| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Special Features |
Division of Medical Sciences (F.O., M.D.G., M.C.S., J.A.F.), Queen Elizabeth Hospital, University of Birmingham, Birmingham, United Kingdom B15 2TH
Address all correspondence and requests for reprints to: Prof. J. A. Franklyn, Department of Medicine, Queen Elizabeth Hospital, Birmingham B15 2TH, United Kingdom. E-mail: . J.A.Franklyn{at}bham.ac.uk
Abstract
Thyrotoxicosis is often perceived as a reversible disorder without long-term consequences, perhaps because of the availability of effective treatments, but recent evidence suggests that there may, in fact, be adverse outcomes. Long-term follow-up studies have revealed increased mortality from cardiovascular and cerebrovascular disease in those with a past history of overt hyperthyroidism treated with radioiodine as well as in those with subclinical hyperthyroidism indicated by a low serum TSH concentration. Thyroid hormones exert direct effects on the myocardium as well as the systemic vasculature predisposing to dysrhythmias, especially supraventricular. Effects of thyroid hormones on the autonomic nervous system may also contribute to arrhythmogenesis. Atrial fibrillation is a recognized complication of hyperthyroidism that predisposes to embolic events. Development of atrial fibrillation, together with other supraventricular dysrhythmias (both clinically obvious and those detected only by Holter monitoring) in those with hyperthyroidism may account for increased vascular mortality. Improved detection of supraventricular dysrhythmias and therapeutic intervention (e.g. anticoagulants, antiarrhythmics) may improve the long-term vascular prognosis, but their role remains to be established in large therapeutic trials.
THYROTOXICOSIS IS A common disorder with a prevalence of 3% in females and 0.3% in males in iodine-replete areas such as the United Kingdom and the United States (1). It is known to induce many cardiovascular effects such as sinus tachycardia, systolic hypertension, changes in ventricular systolic and diastolic function, and predisposition to dysrhythmias, especially atrial fibrillation (AF) (2). The availability of effective treatments for thyrotoxicosis has led to the widespread perception that it is a reversible disorder without long term consequences, but increasing evidence suggests that this is not the case, especially in terms of vascular disease.
Surprisingly, there have been few population-based studies examining the long-term influence of thyroid disease and its treatment on morbidity and mortality. Our recent study (3) of a cohort of 7209 subjects with thyrotoxicosis treated with radioiodine between 1950 and 1989 identified marked excess in mortality from all causes. The underlying cause of death (coded to ICD-9) for the cohort was compared with age-specific mortality data for England and Wales and standardized mortality ratio (SMR) used as a measure of relative risk. The excess mortality was largely accounted for by an excess of deaths because of circulatory diseases, both cardiovascular (SMR 1.2, 95% confidence intervals 1.21.3, P < 0.001) and cerebrovascular (SMR 1.4, 95% confidence intervals 1.21.5, P < 0.001). Increases in risk of death because of rheumatic and hypertensive heart disease were found. The observed increase in relative risk of death because of ischemic heart disease was less striking (although significant), and the absolute risk of death from this cause was high, determining that many deaths in the cohort were attributed to ischemic heart disease (SMR of 1.1, 95% confidence interval 1.01.1, P = 0.03). Deaths secondary to dysrhythmias and congestive cardiac failure (ICD-9 category "other" circulatory diseases) were also significantly increased. In addition to our own study, excess vascular mortality has been reported in a study of 1,762 women with thyrotoxicosis treated with radioiodine, followed for an average of 17.2 yr (4), and another study of 10,552 hyperthyroid patients treated with radioiodine, followed for an average of 15 yr (5). Interestingly, a recent study of patients admitted to the hospital with an acute medical problem revealed that the finding of an elevated serum-free T3 concentration was associated with a 2.6-fold greater likelihood of the presence of a coronary event (6).
Although these findings together suggest a link between overt thyroid hormone excess and occurrence of vascular disease, we have recently described increased mortality from circulatory diseases (and specifically from cardiovascular diseases) in a community-based study of subjects with subclinical hyperthyroidism (not on thyroid hormone replacement) followed over a 10-yr period (7).
It is likely that dysrhythmias contributed to the excess mortality ascribed to cardiovascular and cerebrovascular diseases in these cohort studies, especially in those with the complication of AF, in whom predisposition to embolic events is well described (8).
