Consequences of subclinical hyperthyroidism
	Symptoms of Subclinical Hyperthyroidism While subclinical hyperthyroidism is usually assumed to be asymptomatic, there are a few studies which suggest it may be possible to elicit subtle physical and psychiatric symptoms in these patients. Overtly hyperthyroid patients may have higher rates of depression than their euthyroid counterparts, particularly elderly patients who may present with so-called “apathetic hyperthyroidism“16,17 . A study from the Netherlands found that affective disorders were significantly more prevalent in patients with suppressed TSH levels, most of whom had subclinical rather than overt hyperthyroidism 18 . Schlote-Sautter et al. used several standardized instruments to assess for signs and symptoms of hyperthyroidism in 20 overtly hyperthyroid patients, 15 patients with subclinical hyperthyroidism, and 5 euthyroid controls 19 . They found that subclinically hyperthyroid subjects had more symptoms than the euthyroid patients, though fewer than the overtly hyperthyroid group. These symptoms consisted primarily of tremor, palpitations, and fatigue. The subclinically hyperthyroid patients also resembled overtly hyperthyroid patients in being significantly more depressed than the euthyroid control group. However, when the same group studied workers from a large chemical plant, they found no differences between subclinically hyperthyroid subjects and controls in reported symptoms or in depression scores 20 . Bone Women in the immediate postmenopausal years are at high risk for losing bone density. Up to 3 % of bone mass is commonly lost within each of the first 5 years after menopause 21 . It is estimated that 30 % of Caucasian women in the US have osteoporosis 22 . Concomitant thyrotoxicosis, whether subclinical or overt, has been shown to exacerbate this risk for postmenopausal osteoporosis. In thyrotoxicosis there is an increase in both osteoblast and osteoclast activity. This results in increased bone turnover as measured by increased urine N–telopeptide and osteocalcin, with a net increase in bone resorption 23,24 . In overtly thyrotoxic patients, the increased bone resorption may occasionally lead to hypercalcemia 25 and more commonly to decreased bone density 26,27 . Cortical bone is affected more severely than trabecular bone. Multiple studies of the effects of subclinical hyperthyroidism on bone have provided conflicting results since the first study by Ross et al. demonstrating reduced bone density in patients receiving TSH-suppressive doses of L-T 4 28 . Grant et al. found no significant differences in forearm bone mineral density between postmenopausal women on long-term TSH-suppressive doses of L-T 4 , those on replacement L-T 4 doses, and euthyroid controls 29 . However, in a large study of postmenopausal women, Schneider et al. found significantly lower bone density at multiple skeletal sites in women taking > 200 mcg L-T 4 daily versus those taking smaller doses of thyroid hormone 30 . Studies by Lehmke et al. and Kung et al. found lower bone densities in postmenopausal women with subclinical hyperthyroidism secondary to TSH-suppressive L-T 4 therapy 31,32 , Campbell et al. showed increased markers of bone turnover in hypothyroid postmenopausal women treated with suppressive versus replacement doses of L-T 4 33 . L-T 4 therapy does not seem to have a deleterious effect on bone density, however, when given in replacement, rather than suppressive, doses. In fact, in a 2-year prospective study, Guo et al. were able to show that bone density increased and markers of bone turnover decreased after postmenopausal women were switched from suppressive to replacement doses of L-T 4 35 . Endogenous subclinical hyperthyroidism seems to confer the same risks as suppressive L-T 4 doses. Mudde et al. found that women with subclinical hyperthyroidism due to untreated multinodular goiter had significantly lower bone densities at the forearm than age-and menopausal-status matched euthyroid controls 36 . While overt hyperthyroidism is known to be associated with increased fracture risk, data are less clear for subclinical hyperthyroidism despite the known reductions in bone density and increase in markers of bone turnover. In a Scottish study, primarily of post-menopausal women, no differences in the rate of fractures requiring hospitalization between hypothyroid patients treated with replacement doses of L-thyroxine versus those treated with TSH-suppressive doses were seen over an average 8.6 years of follow-up 37 . A recent large prospective study found that postmenopausal women over age 65 with serum TSH levels =0.1 mIU/L at baseline had increased risk for both vertebral and hip fracture over a mean follow-up period of 3.7 years. T 3 and T 4 were not measured in this study, so it is not clear how many of these women had overt versus subclinical hyperthyroidism. The authors did note that thyroid hormone use was not associated with increased fracture risk if the serum TSH values