Effect of Levothyroxine on Left Ventricular Ejection Fraction in Patients With Subclinical Hypothyroidism and Acute Myocardial Infarction: A Randomized Clinical Trial

Abstract

Importance: Thyroid hormones play a key role in modulating myocardial contractility. Subclinical hypothyroidism in patients with acute myocardial infarction is associated with poor prognosis.

Objective: To evaluate the effect of levothyroxine treatment on left ventricular function in patients with acute myocardial infarction and subclinical hypothyroidism.

Design, setting, and participants: A double-blind, randomized clinical trial conducted in 6 hospitals in the United Kingdom. Patients with acute myocardial infarction including ST-segment elevation and non-ST-segment elevation were recruited between February 2015 and December 2016, with the last participant being followed up in December 2017.

Interventions: Levothyroxine treatment (n = 46) commencing at 25 µg titrated to aim for serum thyrotropin levels between 0.4 and 2.5 mU/L or identical placebo (n = 49), both provided in capsule form, once daily for 52 weeks.

Main outcomes and measures: The primary outcome measure was left ventricular ejection fraction at 52 weeks, assessed by magnetic resonance imaging, adjusted for age, sex, type of acute myocardial infarction, affected coronary artery territory, and baseline left ventricular ejection fraction. Secondary measures were left ventricular volumes, infarct size (assessed in a subgroup [n = 60]), adverse events, and patient-reported outcome measures of health status, health-related quality of life, and depression.

Results: Among the 95 participants randomized, the mean (SD) age was 63.5 (9.5) years, 72 (76.6%) were men, and 65 (69.1%) had ST-segment elevation myocardial infarction. The median serum thyrotropin level was 5.7 mU/L (interquartile range, 4.8-7.3 mU/L) and the mean (SD) free thyroxine level was 1.14 (0.16) ng/dL. The primary outcome measurements at 52 weeks were available in 85 patients (89.5%). The mean left ventricular ejection fraction at baseline and at 52 weeks was 51.3% and 53.8%, respectively, in the levothyroxine group compared with 54.0% and 56.1%, respectively, in the placebo group (adjusted difference in groups, 0.76% [95% CI, -0.93% to 2.46%]; P = .37). None of the 6 secondary outcomes showed a significant difference between the levothyroxine and placebo treatment groups. There were 15 (33.3%) and 18 (36.7%) cardiovascular adverse events in the levothyroxine and placebo groups, respectively.

Conclusions and relevance: In this preliminary study involving patients with subclinical hypothyroidism and acute myocardial infarction, treatment with levothyroxine, compared with placebo, did not significantly improve left ventricular ejection fraction after 52 weeks. These findings do not support treatment of subclinical hypothyroidism in patients with acute myocardial infarction.

Trial registration: isrctn.org Identifier: http://www.isrctn.com/ISRCTN52505169.

Figure 1.. Flow of Participants Through the Thyroxine in Acute Myocardial Infarction (ThyrAMI-2) Randomized Clinical Trial

MRI indicates magnetic resonance imaging; TSH, serum thyrotropin. ^aAge older than 75 years was an exclusion criteria for the first 3 months of recruitment until the protocol was amended.

Figure 2.. Effect of Levothyroxine Compared With Placebo on Left Ventricular Ejection Fraction (LVEF) According to Prespecified and Post Hoc Subgroups

AMI indicates acute myocardial infarction; FT₄, free thyroxine; FT₃, free triiodothyronine; NSTEMI, non–ST-elevation myocardial infarction; STEMI, ST-elevation myocardial infarction. To convert FT₄ to pmol/L, multiply by 12.87; and FT₃ to pmol/L, multiply by 1.54. ^aThe difference in LVEF between the levothyroxine and placebo groups was adjusted for baseline age, sex, type of AMI, baseline LVEF, and infarct territory. ^bThe P for interaction was calculated by the addition of the interaction term (independent variable × allocation to levothyroxine or placebo) to the fully adjusted model above. ^cPost hoc subgroup.