- Email: [email protected]
Subclinical hypothyroidism: the case for treatment
Opinion
TRENDS in Endocrinology and Metabolism
Vol.14 No.6 August 2003
257
Subclinical hypothyroidism: the case for treatment Penelope J.D. Owen and John H. Lazarus Dept Medicine, University of Wales College of Medicine, Llandough Hospital, Penarth, UK CF64 2XX
The issue regarding patients with subclinical hypothyroidism and whether they should be treated has been debated for many years in clinical endocrinology. The main concerns are the cardiovascular and hyperlipidaemic risks associated with this condition, in addition to the risk that these patients have of developing neuropyschiatric features while progressing to overt hypothyroidism. Early detection, potential benefits of longterm treatment and patient satisfaction associated with the alleviation of symptoms are all important aspects of therapy. Here, we detail many studies, including randomized control trials and case –control studies, that demonstrate the benefits of thyroxine replacement therapy on cognition, symptoms such as dry skin and low mood (among others), the cardiovascular system and plasma lipids. We conclude that it is better to treat this not uncommon condition by showing the benefits of early replacement therapy in selected groups. The clinical background to subclinical hypothyroidism (SCH) and its treatment must be considered in relation to the action of thyroid hormone at the cellular level [1]. Briefly, triiodothyronine (T3) formed mainly by the outer ring peripheral deiodination of tetraiodothyronine (T4) acts through two nuclear thyroid hormone receptors (a and b) to regulate appropriate tissue-specific protein synthesis. The situation is complicated by existence of tissue specificity for the receptors and differential isoform tissue distribution. Furthermore, the derivation of T3 also varies; for example, in the brain, cellular T3 is derived from brain T4, whereas in other tissues, circulating T3 is transported to the particular cell in question. This implies that even mild overt hypothyroidism (OH) – that is, raised thyrotrophin (TSH) and low levels of T4 – might have significant effects on the efficiency of thyroid hormone action. What is subclinical hypothyroidism? SCH is defined by the presence of a raised TSH level but normal circulating total or free T4 and T3 levels, usually (but not always) in the absence of clinical symptoms, representing a clinically silent thyroid dysfunction [2]. Its estimated prevalence has been reported to range between 5% and 15% in the general population [3], and it is more prevalent in women, Caucasians, those over 50 years of
age and men and women who have a positive family history of thyroid disease. Aetiology of subclinical hypothyroidism SCH patients can be divided into those who have never received thyroid medication, the so-called endogenous SCH group, and those on medication or with SCH as the result of other iatrogenic procedures, the exogenous SCH group. Box 1 details those individuals with an increased risk for developing SCH. The most common cause of endogenous SCH is autoimmune (Hashimoto’s) thyroiditis [4], which is usually associated with raised levels of antithyroperoxidase antibodies. Inadequately treated OH is one of the most common causes of exogenous SCH, and patients receiving radiotherapy for head and neck tumours can also develop the condition. Diagnosis of subclinical hypothyroidism When diagnosing endogenous SCH it is important that concurrent acute or chronic illnesses and thyroid hormone resistance be excluded, and that there be no pharmacological interference from drugs such as lithium, amiodarone, metoclopramide or domperidone. The ‘cut-off ’ values used in TSH assays affect prevalence, with one study reporting a prevalence as high as 11.3% when a TSH value greater than 3 mU l21 was used [5], and others demonstrating a prevalence as low as 6% when using a TSH value of 7 mU l21 [6]. SCH can progress to OH, and we suggest that avoidance of this is an important benefit (vide infra) of early treatment.
Box 1. Individuals with an increased risk of developing subclinical hypothyroidism † Those with a goitre in the family (e.g. inactivating mutations in thyrotrophin) † Those with treated hyperthyroidism (Graves’ disease) † Those with a history of postpartum thyroiditis † Those with a history of neck irradiation † Those with autoimmune disorders (diabetes mellitus type 1) † Those treated with iodine-containing anti-arrhythmic agents (amiodarone) † Those treated with lithium † Those treated with immune response modulators (interferon a) † Those recovering from severe non-thyroidal illness † Those with poor compliance of thyroxine therapy
Corresponding author: J.H. Lazarus ([email protected]). http://tem.trends.com 1043-2760/03/$ - see front matter q 2003 Elsevier Ltd. All rights reserved. doi:10.1016/S1043-2760(03)00108-5
258
Opinion
TRENDS in Endocrinology and Metabolism
The assessment of the risks associated with SCH and the evaluation of the evidence available in favour of treatment can result in an informed decision about whether to proceed with treatment. The benefits of treatment of subclinical hypothyroidism are the prevention of the onset of OH, the alleviation of symptoms such as dry skin, lethargy and depression, and the reduction of both serum cholesterol levels and cardiovascular risk. Prevention of progression to OH SCH can be regarded as a stage in the development of OH [7], and should be considered as subclinical thyroid failure. An epidemiological study examining the risk of progression of SCH to OH over a 20 year period found a 4.3% risk at one year of developing OH in patients with SCH and antithyroid antibodies and a 38-fold increased risk compared with patients with a normal TSH level and no antibodies [8]. However, other studies have suggested that the risk of progression from SCH to OH is as high as 18% in patients studied over the same period during the first year after the demonstration of a raised TSH [9]. OH can (rarely) lead to extreme morbidity, with complications such as cardiac failure or myxoedema coma occurring in some patients. Alleviation of symptoms of SCH Although patients with SCH are presumed to be asymptomatic by the definition ‘subclinical’ they can have subtle clinical manifestations and non-specific symptomatology. In a previous randomized control trial (RCT) [10], dry skin, cold intolerance, constipation and easy fatigability were noted as the most common presenting symptoms, but other signs and symptoms can be vague and hard to define; because of this, they are sometimes attributed to complaints associated with normal aging. However, patients usually present to a physician for one or more reasons. Canaris et al. [11], in a cross-sectional study involving 25 862 patients, found that those with SCH had significantly more overall symptoms, such as dry skin and lethargy, than did euthyroid patients. Three double-blinded, placebo-controlled RCTs have demonstrated an improvement in symptoms with replacement T4 therapy [9,12,13]. Different methods have been used to assess response. Meier et al. [12] used the ‘hypothyroidism diagnostic index’ of Billewicz et al. [14] and the scores of Zulewski et al. [15] to demonstrate significant improvement ðP , 0:02Þ in symptoms; other RCTs have used questionnaires and have also shown improvement with T4 [10]. The greatest improvement in patients’ symptoms has been shown in those with a baseline TSH .12 mU l21 [12]. It has been suggested that mood disorders and disturbances in cognition and memory are associated with SCH [16]. The lifetime frequency of depression is significantly higher in patients with SCH compared with patients with normal thyroid function [17], suggesting that SCH lowers the threshold for depression. A case –control study has also supported these findings and shown that patients have significantly increased anxiety, somatic complaints and hysteria [18]. Other http://tem.trends.com
