, follicular swelling and fibrosis [Meurisse et al. 1993; Mulligan et al. 1993; Leung et al. 1989; Smyrk et al. 1987].

It can often be difficult to differentiate between these two types of AIT. Indeed, both conditions may coexist in a person, producing a significant diagnostic and treatment challenge. The relative distinguishing features of type I versus type II AIT are shown in Table 2.

Clinical features of amiodarone-induced thyrotoxicosis

The clinical features of AITare similar to those of thyrotoxicosis of any aetiology and include weight loss, heat intolerance, fatigue, muscle weakness, increased frequency of stools, oligomenorrhoea, nervousness, anxiety and palpitations. However, some of these symptoms may be masked due to the underlying β-blocking effect of amiodarone. Goitre (nontender) may be present in patients with underlying thyroid disease such as Graves' disease or multinodular goitre. Type II AIT may be associated with a small and tender goitre and usually has an explosive onset [Newman et al. 1998].

Investigation of amiodarone-induced thyrotoxicosis

Suppressed TSH with elevated FT4 and FT3 is seen in both types of AIT and are not discriminatory between the two. Radioiodine uptake (RAIU) can be normal or raised in type I AIT because of avid uptake by autonomous thyroid tissue (although is usually low in type I AIT due to the diluting effects of large amounts of iodine in the bloodstream released from the amiodarone), but is very low or absent in type II AIT because of destruction of or damage to thyroid tissue [Newman et al. 1998]. Inflammatory markers such as interleukin-6 are markedly raised in type II AIT but may be only mildly elevated in type I AIT and are not generally clinically useful [Bartalena et al. 1994]. Serum thyroglobulins are elevated in both types but markedly elevated in type II. Thyroid ultrasound scanning often reveals an increased thyroid volume, a hypoechoic pattern and nodular lesions in type I AIT but is normal in type II AIT. Colour flow Doppler sonography may reveal high vascularity (indicating hyperfunctioning gland) in type I AIT or absent vascularity (damage to thyroid gland) in type II AIT (Table 2) [Bogazzi et al. 2003; Eaton et al. 2002].

Management of amiodarone-induced thyrotoxicosis

Distinguishing between the two types of AIT is important because it has a major influence on subsequent management (Table 3). Unfortunately there is no large prospective, randomized study to help us in the optimal management of AIT [Martino et al. 1986]. AIT has a threefold increased rate of major adverse cardiovascular events compared with euthyroid patients, hence prompt instigation of treatment is important [Yiu et al. 2009; Conen et al. 2007; Leung et al. 2002].

Table 3.

Treatment of amiodarone-induced thyroid disease dysfunction.

Amiodarone-induced hypothyroidism

Clinical hypothyroidism TSH > 10 mU/l, FT4 low normal or decreased and clinical features of hypothyroidism Treat with levothyroxine and increase the dose until TSH is within the normal range and symptoms have resolved. Amiodarone can be continued in the majority of patients

Subclinical hypothyroidism TSH > 4.5 and ≤ 10 mU/l and normal FT4 with absence of clinical features of hypothyroidism If TPO antibody is positive, consider treating with levothyroxine If antibody is negative and symptoms are attributable to hypothyroidism, trial of T4 and reassessed for symptomatic improvement If asymptomatic and no antibodies, follow up at 6 weeks and 3 months thereafter

Amiodarone-induced thyrotoxicosis Suppressed TSH < 0.1 with raised FT4 and FT3 Management of AIT depends upon type of AIT, severity and duration of clinical features

Type I AIT If possible, withdraw amiodarone Carbimazole (or methimazole), 30–60 mg/day. Definitive treatment after restoration of euthyroidism is either by radioiodine or thyroidectomy.

