OBSTETRICS Atosiban was more effective than placebo in delaying delivery for up to 7 days in women with preterm labour Romero R, Sibai BM, Sanchez-Ramos L, et al. An oxytocin receptor antagonist (atosiban) in the treatment of preterm labor: a randomized, double-blind, placebo-controlled trial with tocolytic rescue. Am J Obstet Gynecol 2000; 182: 1173}1183. OBJECTIVE To determine the efficacy and safety of atosiban for the treatment of pre-term labour. DESIGN Multicentre, randomized, double-blind, placebo-controlled trial. Allocation was computer- generated, blocked, stratified by centre, and carried out at each centre’s pharmacy using sealed envelopes. The study had sufficient power to detect a 30% increase in time to delivery or therapeutic failure. SETTING Thirty-seven centres in the USA. SUBJECTS Five hundred and one women in preterm labour with intact membranes, cervical dilatation 43 cm, and live fetus(es) of gestational age 20 to (34 (mean 31) weeks. 39% of women had Bishop score 54. 16% of pregnancies were multiple gestation. Fifteen additional randomized women in each group did not receive treatment and were excluded from the analysis. INTERVENTION Two hundred and forty-six women were randomized to receive atosiban (6.75 mg i.v. over 1 min, followed by infusion at 300 lg/min for 3 h, then 100 lg/min for up to 45 h) and 255 women to receive placebo. Treatment was discontinued at the occurrence of uterine quiescence, progression of labour after at least 1 h treatment (in which case use of a tocolytic other than atosiban was permitted), or rupture of membranes. Women who achieved uterine quiescence received subcutaneous maintenance therapy with the assigned medication. Additional episodes of preterm labour were treated intravenously with the assigned medication. MAIN OUTCOME MEASURES Time to delivery or therapeutic failure requiring an alternate tocolytic, proportion of women undelivered at 24 h, 48 h, and 7 days, neonatal outcomes. Efficacy analysis was by intention-to-treat, but safety analysis was by drug actually received (3% of subjects received the wrong drug or both drugs). MAIN RESULTS The median (range) time to delivery or therapeutic failure was 25.6 (0}116) days in the atosiban group and 21.0 (0}111) days in the placebo group (P"0.6). For subjects enrolled at gestational age 528 weeks, the proportion of women who remained undelivered without receiving an alternate tocolytic was significantly higher in the atosiban group than the placebo group at all 3 time points: at 24 h, 74 vs 58% (absolute treatment effect [ATE] 16%, 95% CI 7}25, number needed to treat [NNT] to achieve one additional undelivered pregnancy is 6*); at 48 h, 69 vs 55% (ATE 14%, CI 4-23, NNT 7*); at 7 days 65 vs 48% (ATE 17%, CI 7-26, NNT 6*). For subjects enrolled at gestational age (28 weeks (17% of the atosiban group and 13% of the placebo group), atosiban did not provide a significant advantage over placebo at any time point. Alternate tocolytics were used in 42% of women who received atosiban and in 51% of those who received placebo (P"0.04*). There was no significant difference between groups in the proportions that received antibiotics (52 vs 46%) or corticosteroids (46 vs 51%). The fetal/infant mortality rate was 4.5% with atosiban and 1.7% with placebo (P"0.049* ), with all the difference occurring in the subgroup enrolled at (26 weeks gestation (37% vs 0%, respectively, P"0.007*). There was no significant difference between groups in the proportion of women that delivered at (37 weeks gestation (58 vs 51%), in mean birthweight, or in proportions of infants that were admitted to neonatal intensive care (42 vs 38%) or had respiratory distress syndrome (23 vs 18%) or intraventricular hemorrhage (7 vs 8%). CONCLUSION Atosiban did not significantly extend the time to delivery or therapeutic failure in women with preterm labour. However, atosiban was more effective than placebo in delaying delivery for up to 7 days, especially in women at 528 weeks gestation. * Calculated from data in article. ^ 2001 Harcourt Publishers Ltd Evidence-based Obstetrics and Gynecology (2001) 3, 65d66 doi:10.1054/ebog.2001.0228, available online at http://www.idealibrary.com on 65