Vol.14 No.6 August 2003
abnormal mood disorders noted in SCH include an increased incidence of concurrent panic disorder in depressed patients with SCH [19], which might respond less well to antidepressant therapy. Another study [20] showed significant decreases in logical memory in women with SCH compared with euthyroid individuals. A significant improvement of memory skills was induced by T4 treatment in these patients [20]. Reduction of hyperlipidaemia Although OH is usually associated with dyslipidaemia, SCH has a less well recognized association. A previous review has shown that SCH is two to three times more frequent in people with raised total cholesterol (TC) levels [21]. Caraccio et al. [22] found significant increases in TC and low-density lipoprotein cholesterol (LDL-C) in patients compared with euthyroid controls, and this statistical difference has also been confirmed in other case –control studies [23,24]. Therefore, the treatment of SCH could potentially reduce serum cholesterol levels and retard the development of atherosclerosis. Three RCTs comparing T4 and placebo have shown statistically significant reductions in both plasma TC and LDL-C concentrations [12,22,25], with improvement being of greatest benefit in those with higher initial pretreatment TSH and cholesterol levels. There are only limited data on the effects of replacement T4 therapy on apolipoprotein A1, apolipoprotein B (Apo B) and lipoprotein (a). Two studies have shown significant reductions in Apo B following treatment if the initial TSH level is . 10 mU l21 [12,23]. The effects of adding 3-hydroxy-3-methylglutaryl coenzyme A reductase inhibitors (statins) in addition to replacement T4 have been assessed in postmenopausal women. A recent prospective study found that combination therapy caused significant reductions in TC, LDL-C and trigliceride, and a significant increase in HDL-C levels at three months post-treatment [26]. Reduction of cardiovascular risk Cardiac abnormalities in hypothyroidism are well documented [27,28]. More recently, several studies have investigated cardiac changes evident in SCH and even though SCH is considered as ‘mild thyroid failure’ there is evidence to show that slight changes in thyroid hormone status do cause cardiac functional abnormalities. These include impairment of left ventricular (LV) diastolic function at rest (seen as slowing of relaxation of the left ventricle) [3,29] and LV systolic dysfunction [29]. Other changes in LV morphology include increased diastolic interventricular septum thickness [3,30], posterior wall thickness and LV mass [3]. Perhaps of most concern in one study was evidence of impaired intrinsic myocardial contractility [3]. There is evidence that these abnormalities improve with T4 treatment, demonstrating that adequate thyroid hormone replacement improves cardiac output accompanied by a substantial decrease in systemic vascular resistance [31], a reversal of diastolic dysfunction [3,29], and importantly an improvement in LV ejection fraction during exercise [32]. Recently, the technique of
Opinion
TRENDS in Endocrinology and Metabolism
pulsed tissue Doppler echocardiography (used to assess regional myocardial wall velocity) has been used to show that SCH is associated with significant prolongation of the pre-ejection, left ventricular ejection and isovolumetric times [33]. This is related to the fact that ATP-dependent Ca2þ transport in the sarcoplasmic reticulum is controlled by thyroid hormone [34]. Patients with SCH have more cardiovascular risk factors and are more likely to develop atherosclerosis and other cardiovascular diseases [35]. Endothelial dysfunction, known to contribute to the development of atherosclerosis [36] and detected by the use of flow mediated arteriolar dilatation, has been shown in patients with SCH [37]. In view of the clear structural and biological cardiovascular risks associated with the presence of SCH, treatment of this condition would be expected to provide protection against the development of cardiovascular disease, although there are no published longterm outcome studies demonstrating this. Pregnancy Thyroid dysfunction in pregnancy has been associated with both maternal and foetal problems. The impairment of neuro-intellectual development in children born to mothers with SCH during pregnancy has been shown to result from low foetal access to circulating maternal T4 during the first trimester [38]. In this group of patients, it is essential that treatment is recommended and that screening programmes be introduced. Even though screening might not be performed until after conception, it is still important because the precise details of thyroid hormone on brain development have not yet been delineated. Although there are no RTCs in this area, four cohort studies have been performed [39 –42], and Klein et al. [39] reported an inverse correlation between the severity of maternal hypothyroidism and the intelligence quotient of the offspring, and reinforced the need for a screening programme. Reproductive dysfunction or ‘subfertility’ can remain unrecognized in patients with SCH, and women being investigated for infertility should have their thyroid function tested. A variable rate of infertility has been noted in patients with SCH [43]. One study found that patients with SCH have increased prolactin secretion, although the impact of SCH on gonadal function and infertility has yet to be clarified [44]. Screening Screening depends upon the perceived benefits (once SCH is detected) of treating the condition, and ideally should be based on a rigorous cost – benefit analysis. Although population-based screening programmes are generally not recommended, in certain high-risk groups there is evidence that screening is cost effective (e.g. in patients with type 1 diabetes mellitus [45], women over 40 years of age [46], pregnant women and those with a positive thyroid antibody status) [47]. These selected groups of patients (apart from pregnant women) would benefit from periodic screening (probably annually), with treatment instituted if SCH were to be documented. http://tem.trends.com
Vol.14 No.6 August 2003
259
In 2000, the American Thyroid Association recommended that adults should be screened every five years from the age of 35 years by measurement of their TSH levels [48]. We think that this option is too expensive and unproven in terms of population benefit at this time, and therefore suggest a targeted focused approach. Patient choice? Although SCH is considered to be asymptomatic and assumed to be discovered only on laboratory testing, when asked, most patients would volunteer some positive symptomatology [11]. Therefore, when treatment is decided on should patients be offered a choice? In today’s multi media world, with most people having access to the Internet, it seems inevitable that patients will demand treatment. However, it will be of greater importance to reinforce the argument for treatment with clinical evidence. Final decision The decision to treat must be based on the individual patient, once any underlying coexisting morbidity or pharmacological interference has been excluded (see Box 2 for examples that should aid in decision making). The future To date, there are few RCTs and longitudinal studies involving the treatment of SCH, and treatment studies are
Box 2. Two examples that should aid in decision making Examples (1) A 45-year-old, non-smoking woman with a free thyroxine (T4) of 12.3 mmol l21, thyroid-stimulating hormone (TSH) of 10.4 mU l21, antithyroperoxidase (TPO) antibodies at . 600 kU l21 (normal range , 32 kU l21), and total cholesterol of 6.2 mmol l21, complaining of lethargy. Would you treat? (2) A 54-year-old woman, smoker, with a T4 of 14.8 mmol l21, TSH level of 6.8 mU l21, TPO antibodies of , 5 kU l21, complaining of feeling depressed and unable to lose weight. Would you treat?
