Consequences of Subclinical Thyroid Dysfunction
Subclinical hyperthyroidism often reflects ingestion of thyroid hormones, typically thyroxine, and in that context is considered 'exogenous' in origin. If low serum TSH is found in the absence of thyroid hormone use, then it is labelled 'endogenous'. For both categories, given the inverse (but nonlinear) relationship between serum free T4 and TSH, complete suppression of serum TSH (to <0·1 mU/l) is generally considered of more pathophysiological significance than the finding of a low but detectable serum TSH (0·1–0·4 mU/l). Exogenous subclinical hyperthyroidism is more common than endogenous and is present in around 20–40% of the subjects prescribed thyroid hormones. As expected, low serum TSH is more common in those prescribed thyroxine in higher doses, indicating a degree of over-treatment. Studies of those not taking thyroid hormones also reveal a high prevalence of subclinical hyperthyroidism, with variations in frequency depending on age, gender, race and iodine intake within the population. The National Health and Nutrition Survey (NHANES) in the United States revealed 1·8% of the general population to have low but detectable serum TSH and only 0·7% to have fully suppressed serum TSH (after exclusion of 'exogenous' cases), with similar findings from a population prevalence study in Scotland. Both of these studies revealed a higher prevalence in women and a rise in frequency with age. Our own study of almost 6000 community-based subjects aged over 65 years attending general practices in the West Midlands region of England revealed a prevalence of subclinical hyperthyroidism of 2·1% in that age group, again being more common with increasing age. Ninety of 128 subjects with subclinical hyperthyroidism had low but detectable serum TSH, with a relatively small proportion having fully suppressed TSH.
The causes for subclinical hyperthyroidism are shown in Table 1. As well as thyroxine therapy, previous Graves' hyperthyroidism may be associated with suppression of TSH for weeks or even months after successful treatment, although if longstanding indicates persistent thyroid autonomy. Up to 75% of subjects with nodular goitre may also have TSH suppression, again indicating thyroid autonomy. It should be noted, however, that 'non-thyroidal' illnesses and therapies with various drugs probably represent the commonest causes for subclinical hyperthyroidism, especially in hospital outpatient or inpatient populations, the most common biochemical finding being low but detectable serum TSH.
The natural history of subclinical hyperthyroidism depends upon its cause and severity (i.e. the degree of reduction in serum TSH below the reference range). Some subjects in whom TSH suppression (usually complete) is associated with Graves' hyperthyroidism or nodular goitre will progress to overt hyperthyroidism, although the incidence is relatively low at around 1–3% per year. In contrast, those in whom low serum TSH values reflect 'non-thyroidal' illness or drug therapy typically have low but detectable serum TSH, and the biochemical abnormality often disappears after recovery from illness or cessation of drug therapy. A large study demonstrated that reduced serum TSH (<0·35 mU/l) returned to normal in more than half after a follow-up period of 5 years. This finding is compatible with one of our early screening and follow-up studies in the elderly which showed that of those with low but detectable TSH at initial testing, TSH had returned to normal in 76% at one year, compared with those with undetectable TSH of whom 88% had persistently undetectable TSH. A 10-year follow-up of the same group showed that only 4·3% of those with low serum TSH developed overt hyperthyroidism.
If low serum TSH reflects 'non-thyroidal' illness or drug therapy and is consequently transient, it is presumed that it is of little consequence in terms of long-term effects. In these situations, circulating thyroid hormone concentrations (especially T3) are typically low or low normal – the so-called sick euthyroid syndrome. The potential consequences of subclinical hyperthyroidism are probably confined to those in whom suppression of TSH reflects a minor degree of thyroid hormone excess, the major patient groups being those with thyroid autonomy owing to the presence of a nodular goitre or Graves' disease and those receiving thyroxine therapy. These subjects will often have serum T4 values at the upper end of the reference range, and higher than those found in subjects with normal serum TSH. Furthermore, serum T3 concentrations are typically high normal in those with suppressed TSH reflecting thyroid autonomy in contrast to those taking thyroxine in whom T3 is relatively low, a difference which may explain the greater pathophysiological significance of endogenous compared with exogenous subclinical hyperthyroidism.