were normal 38 . In men and premenopausal women , the effects of subclinical hyperthyroidism on bone are of uncertain significance. Marcocci et al. reported that there was no evidence of decreased bone mineral density in men receiving TSH-suppressive doses of L-T 4 39 . Lecomte et al. found a significant increase in serum osteocalcin between premenopausal women on suppressive L-T 4 therapy and euthyroid controls 40 . Földes et al. found that bone density of the lumbar spine, femoral neck, and midshaft radius was not significantly decreased in premenopausal women with endogenous subclinical hyperthyroidism 41 . While Harvey et al. documented increased urinary markers of bone turnover in postmenopausal women on suppressive doses of L-T 4 , they did not find any increase in bone turnover in premenopausal women on similar L-T 4 doses 24 . Similarly, Gurlek and Gedik did not find any increase in bone turnover, or decrease in bone density, in a sample of premenopausal women with endogenous subclinical hyperthyroidism when compared to euthyroid controls 42 although an older study by Paul et al. reported a small but significant decrease in hip but not spine bone mineral density in TSH-suppressed premenopausal women 43 . It is evident that there are some discrepancies in the literature on the effects of TSH-suppressive doses of L-T 4 on bone in pre- and postmenopausal women. Two large meta-analyses reached similar conclusions that TSH-suppressive doses of L-T 4 decrease bone mineral density in postmenopausal but not in premenopausal women 44,45 . Postmenopausal or perimenopausal women with subclinical hyperthyroidism may be particularly good candidates for estrogen or bisphosphonate therapy, which can prevent the thyroid-associated bone loss. Rosen et al. reported that cyclic intravenous pamidronate increased bone density in patients receiving TSH-suppressive doses of L–T 4 46 . Women taking both estrogen replacement therapy and L-T 4 have been shown to have significantly higher bone densities than those taking L-T 4 alone 47 . Lecomte et al. showed no difference in bone turnover or bone mineral density between postmenopausal women on TSH-suppressive doses of L-T 4 and estrogen replacement therapy compared to euthyroid controls 40 . Cardiac Effects Overt thyrotoxicosis, especially when associated with tachycardia or atrial arrhythymias, has long been known to aggravate underlying cardiac disease, worsening angina secondary to increased myocardial oxygen demand. Thyrotoxicosis leads to increased cardiac contractility at rest, although there may be impaired ventricular function with exertion. Tachycardia decreases diastolic coronary perfusion time 48 . All of these factors can contribute to the development of cardiomyopathy and congestive heart failure. Significant cardiac dysfunction may also result from subclinical hyperthyroidism. A recent review summarizes the effects of thyroid hormones on the cardiovascular system 49 . Biondi et al. examined cardiac function and exercise tolerance in a group of 10 subjects, all of whom had been treated with TSH-suppressive doses of L-T 4 for at least 5 years, and all of whom complained of exertional dyspnea 50 . They showed impaired resting left ventricular diastolic filling, increased exertional left ventricular ejection fraction, and significantly reduced exercise capacity in the TSH-suppressed group when compared to controls. Notably, these findings all improved after 4 months of beta-blocker therapy. Another study by the same group noted significantly higher average heart rates, a higher prevalence of atrial premature beats, and left ventricular hypertrophy on echocardiography in patients on longterm TSH-suppressive L-T 4 therapy when compared to controls 51 . However, when Shapiro et al. looked at a group of 17 asymptomatic patients on TSH-suppressive L-T 4 therapy they found no difference between these subjects and controls in mean heart rate, prevalence of atrial premature contractions, or mean ejection fraction. Left ventricular mass index was significantly increased in the L-T 4 -treated group, although still within normal limits 52 . Tenerz et al. studied 40 patients with subclinical hyperthyroidism compared to a control group and found that atrial fibrillation was present in 28 % of the subclinical thyrotoxicosis group as compared to 10 % of the controls 53 . Data from the Framingham Heart Study demonstrated that, over 10 years of follow up in patients over age 60, the cumulative incidence of atrial fibrillation was 28 % in those with low serum TSH values (< 0.1) at baseline, as compared to 11 % of patients with normal TSH levels. Only 2 of the 13 patients with suppressed TSH values in this study who developed atrial fibrillation also developed overt hyperthyroidism during the follow-up period; the rest were subclinically hyperthyroid 54 . This finding has important clinical implications, as thyrotoxicosis-associated atrial fibrillation may be complicated by thromboembolism in up to 15 % of patients 55 .