Introduction
Subclinical thyroid dysfunctions are defined by normal free-triiodothyronine (FT3) and free-thyroxine (FT4) concentrations in the presence of abnormal TSH, which is low-undetectable in subclinical hyperthyroidism (SH) and increased in subclinical hypothyroidism (Sh).
The clinical significance of subclinical thyroid dysfunction is much debated (1-3). The presence of tissue effects, symptoms and signs of mild thyroid hormone excess or deficiency and the management and treatment of these conditions are controversial issues. Similarly, the TSH cut-off point that determines the effects of subclinical thyroid dysfunction remains to be established.
The cardiovascular system, which is a major target of thyroid hormone, is sensitive to the effects of thyroid hormone excess or deficiency at the tissue level. Many symptoms and signs in patients with overt hypo- or hyperthyroidism are related to the reduced or increased action of thyroid hormone on cardiac function. Triiodothyronine (T3) affects the heart and vascular system through genomic and non-genomic mechanisms; it influences heart rate, systolic and diastolic function and systemic vascular resistance, and hence cardiac performance (4,5).
In human overt hyperthyroidism, the increase in left ventricular performance is predominantly sustained by the increased preload that results in enhanced left ventricular diastolic function (6,7). The reduced systemic vascular resistance, coupled with increased venous return and preload, enhances cardiac output (6,7). The decreased cardiac output in hypothyroid patients at rest depends largely on altered diastolic relaxation and hemodynamic loading (5,8). The reduced cardiac preload, combined with bradycardia and slightly depressed myocardial contractility, accounts for a subnormal cardiac output in overt hypothyroidism, whereas peripheral vascular resistance is remarkably increased (4,5,8). Moreover, cardiovascular alterations have been found in individuals with subclinical thyroid disease (9). This review covers the data about the progression of subclinical thyroid dysfunctions and cardiovascular risk. It also deals with the cardiovascular risk and the need for treatment as estimated from epidemiological data on cardiovascular morbidity and mortality.

Subclinical hyperthyroidism
Causes, prevalence and progression
The reported overall prevalence of SH ranges between 0.5 and 6.3%, the prevalence being higher in patients over 65 years. The prevalence can differ in relation to iodine intake and the TSH cut-off point used for diagnosis (10,11).
Subclinical hyperthyroidism may be caused by exogenous or endogenous factors. The exogenous form is usually related to TSH-suppressive therapy with L-thyroxine for the treatment of benign thyroid disease and differentiated thyroid carcinoma. The endogenous form is usually related to the same causes as overt thyrotoxicosis subsequent to autonomously functioning thyroid adenomas, and multinodular goitre. TSH suppressive or unintentional over-replacement L-thyroxine therapy was the most common form of SH (20.7%) among subjects taking L-thyroxine in the Colorado study (11), whereas endogenous factors accounted for a minority of cases and prevailed in areas of iodine insufficiency (12). Endogenous SH is usually a slowly progressive disorder and may last several years before developing into overt thyrotoxicosis. The risk of SH progressing to overt hyperthyroidism varies between 2% and 7% per year in patients with undetectable TSH (12). Unfortunately, there are no data on the progression of SH in patients with TSH between 0.1-0.5 mU/L.

Cardiovascular risk
Exogenous and endogenous SH may lead to signs and symptoms of thyroid hormone excess thereby mimicking adrenergic overactivity and impairing quality of life (12, 13). Subclinical hyperthyroidism affects the cardiovascular system in various ways, and its increased cardiovascular risk is well documented in the elderly (14,15). The cardiovascular abnormalities are similar in stable endogenous and exogenous SH (12)