Symptoms, Cognitive Function Relatively small cross-sectional studies have indicated increased palpitation and heat intolerance to be associated with endogenous and exogenous subclinical hyperthyroidism. Our own large and detailed study of cognition, depression and anxiety failed to demonstrate any association with subclinical thyroid dysfunction in those aged over 65 years. There is some evidence that subclinical hyperthyroidism may be associated with increased risk of dementia, but findings are conflicting.
Cardiovascular System Thyroid hormones are well recognized to exert effects on the cardiovascular system and subclinical hyperthyroidism is associated with similar findings to overt thyroid hormone excess, albeit less marked. Studies of ECG monitoring comparing subjects with low and normal serum TSH have revealed an increase in mean 24 h and nocturnal heart rate, shortening of the systolic time interval and an increase in frequency of atrial premature beats. Echocardiography studies have produced less consistent differences that include increased left ventricular mass and changes in systolic and diastolic function, findings of unclear clinical significance.
Evidence for adverse outcomes associated with subclinical hyperthyroidism has, however, accrued, especially in the context of risk of atrial fibrillation (AF) and its cardiovascular consequences. The seminal study by Sawin et al. of the Framingham cohort of subjects aged over 60 years revealed a 3·1-fold increased relative risk of AF after 10 years associated with suppressed TSH at time zero, with a relative risk of 1·6 in those with low but detectable TSH. Another large retrospective study demonstrated an adjusted relative risk of 2·8 for the finding of AF in subjects with low TSH compared with normal TSH and a further study showed a 2-fold increased incidence of AF in elderly subjects followed for 13 years, which included evidence for an increased incidence in those with low but detectable TSH values. Furthermore, our own cross-sectional study of 5860 subjects aged 65 years and over revealed a higher prevalence of AF confirmed by ECG in participants with subclinical hyperthyroidism than in those with normal serum TSH. We also found that serum free T4 concentration was independently associated with the finding of AF, even in euthyroid subjects with normal serum free T4 and TSH values (Fig. 1), suggesting marked sensitivity in terms of AF risk of even minor degrees of thyroid hormone excess.
(Enlarge Image)
Figure 1.
Prevalence of atrial fibrillation (AF) on resting 12-lead electrocardiogram plotted against serum free thyroxine (T4) concentrations in 5860 subjects aged 65 years and older. The plotted points were obtained by rounding each free T4 measurement to the nearest integer; the superimposed curve is that given by a logistic regression on the actual values of free T4vs the presence/absence of AF. Reproduced with permission from Gammage et al. Association between serum free T4 concentration and atrial fibrillation. Arch Intern Med 2007 May 14;167(9):928–34.
There is also evidence linking subclinical hyperthyroidism and mortality, although findings should be interpreted in the context of potential influences of comorbidities or genetic confounding (as recently discussed in the context of overt thyroid dysfunction.) We reported increased deaths from circulatory diseases (both cardiovascular and cerebrovascular) in association with low TSH in a 10-year follow-up study of 1191 subjects aged more than 60 years (Fig. 2). This increased mortality occurred in the absence of increased deaths from other common causes, suggesting that low TSH reflected a degree of thyroid autonomy rather than 'non-thyroidal' illness. Although a study of 39 subjects with subclinical hyperthyroidism followed for 20 years did not find adverse outcomes in the very elderly (aged >85 years), increased cardiovascular mortality during 4 years of follow-up was reported in another study of those with low TSH levels. Results of meta-analyses have also been conflicting, although the most recent by Collet et al. has linked subclinical hyperthyroidism with not only with incident AF but also increased risks of all-cause mortality and coronary heart disease mortality, the greatest risks being associated with TSH <0·1 mU/l.
(Enlarge Image)
Figure 2.
Kaplan-Meier survival curves showing the relation between survival from circulatory disease and serum TSH concentration. Reproduced with permission from Parle et al. Prediction of all-cause and cardiovascular mortality in elderly people from one low serum thyrotropin result: a 10-year cohort study. Lancet 2001 Sep 15;358(9285):861–5.