Arrhythmic effects of thyroid hormones on the myocardium
Overt hyperthyroidism and atrial fibrillation.
In the adult population, AF is the most common cardiac rhythm disturbance and after sinus tachycardia is the most prevalent dysrhythmia in those with hyperthyroidism. Between 10% and 15% of hyperthyroid patients develop AF (9); this incidence increases with age irrespective of whether there is associated heart disease. AF is well known to be an independent risk factor for cerebrovascular events. Two studies have investigated the rate of embolism observed in thyrotoxicosis, one of which revealed a higher embolic rate in those with AF than those in sinus rhythm (8, 10) (Table 1
).
|
Subclinical hyperthyroidism is common in the community, with the prevalence in iodine replete areas reported to range from 0.5% to 3.9% in adults of all ages (11) and 11.8% in one study of the elderly (12); the prevalence may be higher in areas of iodine deficiency. It is defined as a low-serum TSH concentration in an asymptomatic subject with normal serum T3 and T4 concentrations (9). The most common cause in the general population is the ingestion of exogenous T4 as replacement or suppressive therapy. A low-serum TSH concentration is generally a sensitive marker of thyroid hormone excess and has been reported in a large population-based study to be associated with a 3-fold higher risk of developing AF in the subsequent decade (9). This study followed up 2007 patients aged 60 yr or more (from the Framingham Heart Study) for a 10-yr period. These patients did not have AF at the start of the study and were classified according to their serum TSH level. During the 10-yr follow-up period, 192 (10%) developed AF. The cumulative incidence of AF at 10 yr in subjects with a low TSH concentration (
0.1 mU/liter) was 28%, compared with 11% in those with a normal TSH (P = 0.005). The relative risk for developing AF in those with low TSH was 3.1 (95% confidence intervals 1.7 to 5.5), compared with those with normal TSH (P < 0.001). The incidences of AF in those with a slightly low TSH and a high TSH concentration were 16% and 15%, respectively; these incidences were not significantly different when compared with those in the normal TSH group (Fig. 1
). There are no long-term follow-up mortality or morbidity data on this cohort yet.
|
Other atrial arrhythmias and hyperthyroidism
Supraventricular premature complexes (SVCs) are known to initiate AF, particularly those originating in the pulmonary veins (15), and these have been reported to be more frequent in thyrotoxic patients than in a matched control group (16). The number of patients with supraventricular tachycardia (defined as 10 SVCs in a row, heart rate >130 beats/min) has been reported to decrease after antithyroid therapy, with the prevalence of supraventricular arrhythmias being greater in older patients both before and during therapy (16). Our own preliminary data (17) support this view in that we found the prevalence of significant atrial ectopic beats (defined as >240 SVCs per 24 h) to be higher in thyrotoxic subjects at presentation to clinic, compared with matched controls, and remained elevated 3 months after beginning antithyroid therapy despite restoration of biochemical euthyroidism, suggesting continued arrhythmic substrate.
Effects of thyroid hormones on the ventricles
In contrast to supraventricular arrhythmias, ventricular arrhythmias are uncommon in thyrotoxicosis and are found with a frequency similar to that in the normal population (16, 17). Furthermore, the prevalence of ventricular arrhythmias in thyrotoxic subjects remains unchanged during and after antithyroid therapy (16). Ventricular tachycardia and ventricular fibrillation are exceptional in those with thyrotoxicosis and usually occur only in those with marked heart failure or associated cardiac disease, typically because of ischemic heart disease (18).
Use of ß-adrenoceptor blocker therapy
ß-Adrenoceptor blockers are widely used in the management of patients with thyrotoxicosis, typically in short-term management before euthyroidism is achieved. Such drugs have a well-established role in management of symptoms, including palpitation (19). Whether they have a specific role in dysrhythmia prophylaxis in thyrotoxicosis has not been evaluated formally; likewise, the role of ß-adrenoceptor blockers in subclinical hyperthyroidism is unclear. Some authors have advocated use of ß-adrenoceptor blockers in this condition but only in those with underlying circulatory disease (14). Prospective studies are required to define the role of therapeutic agents including ß-adrenoceptor blockers, and other antiarrhythmics, in reducing the described vascular morbidity and mortality associated with hyperthyroidism.
Should we anticoagulate and attempt cardioversion in those with AF?
