OBJECTIVE: To assess whether cervical cerclage in women deemed to be at increased risk of cervical incompetence prolongs pregnancy and thereby improves fetal and neonatal outcome. DESIGN: Multicentre randomised controlled trial. SETTING: Hospitals in the United Kingdom, France, Hungary, Norway, Italy, Belgium, Zimbabwe, South Africa, Iceland, Ireland, the Netherlands and Canada. SUBJECTS: One thousand two hundred and ninety-two pregnant women whose obstetricians were uncertain whether to recommend cervical cerclage, most of whom had a history of early delivery or cervical surgery. INTERVENTIONS: Cervical cerclage was compared with a policy of withholding the operation unless it was considered to be clearly indicated. MAIN OUTCOME MEASURES: Delivery before 33 completed weeks, preterm delivery (