Answers (1) This patient has subclinical hypothyroidism (SCH) with positive TPO antibodies. This, together with lethargic symptoms and hypercholesterolaemia, are all potential reasons to recommend treatment with T4. (2) This patient represents more of a dilemma. Her SCH is almost not autoimmune in nature, her TSH is below 10 mU l21 and her main symptoms are depression and weight gain. It is important to exclude any comorbidity involved or pharmacological interference, especially agents used in treatment of her depression. We would certainly not wish to start T4 therapy given this clinical scenario, and would ideally wait for repeat thyroid function tests to note any documented trend in T4 concentration or upward trend in TSH. Although the inherent dangers of T4 administration to patients complaining of hypothyroid symptoms who are biochemically euthyroid are recognized by nearly all clinicians, we suggest that, in patients with SCH (as in example 2), a ‘therapeutic trial’ of T4 might be justified.
260
Opinion
TRENDS in Endocrinology and Metabolism
Vol.14 No.6 August 2003
Detection of an elevated TSH level and normal T4 and T3 levels
Repeat TFTs TPO Ab Full lipid profile (exclude renal failure, non thyroidal illness, drugs or interference from iodine-containing agents)
TPO Ab positive
TPO Ab negative
TSH level >10 mUl–1
Treat with T4
TSH level <10 mUl–1
Symptoms evident Goitre Smokers Pregnancy Type 1 DM Cadiovascular risks Ovulatory dysfunction with infertility
No symptoms or other risk factors
Repeat TFTs at six months
TRENDS in Endocrinology & Metabolism
Fig. 1. Recommendations for the treatment of patients with subclinical hypothyroidism. Adapted from Cooper [49]. Abbreviations: DM, diabetes mellitus; T3, triiodothyronine; T4, tetraiodothyronine; TFT, thyroid function test; TPO Ab, antithyroperoxidase antibodies; TSH, thyroid-stimulating hormone.
limited by the lack of placebo controls. In addition, studies undertaken to date usually involve predominately women. This, together with the small sample sizes used and variable definitions of SCH, limits the application of study findings, and it is important that future RCTs assess the benefits of replacement T4 in larger groups of patients. With regular follow up of patients and thyroid function tests performed periodically, any risks of over replacement can be avoided. We take the view that SCH should be treated, definitely in the ‘at risk’ groups of patients (Fig. 1), in view of the outcome data on cardiovascular disease and lipids. We recognize that longterm outcome data are not available in relation to many features of SCH, but they are available for lipid abnormalities. We believe that future large wellcontrolled prospective studies will, in general, confirm the positive suggestive evidence already provided by the shortterm RCTs and accurate observational studies on the abnormal clinical and laboratory findings in this common condition. References 1 Brent, G.A. (2002) Tissue specific actions of thyroid hormone: insights from animal models. Rev. Endocr. Metab. Disord. 1, 27 – 33 2 Benediktsson, R. and Toft, A.D. (1998) Management of the unexpected result: compensated hypothyroidism. Postgrad. Med. J. 74, 729 – 732 http://tem.trends.com
3 Monzani, F. et al. (2001) Effect of levothyroxine on cardiac function and structure in subclinical hypothyroidism. J. Clin. Endocrinol. Metab. 86, 1110 – 1115 4 Rapoport, B. (1991) Pathophysiology of Hashimoto’s thyroiditis and hypothyroidism. Annu. Rev. Med. 42, 91 – 96 5 Rivolta, G. et al. (1999) Prevalence of subclinical hypothyroidism in a population living in the Milan metropolitan area. J. Endocrinol. Invest. 22, 693 – 697 6 Kanaya, A.M. et al. (2002) Association between thyroid dysfunction and total cholesterol level in an older biracial population: the health, ageing and body composition study. Arch. Intern. Med. 162, 773 – 779 7 Samuels, M.H. (1998) Subclinical thyroid disease in the elderly. Thyroid 8, 803– 813 8 Vanderpump, M.P. et al. (1995) The incidence of thyroid disorders in the community: a twenty-year follow up of the Whickham survey. Clin. Endocrinol. (Oxf) 43, 55 – 68 9 Parle, J.V. et al. (1991) Prevalence and follow-up of abnormal thyrotropin (TSH) concentrations in the elderly in the United Kingdom. Clin. Endocrinol. 34, 77 – 83 10 Cooper, D.S. et al. (1984) L -Thyroxine therapy in subclinical hypothyroidism. A double blind, placebo-controlled trial. Ann. Intern. Med. 101, 18 – 24 11 Canaris, G.J. et al. (2000) The Colorado thyroid disease prevalence study. Arch. Intern. Med. 160, 526 – 534 12 Meier, C. et al. (2001) TSH controlled L -thyroxine therapy reduces cholesterol levels. J. Clin. Endocrinol. Metab. 86, 4860 – 4866 13 Nystrom, E. et al. (1988) A double blind cross-over 12 month study of L -thyroxine treatment of women with ‘subclinical’ hypothyroidism. Clin. Endocrinol. 29, 63 – 75
Opinion
TRENDS in Endocrinology and Metabolism
14 Billewicz, W.Z. et al. (1969) Statistical methods applied to the diagnosis of hypothyroidism. Q. J. Med. 38, 255– 266 15 Zulewski, H. et al. (1997) Estimation of tissue hypothyroidism by a new clinical score: evaluation of patients with various grades of hypothyroidism and controls. J. Clin. Endocrinol. Metab. 82, 771 – 776 16 Hendrick, V. et al. (1998) Psychoneuroendocrinology of mood disorders. The hypothalamic– pituitary – thyroid axis. Psychiatr. Clin. North Am. 21, 277 – 292 17 Haggerty, J. et al. (1993) Subclinical hypothyroidism: a modifiable risk factor for depression? Am. J. Psychiatry 150, 508– 510 18 Monzani, F. et al. (1993) Subclinical hypothyroidism: neurobehavioural features and beneficial effect of L -thyroxine treatment. Clin. Investig. 71, 367– 371 19 Joffe, R.T. and Levitt, A.J. (1992) Major depression and subclinical (grade 2) hypothyroidism. Psychoneuroendocrinology 17, 215 – 221 20 Baldini, I. et al. (1997) Psychopathological and cognitive features in subclinical hypothyroidism. Prog. Neuropsychopharmacol. Biol. Psychiatry 21, 925 – 935 21 Tanis, B.C. et al. (1996) Effect of thyroid substitution on hypercholesterolaemia in patients with subclinical hypothyroidism: a reanalysis of interventional studies. Clin. Endocrinol. 44, 643 – 649 22 Carracio, N. et al. (2002) Lipoprotein profile in subclinical hypothyroidism: response to levothyroxine replacement, a randomised placebocontrolled study. J. Clin. Endocrinol. Metab. 87, 1533 – 1538 23 Efstathiadou, Z. et al. (2001) A lipid profile in subclinical hypothyroidism: is L -thyroxine substitution beneficial? Eur. J. Endocrinol. 