Musculoskeletal System Overt hyperthyroidism is associated with increased bone turnover, reflecting a direct effect of thyroid hormone excess on osteoclast function and hence with increased risk of osteoporosis and fracture. Numerous studies have examined potential associations of exogenous and endogenous subclinical hyperthyroidism with changes in bone mineral density (BMD), often with conflicting results probably due to small cohort sizes and heterogeneity with respect to history of previous overt hyperthyroidism, dose and duration of thyroxine therapy and menopausal status. Most studies, and hence meta-analyses, have suggested an adverse effect of subclinical hyperthyroidism in post menopausal women, especially in the context of previous overt hyperthyroidism, with much less evidence for an effect in premenopausal women or in men.
Evidence that such changes in BMD are translated into an increase in risk of osteoporotic fracture, especially fracture of the femur, is relatively limited. A prospective population-based study of fracture of the femur in postmenopausal women identified thyroxine prescription as a possible risk factor for fracture (RR 1·6, 95% CI 1·1–2·3), but this relative risk was no longer significant when a previous history of overt hyperthyroidism was taken into account; previous hyperthyroidism itself being associated with a relative risk of 1·8 (95% CI 1·2–2·6). Our own large study of fracture risk and thyroxine prescription (in which thyroid biochemistry was not available) similarly failed to demonstrate an overall association, although there was an increased risk of femur fracture in the relatively small number of male subjects prescribed thyroxine. Few studies have evaluated possible association between endogenous subclinical hyperthyroidism and fracture risk, although an investigation of 686 women aged >65 years with low serum TSH revealed a 3-fold to 4-fold increased risk of hip or vertebral fracture after adjustment for previous hyperthyroidism and thyroxine prescription when compared with subjects with normal TSH. A recent prospective cohort study has also shown an increase in risk of femur fracture in men aged over 65 years in those with endogenous subclinical hyperthyroidism (HR 4·91, 95% CI, 1·13–21·27) but with no association found in women.
Investigation and Treatment In subjects with subclinical hyperthyroidism taking thyroxine therapy, dose reduction followed by further biochemical testing to ensure that TSH has returned to the reference range is one obvious approach. Whether this is a cost-effective exercise given the frequency of thyroxine prescription in the general population (approximately 1% in the general population and 5% in the over 60 s) and so far inconclusive evidence for adverse long-term clinical consequences (with the possible exception of occurrence of atrial fibrillation) remains unclear. In patients starting thyroxine therapy, it appears appropriate to aim for biochemical as well as clinical euthyroidism, in line with both UK and US guidelines.
Addressing investigation and treatment is more complicated in those in whom suppression of TSH does not reflect thyroid hormone use. Before considering treatment, it is first essential to determine whether TSH suppression reflects autonomous thyroid function, although this is probably only relevant to those with an undetectable TSH value because if TSH is low but detectable this finding is often transient. If undetectable TSH is persistent on repeat testing, then evidence for underlying thyroid disease should be sought. The commonest cause for true endogenous subclinical hyperthyroidism is toxic nodular goitre, especially in the elderly. This may be obvious from history and physical examination but, if not, then thyroid isotope imaging is a reasonable approach to detect a 'hot' nodule.
The topic of treatment of endogenous subclinical hyperthyroidism is controversial as no controlled studies showing benefit in clinical outcomes have been performed, although small studies have shown improvement in echocardiographic parameters and bone mineral density. Because of associations with adverse outcomes, especially atrial fibrillation, expert panels have recommended treatment of subclinical hyperthyroidism, proven to reflect true thyroid disease, specifically in those with persistently undetectable TSH, the elderly and those with cardiac risks, heart disease or osteoporosis. Latterly, the suggestion has been made that treatment should be considered for low but detectable TSH in the elderly or with heart disease, driven by a study showing higher atrial fibrillation risk in such groups. If the decision is made to treat, radioiodine is generally the therapy of choice, especially in toxic nodular goitre, long-term carbimazole at low dose being another option.
Because of the absence of evidence for benefit from treatment, population screening for minor abnormalities of thyroid function is not presently recommended.