The major cardiovascular findings in patients with SH coupled with undetectable TSH are a higher heart rate and a higher risk of supraventricular arrhythmias (15-17). The most consistent cardiac abnormality is a significant increase in left ventricular mass with unchanged or increased at-rest systolic function and, usually, impaired diastolic function (16,18-20). Moreover, reduced systolic performance on effort and decreased exercise tolerance has been reported in patients with SH who had a greater increase in left ventricular mass (21). Thyroid hormone-induced hypertrophy in SH is due primarily to the cardiac response induced by the increased cardiac workload. This is accordance with cardiac hypertrophy induced in rats by thyroid hormone excess (22). Moreover, the significant increase in left ventricular mass with a tendency towards LV concentric remodelling reported in patients with long-standing SH (16,18,19) may counteract the favourable effect acutely exerted by thyroid hormone on diastolic performance, and so lead to impaired ventricular relaxation and systolic dysfunction during effort (20,21). The altered passive elasticity of the ventricle (chamber stiffness) determined by the presence of myocardial hypertrophy is the major determinant of diastolic dysfunction in patients with SH.
The prognostic significance of these cardiovascular alterations in patients with SH remains to be clarified especially in young and middle-aged patients with low TSH. Unfortunately, there is no study of the effects on cardiovascular system of minimally suppressed TSH (i.e., TSH between 0.1-0-4 mU/L). However, the increase in heart rate and in left ventricular mass usually precedes the onset of more severe cardiovascular disease, and is an independent risk factor for increased cardiovascular morbidity and mortality in the general population (23).
The detrimental effects of SH are well documented in the elderly and atrial fibrillation represents an important cardiovascular risk. The Framingham study evaluated the risk of atrial fibrillation during the 10-year follow-up in 2007 people aged 60 years or older with endogenous or exogenous SH (15). The adjusted relative risk for atrial fibrillation was 3.1 times higher in the group with serum TSH ?0.1 mU/L compared with those with normal TSH concentrations (>0.4-5.0mU/L). The relative risk of atrial fibrillation was 1.6 times higher in the group with slightly low TSH concentrations (0.1-0.4 mU/L) (p=0.04) with an an incidence of atrial fibrillation in 16/1000 patients person-years) (p=0.11) (15). Similarly, in a large retrospective study on hospitalized consecutive older subjects, the relative risk of atrial fibrillation was 5.2 (p<0.01) in patients affected by SH with TSH <0.4 mU/L. (17). The combination of subclinical hyperthyroidism and age may have deleterious effects on the heart (12,23). Furthermore, the possible onset of overt hyperthyroidism in hearts previously exposed to longstanding untreated SH may further increase the cardiovascular risk (9). This body of data is in agreement with the increased cardiovascular mortality reported in a community-based review of subjects aged 60 years or older with endogenous SH and TSH values <0.5 mU/L monitored for 10 years (14).
In patients with benign thyroid disease and in low-risk patients with differentiated thyroid cancer, cardiovascular parameters and quality of life can be improved by reducing L-T4 dosage to keep TSH at the lower limit of normal range (24). A cardioselective b-blocking drug, in addition to L-thyroxine, can be used in high-risk thyroid cancer patients to attenuate symptoms caused by mild thyroid hormone excess, and to reduce the risk of atrial arrhythmias and an increase in left ventricular mass (12). Methimazole administration and radioiodine therapy may restore euthyroidism and so improve the cardiovascular risk in patients with endogenous SH (25,26).
A panel of experts recently recommended that treatment of endogenous SH should be considered in case of TSH <0.1 mU/L especially when patients are older than 60 years, and if there are symptoms or risk of heart disease (3). The routine treatment of SH if TSH is between 0.1 and 0.4 mU/L is not recommended (3). In fact, despite increased mortality in the elderly, it remains to be established if a subnormal TSH concentration induces the same adverse effects as suppressed TSH on the heart of young and middle-aged patients. However, it is important to stress that the clinical manifestations of ‘subclinical’ hyperthyroidism may be related to an individual's sensitivity to thyroid hormone excess, which depends on the patient’s thyroid function set-point (12), and may be triggered by individual predisposing conditions (12,27).

Subclinical hypothyroidism
Causes, prevalence and progression
Subclinical hypothyroidism reflects an early and mild form of thyroid failure. Most patients with Sh have chronic autoimmune thyroiditis and test positive for serum antithyroid peroxidase (anti-TPO) antibodies; in these patients, the risk of progression to overt disease is particularly increased. Poor compliance with L-T4 therapy or suboptimal treatment, may also result in Sh. Medications (e.g., lithium, iodine, interferon, etc.), 131I therapy or thyroidectomy and external irradiation of the neck can also cause Sh. The epidemiologic data from three large population-based screening studies (the Whickham Survey, the Colorado Thyroid Disease Prevalence Study and the National Health and Nutrition Examination Survey III) (10,11,28) show that the prevalence of Sh is 4-10% and that this condition increases significantly with age, so that by the ninth decade of life the prevalence is 15%-20%.
A mild TSH increase (between 4-10 mU/L) is present in about 74% of Sh patients, and whether to treat Sh associated with this TSH range is hotly debated. The progression to overt hypothyroidism occurs at a rate of 2-5% per year and it is increased in patients with TSH >6 mU/L and positive thyroid antibodies.
The reasons for treating Sh are: to prevent progression to overt disease, to attenuate symptoms, and to correct lipid profile and cardiovascular abnormalities and so reduce the cardiovascular risk. However, the TSH cut-off at which to start replacement therapy with L-thyroxine is still debated (1-3).

