K. Harjai, A. Licata
1997
Citations
7
Influential Citations
217
Citations
Quality indicators
Journal
Annals of Internal Medicine
Abstract
Amiodarone was approved by the Food and Drug Administration in 1985 for the treatment of serious ventricular arrhythmia. It is also efficacious in the treatment of paroxysmal supraventricular tachycardia and atrial fibrillation and flutter [1]. In addition, use of amiodarone after myocardial infarction may reduce complex ventricular ectopy and cardiac-related mortality [2, 3]. Use of amiodarone may improve survival rates in patients with heart failure [4-6]. However, in view of the results of recent studies, the efficacy of amiodarone in improving survival after myocardial infarction and in patients with heart failure has been questioned ([7, 8]; Camm JA, for the European Myocardial Infarction Amiodarone Trial, paper presented at the American College of Cardiology 1996 Meeting, Orlando, Florida). Side effects of amiodarone are often related to daily or cumulative dose and duration of treatment and include corneal microdeposits, photosensitivity, cutaneous hyperpigmentation, pulmonary toxicity, hepatotoxicity, peripheral neuropathy, drug interactions, hyperthyroidism, and hypothyroidism [9-23]. Smaller doses of amiodarone, such as those used for supraventricular arrhythmias [1], may be associated with fewer side effects. Using a MEDLINE search of articles published from 1975 to 1995, we identified and reviewed English-language articles on the effects of amiodarone on thyroid physiology and the recognition and management of amiodarone-induced thyrotoxicosis and hypothyroidism. Effects of Amiodarone on Thyroid Physiology Amiodarone is an iodine-rich benzofuran derivative (Figure 1). Approximately 37% of amiodarone (by weight) is organic iodine; 10% of the latter is deiodinated to yield free iodine. A maintenance dose of 200 to 600 mg/d results in a daily intake of organic iodide of 75 to 225 mg, at least 10% of which is deiodinated. Because the normal dietary requirement of iodine is only 0.2 to 0.8 mg/d [24], the increased amount of iodine intake associated with amiodarone causes a massive expansion of the iodide pool [25]. In patients treated with amiodarone, urinary and plasma levels of inorganic iodide increase 40-fold, whereas thyroid iodide uptake and clearance decrease significantly [25]. Therefore, thyroid hormone dynamics change in almost all patients receiving amiodarone [26]. Figure 1. Chemical structures of thyroxine, triiodothyronine, and amiodarone. Amiodarone has many effects on thyroid physiology. It decreases the peripheral deiodination of thyroxine to triiodothyronine by inhibiting type I iodothyronine 5-deiodinase [27-30], resulting in an increase of serum levels of thyroxine and reverse triiodothyronine and a decrease of serum levels of triiodothyronine (by 20% to 25%), as seen in the euthyroid sick syndrome [31, 32]. Amiodarone also inhibits entry of thyroxine and triiodothyronine into peripheral tissue. Serum thyroxine levels increase by an average of 40% above pretreatment levels after 1 to 4 months of treatment with amiodarone; in 40% of all patients, the serum thyroxine levels (and free thyroxine index) may increase to levels above the normal range. This is an expected finding and in itself does not constitute evidence of hyperthyroidism [23]. In addition, an increase in thyroid-stimulating hormone levels secondary to inhibition of thyroxine-triiodothyronine deiodination in the pituitary is seen during the early phase of treatment (from 1 to 3 months) [33]. This inhibition is a crucial step in the feedback regulation of secretion of thyroid-stimulating hormone [34]. By themselves, elevated serum levels of thyroid-stimulating hormone are not an indication for thyroxine replacement therapy in these patients. With long-term administration of amiodarone (>3 months), serum levels of thyroid-stimulating hormone often return to normal, and the response of thyroid-stimulating hormone to thyrotropin-releasing hormone may be reduced [33, 35-38]. Changes in thyroid function test results, which occur in euthyroid patients receiving amiodarone, are summarized in Figure 2. Figure 2. Changes in thyroid hormone physiology and thyroid function test results in euthyroid patients who received amiodarone. Abnormal results of thyroid function tests (without overt dysfunction of the thyroid gland) occur more often as the duration of treatment increases and doses accumulate. Serum levels of amiodarone or desethylamiodarone generally do not predict these abnormal test results [39]. One exception is the increase in reverse triiodothyronine levels in the first 2 weeks after commencement of amiodarone therapy; this shows a direct correlation with serum amiodarone levels [40]. In addition, in the absence of factors that may independently affect reverse triiodothyronine metabolism (such as hyperthyroidism, hypothyroidism, surgery, fasting, systemic illnesses, and concomitant use of corticosteroids or -blockers), the efficacy and toxicity of