Reversible diastolic dysfunction after long-term exogenous subclinical hyperthyroidism: a randomized, placebo-controlled study

Subclinical hyperthyroidism has been reported to affect systolic and diastolic cardiac function. However, the reversibility of these effects is not well established. Our objective was to investigate the presence and reversibility of cardiac abnormalities in patients with long-term exogenous subclinical hyperthyroidism. We conducted a prospective, single-blinded, placebo-controlled randomized trial of 6 months duration with two parallel groups. The study occurred at the Leiden University Medical Center, a tertiary referral center for thyroid carcinoma. As a model for subclinical hyperthyroidism, 25 patients with a history of differentiated thyroid carcinoma with more than 10 yr of TSH suppressive therapy with L-T4 were studied. L-T4 dose was replaced by study medication containing L-T4 or placebo. Medication was titrated in a single-blinded fashion to establish continuation of TSH suppression (low-TSH group) or euthyroidism (euthyroid group). We assessed serum levels of free T4 and TSH and used echo Doppler cardiography including tissue Doppler to establish left ventricular (LV) dimensions and function as well as diastolic function. Baseline echocardiography data were compared with 24 controls. There were no differences in baseline cardiac parameters and TSH levels between the two groups. Although mean LV mass index was increased as compared with 24 controls, only four patients had LV hypertrophy at baseline. This was not improved by restoration of euthyroidism. At baseline, diastolic function was impaired in all patients as indicated by abnormal values for the peak flow of the early filling phase (E, 55.3 +/- 9.5 mm/sec), the ratio of E and the peak flow of the atrial filling phase (E/A ratio, 0.87 +/- 0.13), the early diastolic velocity obtained by tissue Doppler (E', 5.7 +/- 1.3 cm/sec), and the peak atrial filling velocity obtained by tissue Doppler (A', 6.8 +/- 1.4 cm/sec), prolonged E deceleration time (234 +/- 34 msec), and isovolumetric relaxation time (121 +/- 15 msec). After 6 months, significant improvements were observed in the euthyroid group in the E/A ratio (+41%; P <0.001), E deceleration time (-18%; P = 0.006), isovolumetric relaxation time (-25%; P <0.001), E' (+31%; P <0.001), and the E'/A' ratio (+40%; P