Subclinical Hyperthyroidism
Epidemiology
Subclinical hyperthyroidism often reflects ingestion of thyroid hormones, typically thyroxine, and in that context is considered 'exogenous' in origin. If low serum TSH is found in the absence of thyroid hormone use, then it is labelled 'endogenous'. For both categories, given the inverse (but nonlinear) relationship between serum free T4 and TSH, complete suppression of serum TSH (to <0·1 mU/l) is generally considered of more pathophysiological significance than the finding of a low but detectable serum TSH (0·1–0·4 mU/l). Exogenous subclinical hyperthyroidism is more common than endogenous and is present in around 20–40% of the subjects prescribed thyroid hormones. As expected, low serum TSH is more common in those prescribed thyroxine in higher doses, indicating a degree of over-treatment. Studies of those not taking thyroid hormones also reveal a high prevalence of subclinical hyperthyroidism, with variations in frequency depending on age, gender, race and iodine intake within the population. The National Health and Nutrition Survey (NHANES) in the United States revealed 1·8% of the general population to have low but detectable serum TSH and only 0·7% to have fully suppressed serum TSH (after exclusion of 'exogenous' cases), with similar findings from a population prevalence study in Scotland. Both of these studies revealed a higher prevalence in women and a rise in frequency with age. Our own study of almost 6000 community-based subjects aged over 65 years attending general practices in the West Midlands region of England revealed a prevalence of subclinical hyperthyroidism of 2·1% in that age group, again being more common with increasing age. Ninety of 128 subjects with subclinical hyperthyroidism had low but detectable serum TSH, with a relatively small proportion having fully suppressed TSH.
The causes for subclinical hyperthyroidism are shown in Table 1. As well as thyroxine therapy, previous Graves' hyperthyroidism may be associated with suppression of TSH for weeks or even months after successful treatment, although if longstanding indicates persistent thyroid autonomy. Up to 75% of subjects with nodular goitre may also have TSH suppression, again indicating thyroid autonomy. It should be noted, however, that 'non-thyroidal' illnesses and therapies with various drugs probably represent the commonest causes for subclinical hyperthyroidism, especially in hospital outpatient or inpatient populations, the most common biochemical finding being low but detectable serum TSH.
The natural history of subclinical hyperthyroidism depends upon its cause and severity (i.e. the degree of reduction in serum TSH below the reference range). Some subjects in whom TSH suppression (usually complete) is associated with Graves' hyperthyroidism or nodular goitre will progress to overt hyperthyroidism, although the incidence is relatively low at around 1–3% per year. In contrast, those in whom low serum TSH values reflect 'non-thyroidal' illness or drug therapy typically have low but detectable serum TSH, and the biochemical abnormality often disappears after recovery from illness or cessation of drug therapy. A large study demonstrated that reduced serum TSH (<0·35 mU/l) returned to normal in more than half after a follow-up period of 5 years. This finding is compatible with one of our early screening and follow-up studies in the elderly which showed that of those with low but detectable TSH at initial testing, TSH had returned to normal in 76% at one year, compared with those with undetectable TSH of whom 88% had persistently undetectable TSH. A 10-year follow-up of the same group showed that only 4·3% of those with low serum TSH developed overt hyperthyroidism.
Consequences of Subclinical Hyperthyroidism
If low serum TSH reflects 'non-thyroidal' illness or drug therapy and is consequently transient, it is presumed that it is of little consequence in terms of long-term effects. In these situations, circulating thyroid hormone concentrations (especially T3) are typically low or low normal – the so-called sick euthyroid syndrome. The potential consequences of subclinical hyperthyroidism are probably confined to those in whom suppression of TSH reflects a minor degree of thyroid hormone excess, the major patient groups being those with thyroid autonomy owing to the presence of a nodular goitre or Graves' disease and those receiving thyroxine therapy. These subjects will often have serum T4 values at the upper end of the reference range, and higher than those found in subjects with normal serum TSH. Furthermore, serum T3 concentrations are typically high normal in those with suppressed TSH reflecting thyroid autonomy in contrast to those taking thyroxine in whom T3 is relatively low, a difference which may explain the greater pathophysiological significance of endogenous compared with exogenous subclinical hyperthyroidism.