145, 705 – 710 24 Yildirimkaya, M. et al. (1996) Lipoprotein (a) concentration in subclinical hypothyroidism before and after L -thyroxine therapy. Endocr. J. 43, 731 – 736 25 Evidence Report Subclinical Thyroid Disease. (1999) Effects of TSHcontrolled thyroxine therapy on lipid profile and peripheral target tissues in patients with subclinical hypothyroidism. The Lewin Group: prepared for the Endocrine Society, pp. 45 26 Evidence Report Subclinical Thyroid Disease. (2002) The comparison of L -thyroxine versus L -thyroxine and statin combinations in lipid status in patients with subclinical hypothyroidism. The Lewin group: prepared for the Endocrine Society, pp. 51 27 Gonzales Vilchez, F. et al. (1998) Cardiac manifestations of primary hypothyroidism. Determinant factors and treatment response. Rev. Esp. Cardiol. 51, 893– 900 28 Ridgway, E.C. et al. (1982) Cardiac function in mild and severe primary hypothyroidism. Life Sci. 30, 651 – 658 29 Biondi, B. et al. (1999) Left ventricular dysfunction in patients with subclinical hypothyroidism. J. Clin. Endocrinol. Metab. 84, 2064 – 2067 30 Varma, R. et al. (1996) Heart in hypothyroidism – an echocardiographic study. J. Assoc. Physicians India 44, 390 – 392 31 Faber, J. et al. (2002) Haemodynamic changes after levothyroxine treatment in subclinical hypothyroidism. Thyroid 12, 19 – 24
Vol.14 No.6 August 2003
261
32 Forfar, J.C. et al. (1985) Left ventricular performance in subclinical hypothyroidism. Q. J. Med. 57, 857 – 865 33 Vitale, G. et al. (2002) Left ventricular myocardial impairment in subclinical hypothyroidism assessed by a new ultrasound tool: pulsed tissue doppler. J. Clin. Endocrinol. Metab. 87, 4350– 4355 34 Warnick, P.R. et al. (1993) Rabbit skeletal muscle sarcoplasmic reticulum Ca(2 þ )-ATPase activity: stimulation in vitro by thyroid hormone analogues and bipyridines. Biochim. Biophys. Acta 1153, 184– 190 35 Staub, J.J. et al. (1993) Spectrum of subclinical and overt hypothyroidism: effect on thyrotropin, prolactin and thyroid reserve and metabolic impact on peripheral target tissues. Am. J. Med. 92, 631– 642 36 Kuikka, J.T. et al. (2001) Imaging of the endothelial dysfunction in coronary atherosclerosis. Eur. J. Nucl. Med. 28, 1567 – 1578 37 Lekakis, J. et al. (1997) Flow mediated endothelium-dependent vasodilation is impaired in subjects with hypothyroidism, borderline hypothyroidism and high normal serum thyrotropin (TSH). Thyroid 7, 411 – 414 38 Calvo, R. et al. (2002) Fetal tissues are exposed to biologically relevant free thyroxine concentrations during early phases of development. J. Clin. Endocrinol. Metab. 87, 1768– 1777 39 Klein, R.Z. et al. (2001) Relation of severity of maternal hypothyroidism to cognitive development of offspring. J. Med. Screen. 8, 18 – 20 40 Abalovich, M. et al. (2002) Overt and subclinical hypothyroidism complicating pregnancy. Thyroid 12, 63 – 68 41 Radetti, G. et al. (2000) Psychomotor and audiological assessment of infants born to mothers with subclinical thyroid dysfunction in early pregnancy. Minerva Pediatr. 52, 691 – 698 42 Kuijpens, J.L. et al. (1998) Prediction of postpartum thyroid dysfunction: can it be improved? Eur. J. Endocrinol. 139, 36 – 43 43 Glinoer, D. (1997) The regulation of thyroid function in pregnancy: pathways of endocrine adaptation from physiology to pathology. Endocr. Rev. 18, 404 – 433 44 Tolina, A. et al. (1991) Subclinical hypothyroidism and hyperprolactinaemia. Acta Eur. Fertil. 22, 275 – 277 45 Perros, P. et al. (1995) Frequency of thyroid dysfunction in diabetic patients: value of annual screening. Diabet. Med. 12, 622– 627 46 Wiersinga, W.M. (1995) Subclinical hypothyroidism and hyperthyroidism. I. Prevalence and clinical relevance. Neth. J. Med. 46, 197 – 204 47 Glinoer, D. et al. (1994) Risk of subclinical hypothyroidism in pregnant women with asymptomatic autoimmune thyroid disorders. J. Clin. Endocrinol. Metab. 79, 197– 204 48 Ladenson, P.W. et al. (2000) American Thyroid Association guidelines for detection of thyroid dysfunction. Arch. Intern. Med. 160, 1573– 1575 49 Cooper, D.S. et al. (2001) Subclinical hypothyroidism. N. Engl. J. Med. 345, 260 – 265
Conference Reporter AACR 2003 Are you unable to attend the 94th Annual Meeting of the American Association for Cancer Research in Washington DC on 11–14 July 2003? Don’t miss out! Sign up for daily reports from the conference at: http://news.bmn.com/alerts Each day’s top stories, written by the BioMedNet Conference Reporter team, can be sent direct from the conference to your inbox in either text or html format. http://tem.trends.com
TRENDS in Endocrinology and Metabolism
Vol.14 No.6 August 2003
257
Subclinical hypothyroidism: the case for treatment Penelope J.D. Owen and John H. Lazarus Dept Medicine, University of Wales College of Medicine, Llandough Hospital, Penarth, UK CF64 2XX
The issue regarding patients with subclinical hypothyroidism and whether they should be treated has been debated for many years in clinical endocrinology. The main concerns are the cardiovascular and hyperlipidaemic risks associated with this condition, in addition to the risk that these patients have of developing neuropyschiatric features while progressing to overt hypothyroidism. Early detection, potential benefits of longterm treatment and patient satisfaction associated with the alleviation of symptoms are all important aspects of therapy. Here, we detail many studies, including randomized control trials and case –control studies, that demonstrate the benefits of thyroxine replacement therapy on cognition, symptoms such as dry skin and low mood (among others), the cardiovascular system and plasma lipids. We conclude that it is better to treat this not uncommon condition by showing the benefits of early replacement therapy in selected groups. The clinical background to subclinical hypothyroidism (SCH) and its treatment must be considered in relation to the action of thyroid hormone at the cellular level [1]. Briefly, triiodothyronine (T3) formed mainly by the outer ring peripheral deiodination of tetraiodothyronine (T4) acts through two nuclear thyroid hormone receptors (a and b) to regulate appropriate tissue-specific protein synthesis. The situation is complicated by existence of tissue specificity for the receptors and differential isoform tissue distribution. Furthermore, the derivation of T3 also varies; for example, in the brain, cellular T3 is derived from brain T4, whereas in other tissues, circulating T3 is transported to the particular cell in question. This implies that even mild overt hypothyroidism (OH) – that is, raised thyrotrophin (TSH) and low levels of T4 – might have significant effects on the efficiency of thyroid hormone action. What is subclinical hypothyroidism? SCH is defined by the presence of a raised TSH level but normal circulating total or free T4 and T3 levels, usually (but not always) in the absence of clinical symptoms, representing a clinically silent thyroid dysfunction [2]. Its estimated prevalence has been reported to range between 5% and 15% in the general population [3], and it is more prevalent in women, Caucasians, those over 50 years of