Type II AIT Prednisolone 40–60 mg/day for 2–3 months Withdraw amiodarone if possible

Mixed AIT Carbimazole (or methimazole) 30–60 mg/day with prednisolone 40–60 mg (pragmatically consider ‘40 and 40’ i.e. 40mg carbimazole (or methimazole) and 40 mg prednisolone) Rapid response suggests type II AIT, hence consider stopping carbimazole (or methimazole). Poor response may indicate type I AIT: taper steroids and continue carbimazole (or methimazole). Consider reducing carbimazole (or methimazole) if clinically indicated

Early or subclinical thyrotoxicosis TSH 0.1 −0.45 with normal FT4 and FT3 (usually at lower end of reference range) Repeat thyroid function tests in 6 weeks, if TSH is suppressed, treat as thyrotoxicosis. If TSH is normal, follow up as normal.

AIT, amiodarone-induced thyrotoxicosis; TPO, thyroid peroxidase; TSH, thyroid-stimulating hormone.

The decision regarding the continuation or otherwise of amiodarone is a complex one with no absolute answer. Consideration must be given to the benefit of amiodarone on life-threatening arrhythmias, the fact it has such a long half life (and hence no immediate benefit on thyroid status if stopped) and the fact that amiodarone reduces T4 to T3 conversion so an initial exacerbation of thyroid symptoms may occur on its cessation [Leger et al. 1994; Mulligan et al. 1993]. If amiodarone can be substituted with other modes of therapy, such as other antiarrhythmic drugs or an implantable defibrillator, then stopping amiodarone may be a viable option.

The final management plan must be discussed jointly by the patient, the cardiologist, the primary care physician, and the endocrinologist [Osman et al. 2002]. An endocrinology opinion is recommended in all patients [Bogazzi et al. 2001]. It is important to remember that successful treatment of AIT can be achieved with antithyroid medications while the patient remains on amiodarone [Trip et al. 1994]. It is also relevant that DEA binds to intracellular T3 receptors and acts as a T3 antagonist, hence continuing amiodarone and treating with antithyroid medications is a feasible option in some patients [Vanbeeren et al. 1995].

Type I amiodarone-induced thyrotoxicosis

Most patients with type I AITare still hyperthyroid 6–9 months after stopping amiodarone [Martino et al. 1987a]. Hence, management of type I AIT in these patients is based on the use of thiourea derivatives, particularly carbimazole (CBZ) or methimazole (MMI), to block hormone synthesis while amiodarone and its associated iodine are cleared from the body [Newnham et al. 1998; Bartalena et al. 1996; Harjai and Licata, 1996; Albert et al. 1987]. Thionamides block further organification of iodine and synthesis of thyroid hormones [Bouvy et al. 2002]. In amiodarone-treated patients the thyroid gland usually contains huge amounts of iodine and hence a very high dose of CBZ (or MMI) up to 60 mg/day may be required.

Propylthiouracil (PTU) was once favoured because it inhibits peripheral ′deiodinase activity. However, the US Food and Drug Administration (FDA) have recently released an advisory on PTU for its liver toxicity potential [Food and Drug Administration, 2010]. Many experts now recommend using either CBZ (or MMI) in all thyrotoxic patients (unless other compelling reasons exist for using PTU such as pregnancy). Patients on CBZ (or MMI) should be cautioned about the life-threatening side effect of bone marrow suppression. Patients should go to their primary care provider or to hospital if they develop a sore throat or fever, or they become systemically unwell. A full blood count should be carried out to test for agranulocytosis. Potassium perchlorate has been used as an adjunct in patients who are given CBZ (or MMI) to a maximum of 5 g/day. Usually 1 g/day is sufficient in most patients. Potassium perchlorate reduces the intrathyroidal iodine stores because it decreases the entry of iodine into the thyroid and competitively inhibits thyroid iodine uptake [Basaria and Cooper, 2005].

As most patients with type I AIT have underlying Graves' disease or toxic multinodular goitre, thyrotoxicosis usually recurs [Newnham et al. 1988]. Definitive treatment, either radioiodine or surgery, is therefore recommended. The timing of this depends on the severity of thyrotoxicosis, the response to antithyroid drugs, the radioiodine uptake level, and the policy of the supervising endocrinologist [Newman et al. 1998; Vanderpump et al. 1996]. The iodine in amiodarone inhibits the uptake of radioactive iodine and hence radioactive iodine can be used only in patients with high radioactive iodine uptake. If amiodarone treatment is an absolute necessity (even after its withdrawal), ablation of the thyroid with radioiodine, before resumption of the drug, is a sensible option [Hermida et al. 2004; Martino et al. 2001]. When thyrotoxicosis is uncontrollable, despite the use of antithyroid drugs and perchlorate, thyroidectomy should be considered. Surgery may be necessary even in the early stages of the disease, particularly during thyroid storm when immediate control of thyrotoxicosis is necessary and could be life saving [Klein et al. 1997; Meurisse et al. 1993].

