Article Text

Review: oxytocin receptor antagonists for preterm labour do not improve infant outcomes more than placebo or other tocolytics
  1. Solange Wyatt, MD1,
  2. Debra A Guinn, MD2
  1. 1Oregon Health Sciences University, Portland, Oregon, USA
  2. 2Northwest Perinatal Center, Portland, Oregon, USA

    Statistics from

 Q In pregnant women with preterm labour, do oxytocin receptor antagonists delay delivery and improve infant outcomes?

    Clinical impact ratings GP/FP/Obstetrics ★★★★★★★ Obstetrics ★★★★★☆☆ Endocrine ★★★★★★☆


    Embedded ImageData sources:

    Cochrane Pregnancy and Childbirth Group’s Trials Register and CENTRAL (to September 2004), Medline and EMBASE/Excerpta Medica (to June 2004), and experts.

    Embedded ImageStudy selection and assessment:

    randomised controlled trials (RCTs) that compared oxytocin receptor antagonists used for tocolysis in the management of women with preterm labour (regular uterine contractions at 20–36 weeks gestation, regardless of membrane or cervical status) with placebo or no treatment, or other tocolytic agents. Quasi-randomised trials and outcomes with <80% follow up were excluded.

    Embedded ImageOutcomes:

    perinatal, neonatal, and infant death; birth within 48 hours; preterm birth; birth weight; adverse neonatal outcomes (including respiratory distress syndrome, intraventricular haemorrhage, and necrotising enterocolitis); admission to neonatal intensive care unit (NICU); and maternal drug reaction.


    2 RCTs (n = 651) compared atosiban with placebo, and 4 RCTs (n = 1044) compared atosiban with β mimetic agents. Women with ruptured membranes, pre-eclampsia, or gestational hypertension were excluded from all RCTs. Compared with placebo, atosiban did not reduce the risk of preterm birth, perinatal death, or infant death in the first year (see table at Maternal drug reactions were increased with atosiban (table). Mean birth weight was lower with atosiban (weighted mean difference −138 g, 95% CI −249 to −28). Atosiban showed no benefit over β mimetics for delaying birth >48 hours or preventing neonatal death (see table at Maternal drug reactions were reduced with atosiban (table). Mean birth weights were similar for the 2 drugs (weighted mean difference 26 g, 95% CI −82 to 135), but birth weight <1500 g was more frequent with atosiban (RRI 96%, CI 15 to 235). Atosiban did not differ from placebo or β mimetics for risk of any adverse infant outcome or NICU admission.


    Atosiban does not prevent preterm birth or improve infant outcomes compared with placebo or β mimetics.


    Tocolytic agents have been widely used to delay preterm delivery. Atosiban, the only new tocolytic in the past 2 decades, has been extensively tested. However, the manufacturer’s 1998 application to the US Food and Drug Administration (FDA) was denied. Several large RCTs had shown atosiban’s superiority to placebo in delaying delivery; its safety profile was encouraging, compared with β mimetics and calcium channel blockers. Why would the FDA not support such promising therapy? Firstly, outcome data were lacking. Secondly, although atosiban delayed delivery for up to 7 days, neonatal outcomes were not improved. Indeed, in pregnancies of <28 weeks’ gestation, atosiban might have increased neonatal mortality.

    The review by Papatsonis et al summarises the current evidence for atosiban and reiterates its lack of benefit for the neonate. Maternal drug reactions with atosiban were increased compared with placebo, but reduced compared with β mimetics. Should the FDA endorse a drug with questionable efficacy simply because it has fewer side effects than currently used drugs? Should tocolytic efficacy be defined as simply the prolongation of pregnancy or is neonatal benefit required? The guidelines of several international organisations1 recommend tocolysis to gain time for use of antenatal corticosteroids or transfer to a tertiary care centre, interventions known to improve neonatal outcomes. β mimetics have been shown to prolong pregnancy; however, the use of antenatal corticosteroids was low in those trials, and delay in delivery did not translate into improved neonatal outcomes. In the atosiban trials, despite high rates of antenatal corticosteroid use in both groups, atosiban did not improve neonatal outcomes. The efficacy of atosiban and all currently used off label tocolytics is questionable and the FDA’s stance was reasonable and justified.

    Clinicians must critically evaluate the widespread practice of tocolysis. We should enroll our patients in placebo controlled RCTs without rescue to evaluate the efficacy of tocolysis compared with no tocolysis. Although some providers suggest that such a trial is unethical and denies women the “standard of care,” it is long overdue and needs to be done.


    Atosiban v placebo or β mimetics for preterm labour*

 Q In pregnant women with preterm labour, do oxytocin receptor antagonists delay delivery and improve infant outcomes?

    Clinical impact ratings GP/FP/Obstetrics ★★★★★★★ Obstetrics ★★★★★☆☆ Endocrine ★★★★★★☆

    View Abstract


    • For correspondence: Dr D Papatsonis, Amphia Hospital Breda, Breda, Netherlands. hoog.pap{at}

    • Source of funding: Department of Health and Ageing, Australia.

    Request permissions

    If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.