Symptoms, Cognitive Function Relatively small cross-sectional studies have indicated increased palpitation and heat intolerance to be associated with endogenous and exogenous subclinical hyperthyroidism. Our own large and detailed study of cognition, depression and anxiety failed to demonstrate any association with subclinical thyroid dysfunction in those aged over 65 years. There is some evidence that subclinical hyperthyroidism may be associated with increased risk of dementia, but findings are conflicting.
Cardiovascular System Thyroid hormones are well recognized to exert effects on the cardiovascular system and subclinical hyperthyroidism is associated with similar findings to overt thyroid hormone excess, albeit less marked. Studies of ECG monitoring comparing subjects with low and normal serum TSH have revealed an increase in mean 24 h and nocturnal heart rate, shortening of the systolic time interval and an increase in frequency of atrial premature beats. Echocardiography studies have produced less consistent differences that include increased left ventricular mass and changes in systolic and diastolic function, findings of unclear clinical significance.
Evidence for adverse outcomes associated with subclinical hyperthyroidism has, however, accrued, especially in the context of risk of atrial fibrillation (AF) and its cardiovascular consequences. The seminal study by Sawin et al. of the Framingham cohort of subjects aged over 60 years revealed a 3·1-fold increased relative risk of AF after 10 years associated with suppressed TSH at time zero, with a relative risk of 1·6 in those with low but detectable TSH. Another large retrospective study demonstrated an adjusted relative risk of 2·8 for the finding of AF in subjects with low TSH compared with normal TSH and a further study showed a 2-fold increased incidence of AF in elderly subjects followed for 13 years, which included evidence for an increased incidence in those with low but detectable TSH values. Furthermore, our own cross-sectional study of 5860 subjects aged 65 years and over revealed a higher prevalence of AF confirmed by ECG in participants with subclinical hyperthyroidism than in those with normal serum TSH. We also found that serum free T4 concentration was independently associated with the finding of AF, even in euthyroid subjects with normal serum free T4 and TSH values (Fig. 1), suggesting marked sensitivity in terms of AF risk of even minor degrees of thyroid hormone excess.
(Enlarge Image)
Figure 1.
Prevalence of atrial fibrillation (AF) on resting 12-lead electrocardiogram plotted against serum free thyroxine (T4) concentrations in 5860 subjects aged 65 years and older. The plotted points were obtained by rounding each free T4 measurement to the nearest integer; the superimposed curve is that given by a logistic regression on the actual values of free T4vs the presence/absence of AF. Reproduced with permission from Gammage et al. Association between serum free T4 concentration and atrial fibrillation. Arch Intern Med 2007 May 14;167(9):928–34.
There is also evidence linking subclinical hyperthyroidism and mortality, although findings should be interpreted in the context of potential influences of comorbidities or genetic confounding (as recently discussed in the context of overt thyroid dysfunction.) We reported increased deaths from circulatory diseases (both cardiovascular and cerebrovascular) in association with low TSH in a 10-year follow-up study of 1191 subjects aged more than 60 years (Fig. 2). This increased mortality occurred in the absence of increased deaths from other common causes, suggesting that low TSH reflected a degree of thyroid autonomy rather than 'non-thyroidal' illness. Although a study of 39 subjects with subclinical hyperthyroidism followed for 20 years did not find adverse outcomes in the very elderly (aged >85 years), increased cardiovascular mortality during 4 years of follow-up was reported in another study of those with low TSH levels. Results of meta-analyses have also been conflicting, although the most recent by Collet et al. has linked subclinical hyperthyroidism with not only with incident AF but also increased risks of all-cause mortality and coronary heart disease mortality, the greatest risks being associated with TSH <0·1 mU/l.
(Enlarge Image)
Figure 2.
Kaplan-Meier survival curves showing the relation between survival from circulatory disease and serum TSH concentration. Reproduced with permission from Parle et al. Prediction of all-cause and cardiovascular mortality in elderly people from one low serum thyrotropin result: a 10-year cohort study. Lancet 2001 Sep 15;358(9285):861–5.