Cardiovascular risk
The cardiovascular risk in patients with Sh results from changes in cardiovascular function and from accelerated atherosclerosis (8,9)



The most consistent cardiac abnormality in patients with Sh is impaired left ventricular diastolic function, which is characterized by slowed myocardial relaxation and impaired early ventricular filling (9,29). Impaired left ventricular relaxation was identified in patients with Sh by echocardiography and radionuclide ventriculography (30-33). All studies of young and middle-aged patients with a mild and persistent TSH increase (4-10 mU/L) due to Hashimoto thyroiditis show that diastolic dysfunction of the left ventricle is a common finding in patients with persistent Sh (30-33). Diastolic function is impaired both at rest and during exercise (33). Slowed left ventricle relaxation is in accordance with the finding that thyroid hormone affects the calcium-regulating proteins SERCA and PLB thereby slowing down calcium re-uptake into the sarcoplasmic reticulum during diastole (29). Altered diastolic function can be reversed by L-thyroxine replacement therapy (8,9,29-33).
Conflicting results on systolic function are reported in patients with overt and Sh (8). Impaired left ventricular systolic function on effort was documented in patients with Sh by using radionuclide ventriculography, Doppler echocardiography and cardiopulmonary exercise testing (29, 34). The negative effect induced by Sh on systolic function at rest and during effort is reverted by restoring euthyroidism with L-T4 therapy (29-34). Ultrasonic myocardial textural analysis indicates that myocardial composition is altered in patients with Sh (35). Many Sh patients are elderly and the onset or progression of the disease in these vulnerable subjects may precipitate cardiac decompensation and promote congestive diastolic heart failure (5,8).
Overt hypothyroidism is associated with premature atherosclerosis and coronary artery disease. Epidemiological studies of the link between Sh and atherosclerosis have yielded conflicting results (36,37). Compelling evidence of a higher prevalence of atherosclerotic cardiovascular disease in patients with Sh (defined as TSH >4.0 mU/L) emerges from a recent large cross-sectional survey of 1,149 women aged 55 years or more, living in Rotterdam (37). It was shown that Sh patients had a significantly increased age-adjusted prevalence of aortic atherosclerosis on chest radiographs and myocardial infarction compared with controls. The attributable risk percentage for Sh associated with myocardial infarction was within the range of the traditionally recognized risk factors for coronary artery disease, including hypercholesterolemia, hypertension, smoking and diabetes mellitus. Moreover, in a cross-sectional analysis, Sh was associated with ischemic heart disease independent of age, systolic blood pressure, body mass index, cholesterol, smoking, or presence of diabetes mellitus (38).
The mechanisms responsible for atherosclerosis and coronary artery disease in patients with Sh are controversial (8,39). Diastolic hypertension (40,41), dyslipidemia (42-45), endothelial dysfunction (46), elevated C reactive protein levels (47) and coagulation abnormalities (48) are atherosclerotic risk factors associated with Sh and may be reversed after L-T4-induced euthyroidism.
In a recent review of guidelines, treatment of Sh was recommended when serum TSH is >10 mU/L so as to prevent progression to overt disease (3). Treatment of subclinical hypothyroidism based on cardiovascular risk was not recommended because the data available were considered insufficient and unconvincing (3). Furthermore, routine levo-thyroxine treatment was not recommended when TSH is between 4.5 and 10 mU/L (3).
However, mild hypothyroidism (TSH <10 mU/L) can negatively affect the cardiovascular system, especially diastolic function, endothelial function and systemic vascular resistance. Treatment of this mild form of hypothyroidism may improve cardiovascular function (8, 49) and it may prevent atherosclerosis and coronary artery disease (50) thereby reducing the cardiovascular risk.