age and men and women who have a positive family history of thyroid disease. Aetiology of subclinical hypothyroidism SCH patients can be divided into those who have never received thyroid medication, the so-called endogenous SCH group, and those on medication or with SCH as the result of other iatrogenic procedures, the exogenous SCH group. Box 1 details those individuals with an increased risk for developing SCH. The most common cause of endogenous SCH is autoimmune (Hashimoto’s) thyroiditis [4], which is usually associated with raised levels of antithyroperoxidase antibodies. Inadequately treated OH is one of the most common causes of exogenous SCH, and patients receiving radiotherapy for head and neck tumours can also develop the condition. Diagnosis of subclinical hypothyroidism When diagnosing endogenous SCH it is important that concurrent acute or chronic illnesses and thyroid hormone resistance be excluded, and that there be no pharmacological interference from drugs such as lithium, amiodarone, metoclopramide or domperidone. The ‘cut-off ’ values used in TSH assays affect prevalence, with one study reporting a prevalence as high as 11.3% when a TSH value greater than 3 mU l21 was used [5], and others demonstrating a prevalence as low as 6% when using a TSH value of 7 mU l21 [6]. SCH can progress to OH, and we suggest that avoidance of this is an important benefit (vide infra) of early treatment.
Box 1. Individuals with an increased risk of developing subclinical hypothyroidism † Those with a goitre in the family (e.g. inactivating mutations in thyrotrophin) † Those with treated hyperthyroidism (Graves’ disease) † Those with a history of postpartum thyroiditis † Those with a history of neck irradiation † Those with autoimmune disorders (diabetes mellitus type 1) † Those treated with iodine-containing anti-arrhythmic agents (amiodarone) † Those treated with lithium † Those treated with immune response modulators (interferon a) † Those recovering from severe non-thyroidal illness † Those with poor compliance of thyroxine therapy
Corresponding author: J.H. Lazarus ([email protected]). http://tem.trends.com 1043-2760/03/$ - see front matter q 2003 Elsevier Ltd. All rights reserved. doi:10.1016/S1043-2760(03)00108-5
258
Opinion
TRENDS in Endocrinology and Metabolism
The assessment of the risks associated with SCH and the evaluation of the evidence available in favour of treatment can result in an informed decision about whether to proceed with treatment. The benefits of treatment of subclinical hypothyroidism are the prevention of the onset of OH, the alleviation of symptoms such as dry skin, lethargy and depression, and the reduction of both serum cholesterol levels and cardiovascular risk. Prevention of progression to OH SCH can be regarded as a stage in the development of OH [7], and should be considered as subclinical thyroid failure. An epidemiological study examining the risk of progression of SCH to OH over a 20 year period found a 4.3% risk at one year of developing OH in patients with SCH and antithyroid antibodies and a 38-fold increased risk compared with patients with a normal TSH level and no antibodies [8]. However, other studies have suggested that the risk of progression from SCH to OH is as high as 18% in patients studied over the same period during the first year after the demonstration of a raised TSH [9]. OH can (rarely) lead to extreme morbidity, with complications such as cardiac failure or myxoedema coma occurring in some patients. Alleviation of symptoms of SCH Although patients with SCH are presumed to be asymptomatic by the definition ‘subclinical’ they can have subtle clinical manifestations and non-specific symptomatology. In a previous randomized control trial (RCT) [10], dry skin, cold intolerance, constipation and easy fatigability were noted as the most common presenting symptoms, but other signs and symptoms can be vague and hard to define; because of this, they are sometimes attributed to complaints associated with normal aging. However, patients usually present to a physician for one or more reasons. Canaris et al. [11], in a cross-sectional study involving 25 862 patients, found that those with SCH had significantly more overall symptoms, such as dry skin and lethargy, than did euthyroid patients. Three double-blinded, placebo-controlled RCTs have demonstrated an improvement in symptoms with replacement T4 therapy [9,12,13]. Different methods have been used to assess response. Meier et al. [12] used the ‘hypothyroidism diagnostic index’ of Billewicz et al. [14] and the scores of Zulewski et al. [15] to demonstrate significant improvement ðP , 0:02Þ in symptoms; other RCTs have used questionnaires and have also shown improvement with T4 [10]. The greatest improvement in patients’ symptoms has been shown in those with a baseline TSH .12 mU l21 [12]. It has been suggested that mood disorders and disturbances in cognition and memory are associated with SCH [16]. The lifetime frequency of depression is significantly higher in patients with SCH compared with patients with normal thyroid function [17], suggesting that SCH lowers the threshold for depression. A case –control study has also supported these findings and shown that patients have significantly increased anxiety, somatic complaints and hysteria [18]. Other http://tem.trends.com
Vol.14 No.6 August 2003
abnormal mood disorders noted in SCH include an increased incidence of concurrent panic disorder in depressed patients with SCH [19], which might respond less well to antidepressant therapy. Another study [20] showed significant decreases in logical memory in women with SCH compared with euthyroid individuals. A significant improvement of memory skills was induced by T4 treatment in these patients [20]. Reduction of hyperlipidaemia Although OH is usually associated with dyslipidaemia, SCH has a less well recognized association. A previous review has shown that SCH is two to three times more frequent in people with raised total cholesterol (TC) levels [21]. Caraccio et al. [22] found significant increases in TC and low-density lipoprotein cholesterol (LDL-C) in patients compared with euthyroid controls, and this statistical difference has also been confirmed in other case –control studies [23,24]. Therefore, the treatment of SCH could potentially reduce serum cholesterol levels and retard the development of atherosclerosis. Three RCTs comparing T4 and placebo have shown statistically significant reductions in both plasma TC and LDL-C concentrations [12,22,25], with improvement being of greatest benefit in those with higher initial pretreatment TSH and cholesterol levels. There are only limited data on the effects of replacement T4 therapy on apolipoprotein A1, apolipoprotein B (Apo B) and lipoprotein (a). Two studies have shown significant reductions in Apo B following treatment if the initial TSH level is . 