Type II amiodarone-induced thyrotoxicosis

If amiodarone is stopped in patients with type II AIT the majority will become and remain euthyroid within 3–5 months of amiodarone withdrawal [Bartalena et al. 1996; Martino et al. 1987a]. Some may eventually become hypothyroid either spontaneously or after re-exposure to iodine [Bogazzi et al. 2001]. In patients in whom amiodarone is continued, prednisolone 40–60 mg/day should be used, usually with a good clinical effect starting within the first week or two. The dose of prednisolone is gradually tapered once control has been achieved, usually after 2–3 months. Follow up of these patients is vital because disease may recur, which should be promptly treated [Bartalena et al. 1996; Harjai and Licata, 1996; Albert et al. 1987]. In most cases, biochemical and clinical resolution of thyrotoxicosis begins within days of commencing steroids and is complete within the first month of treatment [Bartalena et al. 1996; Roti et al. 1993]. If thyrotoxicosis is mild with no cardiac symptoms, amiodarone need not be stopped when steroids are given concomitantly, but regular follow up of these patients is important.

Steroids, either alone or in combination with anti-thyroid drugs or occasionally plasmapheresis, are generally favourable in treating type II AIT [Rao et al. 1986]. Iopanoic acid is an iodinated oral cholecystographic agent that inhibits type II deiodinase activity and is occasionally used in the short term in some thyrotoxic conditions (if available) [Bogazzi et al. 2002]. However, because of the escape phenomenon when persistent use leads to a recurrence of thyrotoxicosis, long-term treatment with this drug is not feasible. Prednisolone is more effective than iopanoic acid but the latter may be useful if type II AIT needs to be rapidly controlled, perhaps preceding thyroidectomy [Bogazzi et al. 2002].

Mixed type amiodarone-induced thyrotoxicosis

In mixed type AIT or if diagnosis is uncertain, it is reasonable to start therapy with CBZ (or MMI) 40–60 mg/day and prednisolone 40–60 mg/day (pragmatically start with ‘40 and 40’, i.e. 40 mg CBZ [or MMI] and 40 mg prednisolone). If there is rapid response, it may be suggestive of type II AITand consideration can be given to stopping CBZ (or MMI) in these patients. A poor response may indicate type I AIT. Gradual tapering of the dose of prednisolone and continuation of CBZ (or MMI) may then be necessary. Close follow up of these patients is required because further adjustment of carbimazole may be necessary depending on clinical response to this medication. Perchlorate, radioactive iodine or surgery should be considered if the response is not desirable [Osman et al. 2002; Brennan et al. 1987].

Monitoring patients on amiodarone

There has been no clear consensus regarding the frequency of screening for amiodarone-induced thyroid dysfunction in patients on chronic amiodarone treatment. Given the time course of the onset of dysfunction and its unpredictability from baseline patient demographics, thyroid function tests (T4 and T3) should be checked every 6 months or sooner if clinically indicated [Pazin-Filho et al. 2009]. Furthermore, before initiating therapy with amiodarone a complete thyroid examination should be performed along with baseline measurements of serum TSH, free T4, T3, and anti-thyroid peroxidase (TPO) antibodies. This baseline evaluation will not only detect existing underlying thyroid dysfunction but may also help identify patients who may be predisposed to developing thyroid dysfunction while on amiodarone [Basarai and Cooper, 2005].

