Introduction Preterm birth complicates >15 million pregnancies annually worldwide. In many countries, women who present with signs of preterm labour are treated with tocolytics for 48 hours. Although this delays birth, it has never been shown to improve neonatal outcome. In 2015, the WHO stated that the use of tocolytics should be reconsidered and that large placebo-controlled studies to evaluate the effectiveness of tocolytics are urgently needed. Methods and analysis We designed an international, multicentre, randomised, double-blinded, placebo-controlled clinical trial. Women with threatened preterm birth (gestational age 30-34 weeks), defined as uterine contractions with (1) a cervical length of < 15 mm or (2) a cervical length of 15-30 mm and a positive fibronectin test or (3) in centres where cervical length measurement is not part of the local protocol: a positive fibronectin test or insulin-like growth factor binding protein-1 (Actim-Partus test) or (4) ruptured membranes, will be randomly allocated to treatment with atosiban or placebo for 48 hours. The primary outcome is a composite of perinatal mortality and severe neonatal morbidity. Analysis will be by intention to treat. A sample size of 1514 participants (757 per group) will detect a reduction in adverse neonatal outcome from 10% to 6% (alpha 0.05, beta 0.2). A cost-effectiveness analysis will be performed from a societal perspective. Ethics and dissemination This study has been approved by the Research Ethics Committee (REC) of the Amsterdam University Medical Centres, location AMC, as well as the REC's in Dublin and the UK. The results will be presented at conferences and published in a peer-reviewed journal. Participants will be informed about the results. Trial registration number Nederlands Trial Register (Trial NL6469).