10 mU l21 [12,23]. The effects of adding 3-hydroxy-3-methylglutaryl coenzyme A reductase inhibitors (statins) in addition to replacement T4 have been assessed in postmenopausal women. A recent prospective study found that combination therapy caused significant reductions in TC, LDL-C and trigliceride, and a significant increase in HDL-C levels at three months post-treatment [26]. Reduction of cardiovascular risk Cardiac abnormalities in hypothyroidism are well documented [27,28]. More recently, several studies have investigated cardiac changes evident in SCH and even though SCH is considered as ‘mild thyroid failure’ there is evidence to show that slight changes in thyroid hormone status do cause cardiac functional abnormalities. These include impairment of left ventricular (LV) diastolic function at rest (seen as slowing of relaxation of the left ventricle) [3,29] and LV systolic dysfunction [29]. Other changes in LV morphology include increased diastolic interventricular septum thickness [3,30], posterior wall thickness and LV mass [3]. Perhaps of most concern in one study was evidence of impaired intrinsic myocardial contractility [3]. There is evidence that these abnormalities improve with T4 treatment, demonstrating that adequate thyroid hormone replacement improves cardiac output accompanied by a substantial decrease in systemic vascular resistance [31], a reversal of diastolic dysfunction [3,29], and importantly an improvement in LV ejection fraction during exercise [32]. Recently, the technique of
Opinion
TRENDS in Endocrinology and Metabolism
pulsed tissue Doppler echocardiography (used to assess regional myocardial wall velocity) has been used to show that SCH is associated with significant prolongation of the pre-ejection, left ventricular ejection and isovolumetric times [33]. This is related to the fact that ATP-dependent Ca2þ transport in the sarcoplasmic reticulum is controlled by thyroid hormone [34]. Patients with SCH have more cardiovascular risk factors and are more likely to develop atherosclerosis and other cardiovascular diseases [35]. Endothelial dysfunction, known to contribute to the development of atherosclerosis [36] and detected by the use of flow mediated arteriolar dilatation, has been shown in patients with SCH [37]. In view of the clear structural and biological cardiovascular risks associated with the presence of SCH, treatment of this condition would be expected to provide protection against the development of cardiovascular disease, although there are no published longterm outcome studies demonstrating this. Pregnancy Thyroid dysfunction in pregnancy has been associated with both maternal and foetal problems. The impairment of neuro-intellectual development in children born to mothers with SCH during pregnancy has been shown to result from low foetal access to circulating maternal T4 during the first trimester [38]. In this group of patients, it is essential that treatment is recommended and that screening programmes be introduced. Even though screening might not be performed until after conception, it is still important because the precise details of thyroid hormone on brain development have not yet been delineated. Although there are no RTCs in this area, four cohort studies have been performed [39 –42], and Klein et al. [39] reported an inverse correlation between the severity of maternal hypothyroidism and the intelligence quotient of the offspring, and reinforced the need for a screening programme. Reproductive dysfunction or ‘subfertility’ can remain unrecognized in patients with SCH, and women being investigated for infertility should have their thyroid function tested. A variable rate of infertility has been noted in patients with SCH [43]. One study found that patients with SCH have increased prolactin secretion, although the impact of SCH on gonadal function and infertility has yet to be clarified [44]. Screening Screening depends upon the perceived benefits (once SCH is detected) of treating the condition, and ideally should be based on a rigorous cost – benefit analysis. Although population-based screening programmes are generally not recommended, in certain high-risk groups there is evidence that screening is cost effective (e.g. in patients with type 1 diabetes mellitus [45], women over 40 years of age [46], pregnant women and those with a positive thyroid antibody status) [47]. These selected groups of patients (apart from pregnant women) would benefit from periodic screening (probably annually), with treatment instituted if SCH were to be documented. http://tem.trends.com
Vol.14 No.6 August 2003
259
In 2000, the American Thyroid Association recommended that adults should be screened every five years from the age of 35 years by measurement of their TSH levels [48]. We think that this option is too expensive and unproven in terms of population benefit at this time, and therefore suggest a targeted focused approach. Patient choice? Although SCH is considered to be asymptomatic and assumed to be discovered only on laboratory testing, when asked, most patients would volunteer some positive symptomatology [11]. Therefore, when treatment is decided on should patients be offered a choice? In today’s multi media world, with most people having access to the Internet, it seems inevitable that patients will demand treatment. However, it will be of greater importance to reinforce the argument for treatment with clinical evidence. Final decision The decision to treat must be based on the individual patient, once any underlying coexisting morbidity or pharmacological interference has been excluded (see Box 2 for examples that should aid in decision making). The future To date, there are few RCTs and longitudinal studies involving the treatment of SCH, and treatment studies are
Box 2. Two examples that should aid in decision making Examples (1) A 45-year-old, non-smoking woman with a free thyroxine (T4) of 12.3 mmol l21, thyroid-stimulating hormone (TSH) of 10.4 mU l21, antithyroperoxidase (TPO) antibodies at . 600 kU l21 (normal range , 32 kU l21), and total cholesterol of 6.2 mmol l21, complaining of lethargy. Would you treat? (2) A 54-year-old woman, smoker, with a T4 of 14.8 mmol l21, TSH level of 6.8 mU l21, TPO antibodies of , 5 kU l21, complaining of feeling depressed and unable to lose weight. Would you treat?
