The addition of thyrotropin-releasing hormone (TRH) to antenatal glucocorticoid treatment of women at risk of preterm delivery has been reported to lower the risk of respiratory distress syndrome (RDS) in the infant. We have assessed the efficacy of 200 micrograms TRH in a multicentre randomised trial. 1234 women at 24 weeks to 31 weeks 6 days of gestation with a singleton or twin pregnancy and at risk of preterm delivery were randomly allocated to groups receiving 200 micrograms TRH or placebo intravenously every 12 h up to a maximum of four doses. Randomisation was stratified by duration of gestation and centre. All women received glucocorticoids. Clinical outcome is known for 1231 women and their 1397 infants. The frequencies of the main prespecified study outcomes RDS (relative risk 1.17 [95% CI 1.00-1.36], p = 0.05) and need for ventilation (1.15 [1.01-1.31], p = 0.04) were higher in TRH-group infants than in control infants. The excess risk in the TRH group was greater in infants who were born more than 10 days after treatment. Multivariate analysis adjusting for duration of gestation at randomisation, time from randomisation to delivery, parity, history of perinatal death, and infant's sex did not affect the risk estimates. TRH administration was associated with increased risks of maternal nausea, vomiting, lightheadedness, and a rise in blood pressure to 140/90 mm Hg or higher. Antenatal TRH given with glucocorticoids to women at high risk of preterm delivery is associated with maternal and perinatal risks and cannot be recommended for widespread clinical use.