amiodarone can be monitored by serial measurements of serum reverse triiodothyronine levels [41]. Serum levels of reverse triiodothyronine that are threefold to fivefold greater than baseline levels are associated with adequate antiarrhythmic response; levels that are more than five times the baseline values are associated with a greater chance for drug toxicity. Effects of Amiodarone on Cardiac Tissue Receptors Independent of its effects on thyroid hormone physiology, amiodarone has some electrophysiologic effects on cardiac muscle cells that simulate those of hypothyroidism [42]. The effects seen with long-term administration may be mediated by amiodarone itself, its active metabolite desethylamiodarone, or both. In the hearts of pigs treated with amiodarone, the maximum binding capacity of -receptors and calcium channels is reduced [43]. The maximum binding capacity for triiodothyronine is unchanged, suggesting that no functional reduction in the number of triiodothyronine receptors occurs. However, desethylamiodarone competitively inhibits the binding of triiodothyronine to its nuclear receptors and may be responsible for the local hypothyroid-like effects [43]. In a comparison of rats with normal thyroid function and those that had had thyroidectomy [44], amiodarone reduced cardiac -receptor density and heart rate in the former but not the latter group. This finding implies that a minimum serum thyroid hormone level is necessary for the drug to produce some of its cardiac effects. These changes occur independently of alterations in thyroid secretion and serum triiodothyronine levels. Exogenous triiodothyronine-mediated increase in -receptor density and heart rate is also partly inhibited by amiodarone [45]. These observations suggest that the lowering of -receptor density by amiodarone is related to triiodothyronine antagonism at the cardiac cellular level. Incidence of Clinical Thyroid Dysfunction in Patients Receiving Amiodarone In various studies [10, 35, 46-50], the incidence of amiodarone-induced thyrotoxicosis has been reported to be 1% to 23% and that of hypothyroidism has been reported to be 1% to 32%. As many as 49% of patients in a study in which phenytoin was used as a supplementary antiarrhythmic agent [26] developed thyroid dysfunction within 60 months of follow-up. However, the overall incidence of amiodarone-induced thyroid dysfunction is more reasonably estimated to be 2% to 24% [26]. Amiodarone-induced thyrotoxicosis prevails in areas with low iodine intake, and hypothyroidism is prevalent in areas with high iodine intake. Thus, thyrotoxicosis is more common in Italian than American patients (10% compared with 2%), but hypothyroidism is less common (2% compared with 22%) [35]. This difference is generally similar to the difference in the incidence of iodide-induced thyrotoxicosis, which is more common in iodide-deficient areas than in iodide-replete areas [51]. However, this geographic predilection is not substantiated by all studies [52, 53]. Although amiodarone crosses the placental barrier, its use in nine pregnant women was not associated with clinical thyroid dysfunction in their neonates [54]. Thyroid function test results were normal in all neonates except one who had clearly abnormal serum levels of thyroxine and thyroid-stimulating hormone. In another series of five neonates born to women receiving amiodarone [55], one was found to have hypothyroidism requiring treatment with triiodothyronine for a few weeks. In a review of adverse effects associated with amiodarone therapy in 34 pregnant women [56], hypothyroidism was reported in three neonates (9%) and hyperthyroidism was reported in none. Amiodarone-Induced Thyrotoxicosis The factors leading to the development of thyrotoxicosis in some patients treated with amiodarone are not completely understood. Amiodarone-induced thyrotoxicosis occurs in patients with underlying goiter and those with no apparent thyroid disorder [57]. A predominance among men is sometimes reported [26, 58]. Lack of response of thyroid-stimulating hormone to stimulation of thyrotropin-reducing hormone may predict development of amiodarone-induced thyrotoxicosis [59], but this is not uniformly accepted [26, 60]. Pathogenesis Amiodarone-induced thyrotoxicosis is often caused by excessive synthesis of thyroid hormone induced by iodine, especially in patients with underlying thyroid disease. However, various other mechanisms have been proposed. Disturbance of Thyroid Iodine Autoregulation Intrinsic autoregulatory mechanisms in the thyroid modulate the gland's iodine handling according to its iodine content [61]. Alterations in these mechanisms may cause thyroid dysfunction in the presence of excess iodine [62, 63]. A disturbance of these mechanisms is suggested by the high iodine content of the thyroid in patients with amiodarone-induced thyrotoxicosis (compared with those who have normal thyroid function) while they are receiving amiodarone [64] and by return of iodine content to normal during resolution of thyrotoxicosis [38]. T