Answers (1) This patient has subclinical hypothyroidism (SCH) with positive TPO antibodies. This, together with lethargic symptoms and hypercholesterolaemia, are all potential reasons to recommend treatment with T4. (2) This patient represents more of a dilemma. Her SCH is almost not autoimmune in nature, her TSH is below 10 mU l21 and her main symptoms are depression and weight gain. It is important to exclude any comorbidity involved or pharmacological interference, especially agents used in treatment of her depression. We would certainly not wish to start T4 therapy given this clinical scenario, and would ideally wait for repeat thyroid function tests to note any documented trend in T4 concentration or upward trend in TSH. Although the inherent dangers of T4 administration to patients complaining of hypothyroid symptoms who are biochemically euthyroid are recognized by nearly all clinicians, we suggest that, in patients with SCH (as in example 2), a ‘therapeutic trial’ of T4 might be justified.
260
Opinion
TRENDS in Endocrinology and Metabolism
Vol.14 No.6 August 2003
Detection of an elevated TSH level and normal T4 and T3 levels
Repeat TFTs TPO Ab Full lipid profile (exclude renal failure, non thyroidal illness, drugs or interference from iodine-containing agents)
TPO Ab positive
TPO Ab negative
TSH level >10 mUl–1
Treat with T4
TSH level <10 mUl–1
Symptoms evident Goitre Smokers Pregnancy Type 1 DM Cadiovascular risks Ovulatory dysfunction with infertility
No symptoms or other risk factors
Repeat TFTs at six months
TRENDS in Endocrinology & Metabolism
Fig. 1. Recommendations for the treatment of patients with subclinical hypothyroidism. Adapted from Cooper [49]. Abbreviations: DM, diabetes mellitus; T3, triiodothyronine; T4, tetraiodothyronine; TFT, thyroid function test; TPO Ab, antithyroperoxidase antibodies; TSH, thyroid-stimulating hormone.
limited by the lack of placebo controls. In addition, studies undertaken to date usually involve predominately women. This, together with the small sample sizes used and variable definitions of SCH, limits the application of study findings, and it is important that future RCTs assess the benefits of replacement T4 in larger groups of patients. With regular follow up of patients and thyroid function tests performed periodically, any risks of over replacement can be avoided. We take the view that SCH should be treated, definitely in the ‘at risk’ groups of patients (Fig. 1), in view of the outcome data on cardiovascular disease and lipids. We recognize that longterm outcome data are not available in relation to many features of SCH, but they are available for lipid abnormalities. We believe that future large wellcontrolled prospective studies will, in general, confirm the positive suggestive evidence already provided by the shortterm RCTs and accurate observational studies on the abnormal clinical and laboratory findings in this common condition. References 1 Brent, G.A. (2002) Tissue specific actions of thyroid hormone: insights from animal models. Rev. Endocr. Metab. Disord. 1, 27 – 33 2 Benediktsson, R. and Toft, A.D. (1998) Management of the unexpected result: compensated hypothyroidism. Postgrad. Med. J. 74, 729 – 732 http://tem.trends.com
3 Monzani, F. et al. (2001) Effect of levothyroxine on cardiac function and structure in subclinical hypothyroidism. J. Clin. Endocrinol. Metab. 86, 1110 – 1115 4 Rapoport, B. (1991) Pathophysiology of Hashimoto’s thyroiditis and hypothyroidism. Annu. Rev. Med. 42, 91 – 96 5 Rivolta, G. et al. (1999) Prevalence of subclinical hypothyroidism in a population living in the Milan metropolitan area. J. Endocrinol. Invest. 22, 693 – 697 6 Kanaya, A.M. et al. (2002) Association between thyroid dysfunction and total cholesterol level in an older biracial population: the health, ageing and body composition study. Arch. Intern. Med. 162, 773 – 779 7 Samuels, M.H. (1998) Subclinical thyroid disease in the elderly. Thyroid 8, 803– 813 8 Vanderpump, M.P. et al. (1995) The incidence of thyroid disorders in the community: a twenty-year follow up of the Whickham survey. Clin. Endocrinol. (Oxf) 43, 55 – 68 9 Parle, J.V. et al. (1991) Prevalence and follow-up of abnormal thyrotropin (TSH) concentrations in the elderly in the United Kingdom. Clin. Endocrinol. 34, 77 – 83 10 Cooper, D.S. et al. (1984) L -Thyroxine therapy in subclinical hypothyroidism. A double blind, placebo-controlled trial. Ann. Intern. Med. 101, 18 – 24 11 Canaris, G.J. et al. (2000) The Colorado thyroid disease prevalence study. Arch. Intern. Med. 160, 526 – 534 12 Meier, C. et al. (2001) TSH controlled L -thyroxine therapy reduces cholesterol levels. J. Clin. Endocrinol. Metab. 86, 4860 – 4866 13 Nystrom, E. et al. (1988) A double blind cross-over 12 month study of L -thyroxine treatment of women with ‘subclinical’ hypothyroidism. Clin. Endocrinol. 29, 63 – 75
Opinion
TRENDS in Endocrinology and Metabolism