Dronedarone

Pharmacology and pharmacokinetics of dronedarone

Dronedarone (marketed as Multaq) is a new anti-arrhythmic drug which was designed to maintain the potent antiarrhythmic and rate-controlling effects of amiodarone whilst reducing its toxic effects. Dronedarone is a noniodinated benzofuran derivative of amiodarone in which the iodine moieties, observed with amiodarone, are substituted with a methyosulfonamide group (Figure 1(d)). Hence, dronedarone is less lipophilic than amiodarone, with a much shorter half life (24 h) than amiodarone (several weeks) [Singh et al. 2010]. Dronedarone is administered in a simple twice daily (400 mg) oral regime, without the need for loading as with amiodarone.

As with amiodarone, dronedarone is also extensively metabolized primarily by the cytochrome P450 3A4 system and excreted in the bile with minimal renal excretion [Singh et al. 2010; FDA, 2009]. However, akin to its parent compound amiodarone, 10–15% transient increase in serum creatinine can be seen with dronedarone and this relates to the inhibition of tubular secretion of creatinine by the drug and not to a decrease in the glomerular filtration rate [FDA, 2009; Tschuppert et al. 2007]. Dronedarone does not appear to cause any of the thyroid, pulmonary and neurological adverse effects observed with amiodarone. However, it does cause more gastrointestinal side effects (nausea, vomiting and abdominal pain) [FDA, 2009].

Clinical efficacy of dronedarone

Dronedarone has been the subject of seven randomized controlled phase II/III clinical trials assessing its clinical efficacy in over 7000 patients (see Singh et al. 2010 for review). It has proven to be superior to placebo in terms of rate control among patients with AF and in the prevention of AF recurrence following cardioversion. However, in the only head-to-head comparison with amiodarone, the recently completed DIONYSOS trial (Efficacy and Safety of Dronedarone Versus Amiodarone for the Maintenance of Sinus Rhythm in Patients with Atrial Fibrillation) of 504 patients demonstrated the superiority of amiodarone over dronedarone in preventing AF recurrence. In addition, in the ANDROMEDA study (Antiarrhythmic Trial With Dronedarone in Moderate-to-Severe Congestive Heart Failure Evaluating Morbidity Decrease) dronedarone was linked to higher mortality than placebo amongst 650 patients with symptomatic heart failure and AF history randomized to treatment with dronedarone versus placebo [Kober et al. 2008]. However, in the recent ATHENA trial of 4628 patients with a history of nonpermanent AF, dronedarone was significantly more effective than placebo in reducing the composite endpoint of first hospitalization caused by cardiovascular events or death (not death alone) [Hohnloser et al. 2009]. The mechanisms of this benefit (because of rate control, antiarrhythmic or other effects) remain the subject of debate. Nevertheless the FDA approved dronedarone on 2 July 2009 and the National Institute for Health and Clinical Excellence (NICE) in 2010 [NICE, 2010]. Dronedarone should be avoided in patients with symptomatic heart failure and/or ejection fraction less than 35%.

Conclusions

Amiodarone continues to be a very useful and extensively utilized antiarrhythmic drug. Its superior potency and tolerability among patients with significant left ventricular systolic dysfunction, compared with its congener dronedarone, suggests that it will continue to remain at the forefront of antiarrhythmic therapy for the foreseeable future. Concern about its effect on thyroid function should not necessarily preclude its use.

A common-sense approach to thyroid status assessment before starting amiodarone and subsequent monitoring of thyroid function will allow identification of patients at increased risk of thyroid dysfunction and facilitate early detection and treatment. In the event of thyroid dysfunction on amiodarone, the decision of whether to continue or stop amiodarone remains complex, particularly in cases of AIT. Due consideration must be given to the benefit of amiodarone in life-threatening arrhythmias and that its cessation may initially aggravate thyroid dysfunction at worst and have no effect for months at best. The development of clinically and biochemically significant AIH is relatively simple to treat with T4. However, the development of AIT remains a diagnostic and therapeutic challenge and in some cases the empirical use of combined CBZ (or MMI) (40–60 mg) and prednisolone (40–60 mg) may be needed (pragmatically start with ‘40 and 40’, i.e. 40 mg CBZ [or MMI] and 40 mg prednisolone).