Article Text

Randomised controlled trial
For pregnant women near full term with intrauterine growth restriction, induction of labour does not affect risk of adverse neonatal outcome or caesarean section compared with expectant management
  1. Janet Hirst
  1. School of Healthcare, University of Leeds, West Yorkshire, UK
  1. Correspondence to Janet Hirst
    School of Healthcare, Baines Wing, University of Leeds, Leeds, West Yorkshire LS2 9UT, UK; j.hirst{at}

Statistics from

Commentary on: .

Implications for practice and research

  • No differences were found in relation to adverse neonatal outcomes, the rates of instrumental vaginal delivery or caesarean section for women with suspected intrauterine growth restriction (IUGR) who were induced and those who were managed expectantly.

  • These results suggest that women with suspected IUGR should be given the choice of the two care options.

  • As this is the first randomised controlled trial in this area, further studies are required to confirm the findings.

  • Further research is also needed to determine the most effective way of identifying IUGR.


This study focused on pregnant women with a singleton fetus suspected of IUGR (failure to reach growth potential) between 36 and 41 weeks gestation. Comparisons between two forms of care, induction of labour versus expectant management with enhanced antenatal care, were needed to support obstetrician's decision making regarding mode of delivery1 and inform women's choice towards mode of birthing. IUGR is associated with serious intrapartum and neonatal complications.2 3 Growth restriction is attributed to 1–2% of neonatal deaths worldwide4; a rate of 3% of all pregnancies is reported,5 and based on their sampling procedure, it appears that Boers and colleagues have assumed an incidence about 4%. The cause is associated with poor placental function, as a consequence of other obstetric or medical conditions; suspected fetuses are often identified by midwives assessing symphysis-fundal height using a tape measure. Subsequent diagnosis is by serial ultrasounds.


This was a multicentred, randomised, equivalence trial conducted in The Netherlands over a 4-year period; 52 maternity units took part. Primary outcomes were a composite of neonatal outcomes (admission to neonatal intensive care, APGAR score <7 at 5 min, umbilical artery pH <7.05 and death before discharge). Secondary outcomes included mode of delivery and duration of hospital stay. Clear inclusion and exclusion criteria were evident, and a robust and secure, web-based random allocation procedure employed to minimise allocation bias. Detailed power calculations assure the reader of robust intentions, and the sample size was adequate. Blinding could not be achieved while outcomes were sufficiently objective. The impracticality of designing a trial sufficiently powered to detect differences between groups in rates of stillbirth5 or neonatal death and problematic diagnosis limited the study.

The induction arm of the trial included 321 pregnant women, and there were 329 in the expectant arm. More interestingly, 452 women declined randomisation. Why women refused or the extent to which this varied between recruiting hospitals is not clear. Therefore, we cannot rule out reluctance among obstetricians in offering entry to the trial to eligible women. All data were routinely available, and analyses were by intention to treat, reporting appropriate techniques for proportional differences, summaries, comparisons and checking for equivalence between margins.


The results have ruled out a difference in composite, adverse neonatal outcomes larger than 4.3% (not stillbirth or neonatal death), as a total of 17 (5.3%) neonates in the induction arm and 20 (6.1%) neonates in the expectant arm had the composite neonatal outcome (difference −0.8%, 95% CI −4.3% to 2.8%). Babies from the induction arm were, on average, born 10 days earlier and lighter. This has caused others to question the initial diagnosis, that is, that babies born in the expectant group were 130 g heavier as they were still growing.6 7 It is notable that 51% of women in the expectant arm had labour induced, whereas spontaneous onset of labour in the induction arm was only 4%. Induction rates among the general population of pregnant women in The Netherlands are not reported making it difficult to assess any threats to the planned analyses with this crossover.


This study is important as it leads this field. It has demonstrated women's willingness to engage in trials of a controversial nature. The diagnosis of IUGR in this study may have been flawed although the strategies used represent current practice, and the results are therefore likely to be generalisable. However, other commentators have expressed concern about accuracy of diagnosis.6 7 How women can utilise such information to inform an optimum birth plan is not clear even though it is an anticipated benefit of the study. How health services can justify resources to enhance antenatal care, as in the expectant group, to reassure women (and healthcare professionals) is also unclear; an economic evaluation is needed.

It is notable that women in all groups in the study had a normal body mass index (BMI) at trial entry. It is unclear when this occurred, but, given that the average BMI across all groups was 21–22, it suggests trial entry was at ‘booking’ during early pregnancy (ie, all pregnant women were potentially eligible). Thus, as pregnancy progressed, midwives would have had a reasonable chance of recognising fetuses at suspected risk of IUGR if weight gain was not excessive. Recognition of IUGR is paramount although it is likely to become more difficult as other studies have found one in five pregnant women classed as obese at booking.8 This will make screening for growth-restricted fetuses, using maternal symphysis-fundal height measurement using a tape measure from 24 weeks gestation, increasingly problematic.


View Abstract


  • Competing interests None.

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.