Subclinical hyperthyroidism and fracture risk in women.

Abstract

Although overt thyrotoxicosis is known to cause loss of bone mineral density (BMD), osteoporosis, and propensity to fracture, it remains controversial whether similar risk applies to mild states of T4 excess with normal reference range T4 and T3 levels but suppressed TSH levels, ie, “subclinical” hyperthyroidism (1–7). That controversy is perhaps further fueled by two papers appearing in this issue of the JCEM, one by Garin et al (8) indicating no greater risk of bone mineral loss or hip fracture in older men or women, and the other by Leader et al (9) concluding that there is an increased risk of hip fracture with subclinical hyperthyroidism, at least in older women but not men. How can we reconcile the differing conclusions of these two carefully performed studies? There is a body of literature indicating that suppressive doses of T4 in certain settings can be associated with loss of BMD (3). The basis for differing reports in the literature on this issue may be due to a variety of potentially confounding parameters including: study design (cross-sectional vs longitudinal), BMD methodology, patient population characteristics (age, race, ethnicity, sex, body mass index, menopausal status), type or duration of T4 therapy (replacement or suppressive), and other confounding factors known to alter BMD (corticosteroid use, smoking, thiazide diuretics, nonexercise, prior thyrotoxicosis, history of use of bone-enhancing therapies, eg, bisphosphonates, and possibly, prior thyroidectomy with calcitonin deficiency). Some of these confounders apply to the populations studied in the two abovementioned reports. Apropos the current flurry of interest in vitamin D is a recent study indicating that concomitant vitamin D deficiency in hyperthyroid subjects is another factor associated with lower BMD (10). In comparing the patient populations in the two current JCEM reports, it is often unclear how well these potentially confounding effects were controlled. Circulating T4 concentrations will affect bone remodeling, a normal ongoing process of cycles of bone activation-resorption-formation (11). The cycles are shortened in hyperthyroidism, leading to an accelerated rate of bone turnover. The ultimate clinical expression of these shortened activation-resorption-formation cycles and increasedbone turnover (eg, osteopeniaand/or fracture)will depend upon the severity and duration of the thyroid hormone excess. The imagination is not stretched to envision the same effect in subclinical hyperthyroidism, albeit to a milder or attenuated degree, presumably predicated on the degree and duration of relative thyroid hormone excess. In contrast to the situation in subclinical hyperthyroidism, higher TSH levels have been associated with reduced bone remodeling and turnover (12, 13). The effects of physiologically excessive circulating T4 or T3 levels are intimately linked as well to the estrogen status in periand postmenopausal women (14). Similar to the effect of thyroid hormone, estrogen deficiency is associated with cytokinestimulated osteoclast formation with increased rates of bone resorption (15). Thus, the adverse effect on bone of mild excesses of thyroid hormone will be augmented by concomitant estrogen deficiency, and the preponderance of data suggests that treatment with suppressive doses of levothyroxine does not materially affect BMD in premenopausal women or men but will do so in postmenopausal women (3, 16, 17), although this has not been noted in all studies (18). However, earlier studies did not indicate an increased risk of fracture in patients with subclinical hyperthyroidism (19, 20), and it has been suggested that when fractures are seen, they could be related to a prior history of overt