Background Thyroid autoantibodies, specifically thyroid peroxidase antibodies, have been associated with miscarriage and pre-term birth in women with a normal thyroid function. Small randomised controlled trials have found that treatment with levothyroxine may reduce such adverse outcomes in pregnancy. Objectives The Thyroid AntiBodies and LEvoThyroxine (TABLET) trial was conducted to explore the effects of levothyroxine in euthyroid women with thyroid peroxidase antibodies. A concurrent mechanistic study was conducted to examine the effect of levothyroxine on immune responses. Design This was a randomised, double-blind, placebo-controlled, multicentre study. Setting The TABLET trial was conducted in 49 hospitals across the UK between 2011 and 2016. Participants Euthyroid women who tested positive for thyroid peroxidase antibodies, were aged between 16 and 41 years and were trying to conceive either naturally or through assisted conception were eligible. Intervention Participants were randomised to levothyroxine at a dose of 50 µg daily or placebo. The intervention was commenced preconception and continued until the end of a pregnancy. Women were given a 12-month period to conceive from randomisation. Main outcome measures The primary outcome was live birth at ≥ 34 completed weeks of gestation. The secondary outcomes included miscarriage at < 24 weeks; clinical pregnancy at 7 weeks; ongoing pregnancy at 12 weeks; gestation at delivery; birthweight; appearance, pulse, grimace, activity and respiration (Apgar) scores; congenital abnormalities; and neonatal survival at 28 days of life. Methods Participants were randomised in a 1 : 1 ratio. Minimisation was implemented for age (< 35 or ≥ 35 years), number of previous miscarriages (0, 1 or 2, ≥ 3), infertility treatment (yes/no) and baseline thyroid-stimulating hormone concentration (≤ 2.5 or > 2.5 mlU/l) to achieve balanced trial arms. Women were followed up every 3 months while trying to conceive to check thyroid function and general well-being, and, once pregnant, were seen each trimester: 6–8 weeks, 16–18 weeks and 28 weeks. Any abnormal thyroid results were managed in line with clinical guidance at the time. Results Of the 19,556 women screened, 1420 women were eligible and 952 were randomised to receive levothyroxine (n = 476) or placebo (n = 476). Six women from each arm either were lost to follow-up or withdrew from the trial. A total 540 women became pregnant: 266 in the levothyroxine arm and 274 in the placebo arm. The live birth rate was 37% (176/470) in the levothyroxine group and 38% (178/470) in the placebo group, translating to a relative risk of 0.97 (95% confidence interval 0.83 to 1.14; p = 0.74) and an absolute risk difference of –0.4% (95% confidence interval –6.6% to 5.8%). A subset of 49 trial participants (26 in the levothyroxine arm and 23 in the placebo arm) were recruited to assess changes in their serum chemocytokine concentrations. Treatment with levothyroxine resulted in some changes in chemocytokine concentrations in the non-pregnant state and in early pregnancy, but these had no association with clinical outcome. Conclusions Levothyroxine therapy in a dose of 50 µg per day does not improve live birth rate in euthyroid women with thyroid peroxidase antibodies. Limitations Titration of the levothyroxine dose based on thyroid-stimulating hormone/thyroid peroxidase concentrations was not explored. Future work Future research could explore the efficacy of levothyroxine administered for the treatment of subclinical hypothyroidism. Trial registration Current Controlled Trials ISRCTN15948785 and EudraCT 2011-000719-19. Funding This project was funded by the Efficacy and Mechanism Evaluation programme, a Medical Research Council and National Institute for Health Research partnership.