The published data examining AF and embolic risk in thyrotoxicosis are limited (8, 10) but considered together suggest that the rate of embolism in thyrotoxic AF exceeds that for nonthyrotoxic AF not associated with rheumatic heart disease (20). Furthermore, the majority of clinically evident emboli in thyrotoxic AF involves the central nervous system and occurs most commonly early in the course of the disease (20). These findings are in keeping with data demonstrating highest rates of mortality from cerebrovascular causes within the first year following treatment for thyrotoxicosis (3).
It is generally recommended that anticoagulation be considered in those with AF and thyrotoxicosis (19); there are, however, no published data specifically examining the use of anticoagulants in thyrotoxic AF, so the risk/benefit ratio of such therapy remains to be established. The decision to treat patients who have thyrotoxicosis-induced AF with short- or long-term anticoagulation remains one to be made on an individual basis, taking into consideration age, associated heart disease, and risks of such therapy. Antiplatelet and anticoagulant agents differentially affect cardioembolic and noncardioembolic stroke in AF (21). Aspirin has a greater effect on noncardioembolic stroke than cardioembolic but may offer some degree of protection against cardioembolic stroke in AF patients, although these are more effectively prevented by anticoagulant therapy (21).
Treatment of thyrotoxicosis has been reported to lead to spontaneous reversion to sinus rhythm in nearly two-thirds of those with associated AF within 810 wk, although beyond 3 months few revert spontaneously to sinus rhythm (22). Cardioversion should therefore not be delayed if the patient is rendered euthyroid and still has AF after this period (22). Factors in addition to duration of AF should influence the decision to attempt cardioversion, in particular the presence or absence of concomitant heart disease.
Cellular mechanisms determining arrhythmogenesis
Cardiac myocytes constitute only one-third of the total cells that make up the myocardium. Fibroblasts, smooth muscle cells, endothelial cells, and other cell types constitute the majority of cardiac cells. Thyroid hormone-responsive proteins that have been investigated are largely of myocytic origin, and thyroid hormone influences on nonmyocytic cardiac cells have not been investigated in detail.
The biologically active hormone T3 mediates thyroid hormone action. Once inside the cardiac myocyte, T3 enters the nucleus and binds to nuclear receptors that are bound to DNA response elements of target genes. T3-responsive genes encode both structural and regulatory proteins in the heart. Several cardiac genes are modulated by thyroid hormone at transcriptional and posttranscriptional levels (Table 2
).
|
After commencing antithyroid therapy, biochemical euthyroidism may be achieved after 46 wk; the excess vascular mortality noted following treatment of thyrotoxicosis might be related to the persistence of cellular effects, despite the restoration of biochemical euthyroidism. There is evidence that the on-rate or induction of transcription of specific genes by thyroid hormones differs from the off-rate of transcription (following withdrawal of the hormones) (25) with the latter taking more time. Cellular changes could thus persist with continuing arrhythmic consequences because of electrical remodeling, especially of the atria.
Potential cellular differences between the atria and ventricles
The difference between the prevalence of arrhythmias arising in the atria and ventricles may be owing to a difference in sensitivity of the two tissues to the effects of thyroid hormones. Golf et al. (26) found the ß-adrenoceptor-binding capacity in the right atrium to be more than twice that of the left ventricle. These findings are consistent with those of another study demonstrating the turnover of noradrenaline to be markedly higher in atrial than ventricular tissue from laboratory animals. Furthermore, cardiac tissue is known to contain both ß1 and ß2 adrenoceptors. Stiles et al. (27) found about 26% of the receptors to be of ß2 subtype in the right atrium with about 14% in left ventricular tissue. Effects of thyroid hormone on the expression of these receptors will affect impulse generation and propagation and hence arrhythmogenesis. The level of expression of various voltage-gated potassium channels (especially Kv1.5) has been shown to be 30% higher in atrial, compared with ventricular, myocardium (25), which may also explain the observed discrepancy.