14 Billewicz, W.Z. et al. (1969) Statistical methods applied to the diagnosis of hypothyroidism. Q. J. Med. 38, 255– 266 15 Zulewski, H. et al. (1997) Estimation of tissue hypothyroidism by a new clinical score: evaluation of patients with various grades of hypothyroidism and controls. J. Clin. Endocrinol. Metab. 82, 771 – 776 16 Hendrick, V. et al. (1998) Psychoneuroendocrinology of mood disorders. The hypothalamic– pituitary – thyroid axis. Psychiatr. Clin. North Am. 21, 277 – 292 17 Haggerty, J. et al. (1993) Subclinical hypothyroidism: a modifiable risk factor for depression? Am. J. Psychiatry 150, 508– 510 18 Monzani, F. et al. (1993) Subclinical hypothyroidism: neurobehavioural features and beneficial effect of L -thyroxine treatment. Clin. Investig. 71, 367– 371 19 Joffe, R.T. and Levitt, A.J. (1992) Major depression and subclinical (grade 2) hypothyroidism. Psychoneuroendocrinology 17, 215 – 221 20 Baldini, I. et al. (1997) Psychopathological and cognitive features in subclinical hypothyroidism. Prog. Neuropsychopharmacol. Biol. Psychiatry 21, 925 – 935 21 Tanis, B.C. et al. (1996) Effect of thyroid substitution on hypercholesterolaemia in patients with subclinical hypothyroidism: a reanalysis of interventional studies. Clin. Endocrinol. 44, 643 – 649 22 Carracio, N. et al. (2002) Lipoprotein profile in subclinical hypothyroidism: response to levothyroxine replacement, a randomised placebocontrolled study. J. Clin. Endocrinol. Metab. 87, 1533 – 1538 23 Efstathiadou, Z. et al. (2001) A lipid profile in subclinical hypothyroidism: is L -thyroxine substitution beneficial? Eur. J. Endocrinol. 145, 705 – 710 24 Yildirimkaya, M. et al. (1996) Lipoprotein (a) concentration in subclinical hypothyroidism before and after L -thyroxine therapy. Endocr. J. 43, 731 – 736 25 Evidence Report Subclinical Thyroid Disease. (1999) Effects of TSHcontrolled thyroxine therapy on lipid profile and peripheral target tissues in patients with subclinical hypothyroidism. The Lewin Group: prepared for the Endocrine Society, pp. 45 26 Evidence Report Subclinical Thyroid Disease. (2002) The comparison of L -thyroxine versus L -thyroxine and statin combinations in lipid status in patients with subclinical hypothyroidism. The Lewin group: prepared for the Endocrine Society, pp. 51 27 Gonzales Vilchez, F. et al. (1998) Cardiac manifestations of primary hypothyroidism. Determinant factors and treatment response. Rev. Esp. Cardiol. 51, 893– 900 28 Ridgway, E.C. et al. (1982) Cardiac function in mild and severe primary hypothyroidism. Life Sci. 30, 651 – 658 29 Biondi, B. et al. (1999) Left ventricular dysfunction in patients with subclinical hypothyroidism. J. Clin. Endocrinol. Metab. 84, 2064 – 2067 30 Varma, R. et al. (1996) Heart in hypothyroidism – an echocardiographic study. J. Assoc. Physicians India 44, 390 – 392 31 Faber, J. et al. (2002) Haemodynamic changes after levothyroxine treatment in subclinical hypothyroidism. Thyroid 12, 19 – 24
Vol.14 No.6 August 2003
261
32 Forfar, J.C. et al. (1985) Left ventricular performance in subclinical hypothyroidism. Q. J. Med. 57, 857 – 865 33 Vitale, G. et al. (2002) Left ventricular myocardial impairment in subclinical hypothyroidism assessed by a new ultrasound tool: pulsed tissue doppler. J. Clin. Endocrinol. Metab. 87, 4350– 4355 34 Warnick, P.R. et al. (1993) Rabbit skeletal muscle sarcoplasmic reticulum Ca(2 þ )-ATPase activity: stimulation in vitro by thyroid hormone analogues and bipyridines. Biochim. Biophys. Acta 1153, 184– 190 35 Staub, J.J. et al. (1993) Spectrum of subclinical and overt hypothyroidism: effect on thyrotropin, prolactin and thyroid reserve and metabolic impact on peripheral target tissues. Am. J. Med. 92, 631– 642 36 Kuikka, J.T. et al. (2001) Imaging of the endothelial dysfunction in coronary atherosclerosis. Eur. J. Nucl. Med. 28, 1567 – 1578 37 Lekakis, J. et al. (1997) Flow mediated endothelium-dependent vasodilation is impaired in subjects with hypothyroidism, borderline hypothyroidism and high normal serum thyrotropin (TSH). Thyroid 7, 411 – 414 38 Calvo, R. et al. (2002) Fetal tissues are exposed to biologically relevant free thyroxine concentrations during early phases of development. J. Clin. Endocrinol. Metab. 87, 1768– 1777 39 Klein, R.Z. et al. (2001) Relation of severity of maternal hypothyroidism to cognitive development of offspring. J. Med. Screen. 8, 18 – 20 40 Abalovich, M. et al. (2002) Overt and subclinical hypothyroidism complicating pregnancy. Thyroid 12, 63 – 68 41 Radetti, G. et al. (2000) Psychomotor and audiological assessment of infants born to mothers with subclinical thyroid dysfunction in early pregnancy. Minerva Pediatr. 52, 691 – 698 42 Kuijpens, J.L. et al. (1998) Prediction of postpartum thyroid dysfunction: can it be improved? Eur. J. Endocrinol. 139, 36 – 43 43 Glinoer, D. (1997) The regulation of thyroid function in pregnancy: pathways of endocrine adaptation from physiology to pathology. Endocr. Rev. 18, 404 – 433 44 Tolina, A. et al. (1991) Subclinical hypothyroidism and hyperprolactinaemia. Acta Eur. Fertil. 22, 275 – 277 45 Perros, P. et al. (1995) Frequency of thyroid dysfunction in diabetic patients: value of annual screening. Diabet. Med. 12, 622– 627 46 Wiersinga, W.M. (1995) Subclinical hypothyroidism and hyperthyroidism. I. Prevalence and clinical relevance. Neth. J. Med. 46, 197 – 204 47 Glinoer, D. et al. (1994) Risk of subclinical hypothyroidism in pregnant women with asymptomatic autoimmune thyroid disorders. J. Clin. Endocrinol. Metab. 79, 197– 204 48 Ladenson, P.W. et al. (2000) American Thyroid Association guidelines for detection of thyroid dysfunction. Arch. Intern. Med. 160, 1573– 1575 49 Cooper, D.S. et al. (2001) Subclinical hypothyroidism. N. Engl. J. Med. 345, 260 – 265
Conference Reporter AACR 2003 Are you unable to attend the 94th Annual Meeting of the American Association for Cancer Research in Washington DC on 11–14 July 2003? Don’t miss out! Sign up for daily reports from the conference at: http://news.bmn.com/alerts Each day’s top stories, written by the BioMedNet Conference Reporter team, can be sent direct from the conference to your inbox in either text or html format. http://tem.trends.com