Thyroid and the Heart: The Intimacy Is Strained

D ating back to the earliest description by Caleb Parry and Robert Graves, it is well known that thyroid hormone exerts profound effects on the heart and cardiovascular system (1). Acting through a classical nuclear mechanism of action, triiodothyronine (T₃), but not T₄, can alter myocyte expression of important cardiac structural and regulatory proteins (2). As seen in Table 1, the increased expression of the fast isoform of the myosin heavy chain (α), sarcoplasmic reticulum calcium-activated ATPase (3), and β1-adrenergic receptor density would all be expected to enhance both systolic and diastolic left ventricular (LV) function. In contrast, and in the absence of thyroid hormone, a shift to the slow myosin heavy chain isoform (β) with a decrease in SERCA2 and an increase in its inhibitory protein phospholamban (4) would be expected to impair cardiac function. The clinical correlation for these changes in both hyper- and hypothyroidism have been previously well documented (5,6). In addition to these cardiovascular hemodynamic changes of overt thyroid disease, subclinical thyroid disease can also change cardiac contractile performance. From the earliest studies of Biondi and colleagues (7), it has been well established that subclinical hyperthyroidism can alter heart rate and rhythm and cardiac mass. Subclinical hypothyroidism can also alter cardiac contractility including LV diastolic function, diastolic blood pressure, and systemic vascular resistance in a manner potentially leading to heart failure (8). Studies designed to further assess the underlying mechanisms for these changes are of obvious clinical importance.

In the current issue of Thyroid the Endocrinology and Cardiology Sections at Leiden University have shown the presence of decreased strain in the LV myocardium of thyroid cancer patients receiving thyrotropin (TSH)-suppressive doses of T₄ using a state of the art speckle tracking echocardiographic technique (9). While two-dimensional speckle tracking can be used to measure both changes in systolic function as well as diastolic function, the authors currently report a change in the former, while their prior studies have shown the expected impairment in diastolic function in this same patient population studied by standard Doppler echocardiography. An interesting point in this current study is the observation that the impaired strain with subclinical hyperthyroidism appears to reverse with thyroid hormone withdrawal (partial thyroid hormone normalization) and then reappear after a longer period of drug discontinuance contemporaneous with the development of chemical hypothyroidism. While the significance of these findings remains to be determined (and the authors concur), the relevance to practicing thyroidologists can be explored. Firstly, the term strain may be misleading to a reader not familiar with the field of echocardiography. It refers to the magnitude of ventricular size change (%) during both contraction and relaxation. Thus a lower percentage implies impaired function. While initially assessed in patients with hypertensive, ischemic, and myopathic heart disease, it may have implications in the realm of thyroid disease. The ability of thyroid hormone (T₃) to regulate cardiac gene expression would suggest that patients with subclinical hyperthyroidism should have enhanced cardiac contractility (1). However, as shown previously, the increase in heart rate that accompanies a suppressed TSH may, of and by itself, impair LV strain (7). If this or the pro-arrhythmic effects of subclinical hyperthyroidism are to put our thyroid cancer patients at risk for increased cardiac mortality, as reported by Parle et al. (10) for subclinical hyperthyroid patients in general, then treatment strategies to reduce this risk may be desirable. This would include lowering the dose of levothyroxine to relax the degree of TSH suppression into the range of 0.1 to 0.4 mIU/mL with continued thyroglobulin monitoring (11). An alternative would be to initiate β-adrenergic blockade with any of a number of the currently used drugs to both lower heart rate and decrease cardiac mass (12). It would be important to then document that such treatment could reverse the alterations in strain currently observed by Abdulrahman et al. (9).

Secondly, the finding that overt hypothyroidism also lowers LV strain further supports the accumulating body of evidence that both mild (subclinical) as well as overt hypothyroidism can lead to heart failure. In older patients it is well established that hypothyroidism is associated with increased cardiac morbidity, and the demonstration of lower LV strain in this patient population should not be interpreted as a U-shaped curve but rather the emergence of impaired cardiac function of the type that would be predicted by changes in ventricular phenotype (1). Indeed, in contrast to the hyperthyroid cardiovascular system, the hypothyroid heart faces an increase in diastolic blood pressure, afterload, and systemic vascular resistance, all of which can independently impair pump function. The therapeutic implication for this set of findings is clearly a need for thyroid hormone replacement to restore the cardiovascular system to a euthyroid state. While this is well documented for overt hypothyroidism and for patients with TSH levels >10, further studies will be required to demonstrate the appropriateness of such treatment in patients with milder disease. A recent study from Razvi et al. (13) certainly supports this suggestion and ratifies the notion that an intimate relationship between the heart and the cardiovascular system endures.