Effects of thyroid status on the autonomic nervous system
Several symptoms and signs in patients with hyperthyroidism suggest an abnormality of the autonomic nervous system. Alterations in vagal and sympathetic innervation can influence the development of arrhythmias with areas of sympathetic supersensitivity being conducive to the development of arrhythmias. Although the effect of thyroid hormone on the autonomic nervous system has been the subject of study for many years, there remain doubts about the nature of this interaction. It has been suggested that there is high adrenergic and low vagal activity in thyrotoxicosis, but circulating concentrations of catecholamines in hyperthyroidism suggest that secretion is normal or even reduced (18). To explain this paradox, some have suggested that thyroid hormones and catecholamines can exert similar effects because of their structural similarity (28). Increased tissue sensitivity to catecholamines, secondary to increased ß-adrenoceptors and reduced parasympathetic activity (29), have also been put forward as possible explanations.
Heart rate variability is a useful, noninvasive tool in the detection of autonomic activity on the sinus node. Cacciatori et al. (30) described patients with thyrotoxicosis who had reduced parasympathetic activity that returned to normal following antithyroid therapy. These findings were in keeping with the data of Kollai and Kollai (31), who found a low state of excitability in vagal motor neurons in response to baroreceptor stimulation in hyperthyroid patients, suggesting that the reduction in tonic vagal activity may be a consequence of secondary baroreceptor rearrangement. In contrast, Pitzalis et al. (32) found vagal activity (assessed by the same method) to be unaltered in patients with hyperthyroidism. Our own data suggest reduced heart rate variability (and vagal activity) in patients with hyperthyroidism that persists despite restoration of euthyroidism (17). Studies conducted in hypothyroid patients revealed significant reductions in normal-normal interval variations, suggesting reduced function of the parasympathetic nervous system, which was reversible with treatment of the hypothyroidism (33).
Conclusion
Thyrotoxicosis exerts major effects on the cardiovascular system, and many of these are reversible with effective antithyroid therapy. For this reason, thyrotoxicosis has been perceived as a benign disease, but long-term follow-up of those with both overt and subclinical thyroid hormone excess has revealed excess vascular mortality. The role of dysrhythmias (particularly supraventricular) may be critical in accounting for some of the excess cardiovascular and cerebrovascular mortality observed. Thyroid hormones are known to have both direct and indirect effects on the myocardium, affect the autonomic nervous system, and predispose to a number of arrhythmias. Routine ECG and 24-hr Holter monitoring (with heart rate variability analysis as a noninvasive marker of autonomic function) may help identify those at particular risk. The role of therapeutic intervention such as antiplatelet, antithrombotic, and/or antiarrhythmic therapy remains to be established in prospective therapeutic trials. A multidisciplinary approach to the management of these patients appears to be very important with involvement of both an endocrinologist and cardiologist.
Acknowledgments
We acknowledge the support of the British Heart Foundation, Endowment Fund of the former United Birmingham Hospitals, and research nurse Mrs. J. Daykin for invaluable contribution to our clinical studies.
Footnotes
F.O. is supported by a British Heart Foundation Research Fellowship.
Abbreviations: AF, Atrial fibrillation; SMR, standardized mortality ratio; SVC, supraventricular premature complexes.
Received August 27, 2001.
Accepted October 29, 2001.
References
This article has been cited by other articles:
![]() |
B. Biondi and D. S. Cooper The Clinical Significance of Subclinical Thyroid Dysfunction Endocr. Rev., February 1, 2008; 29(1): 76 - 131. [Abstract] [Full Text] [PDF] |
||||
![]() |
S. Metso, P. Jaatinen, H. Huhtala, A. Auvinen, H. Oksala, and J. Salmi Increased Cardiovascular and Cancer Mortality after Radioiodine Treatment for Hyperthyroidism J. Clin. Endocrinol. Metab., June 1, 2007; 92(6): 2190 - 2196. [Abstract] [Full Text] [PDF] |
||||
![]() |
P.-Q. Yuan and H. Yang Hypothyroidism increases Fos immunoreactivity in cholinergic neurons of brain medullary dorsal vagal complex in rats Am J Physiol Endocrinol Metab, November 1, 2005; 289(5): E892 - E899. [Abstract] [Full Text] [PDF] |
||||
![]() |
M. Helfand Screening for Subclinical Thyroid Dysfunction in Nonpregnant Adults: A Summary of the Evidence for the U.S. Preventive Services Task Force Ann Intern Med, January 20, 2004; 140(2): 128 - 141. [Abstract] [Full Text] [PDF] |
||||
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Endocrinology | Endocrine Reviews | J. Clin. End. & Metab. |
| Molecular Endocrinology | Recent Prog. Horm. Res. | All Endocrine Journals |