Musculoskeletal System Overt hyperthyroidism is associated with increased bone turnover, reflecting a direct effect of thyroid hormone excess on osteoclast function and hence with increased risk of osteoporosis and fracture. Numerous studies have examined potential associations of exogenous and endogenous subclinical hyperthyroidism with changes in bone mineral density (BMD), often with conflicting results probably due to small cohort sizes and heterogeneity with respect to history of previous overt hyperthyroidism, dose and duration of thyroxine therapy and menopausal status. Most studies, and hence meta-analyses, have suggested an adverse effect of subclinical hyperthyroidism in post menopausal women, especially in the context of previous overt hyperthyroidism, with much less evidence for an effect in premenopausal women or in men.
Evidence that such changes in BMD are translated into an increase in risk of osteoporotic fracture, especially fracture of the femur, is relatively limited. A prospective population-based study of fracture of the femur in postmenopausal women identified thyroxine prescription as a possible risk factor for fracture (RR 1·6, 95% CI 1·1–2·3), but this relative risk was no longer significant when a previous history of overt hyperthyroidism was taken into account; previous hyperthyroidism itself being associated with a relative risk of 1·8 (95% CI 1·2–2·6). Our own large study of fracture risk and thyroxine prescription (in which thyroid biochemistry was not available) similarly failed to demonstrate an overall association, although there was an increased risk of femur fracture in the relatively small number of male subjects prescribed thyroxine. Few studies have evaluated possible association between endogenous subclinical hyperthyroidism and fracture risk, although an investigation of 686 women aged >65 years with low serum TSH revealed a 3-fold to 4-fold increased risk of hip or vertebral fracture after adjustment for previous hyperthyroidism and thyroxine prescription when compared with subjects with normal TSH. A recent prospective cohort study has also shown an increase in risk of femur fracture in men aged over 65 years in those with endogenous subclinical hyperthyroidism (HR 4·91, 95% CI, 1·13–21·27) but with no association found in women.
Investigation and Treatment In subjects with subclinical hyperthyroidism taking thyroxine therapy, dose reduction followed by further biochemical testing to ensure that TSH has returned to the reference range is one obvious approach. Whether this is a cost-effective exercise given the frequency of thyroxine prescription in the general population (approximately 1% in the general population and 5% in the over 60 s) and so far inconclusive evidence for adverse long-term clinical consequences (with the possible exception of occurrence of atrial fibrillation) remains unclear. In patients starting thyroxine therapy, it appears appropriate to aim for biochemical as well as clinical euthyroidism, in line with both UK and US guidelines.
Addressing investigation and treatment is more complicated in those in whom suppression of TSH does not reflect thyroid hormone use. Before considering treatment, it is first essential to determine whether TSH suppression reflects autonomous thyroid function, although this is probably only relevant to those with an undetectable TSH value because if TSH is low but detectable this finding is often transient. If undetectable TSH is persistent on repeat testing, then evidence for underlying thyroid disease should be sought. The commonest cause for true endogenous subclinical hyperthyroidism is toxic nodular goitre, especially in the elderly. This may be obvious from history and physical examination but, if not, then thyroid isotope imaging is a reasonable approach to detect a 'hot' nodule.
The topic of treatment of endogenous subclinical hyperthyroidism is controversial as no controlled studies showing benefit in clinical outcomes have been performed, although small studies have shown improvement in echocardiographic parameters and bone mineral density. Because of associations with adverse outcomes, especially atrial fibrillation, expert panels have recommended treatment of subclinical hyperthyroidism, proven to reflect true thyroid disease, specifically in those with persistently undetectable TSH, the elderly and those with cardiac risks, heart disease or osteoporosis. Latterly, the suggestion has been made that treatment should be considered for low but detectable TSH in the elderly or with heart disease, driven by a study showing higher atrial fibrillation risk in such groups. If the decision is made to treat, radioiodine is generally the therapy of choice, especially in toxic nodular goitre, long-term carbimazole at low dose being another option.
Because of the absence of evidence for benefit from treatment, population screening for minor abnormalities of thyroid function is not presently recommended.
Source...